OpenCV version++

3.4.0
Merge pull request #10398 from alalek:ml_simplify_simulated_annealing
2026-07-24 12:53:04 +04:00 · 2017-12-22 19:46:21 +00:00 · 2017-12-22 22:43:59 +03:00 · 2017-12-22 18:50:04 +00:00 · 2017-12-22 16:18:50 +00:00 · 2017-12-22 15:58:56 +00:00
1501 changed files with 141322 additions and 71629 deletions
@@ -1,10 +1,11 @@
+# ignore dot files/directories
+.*
+!.gitignore
+
 *.autosave
 *.pyc
 *.user
 *~
-.*.swp
-.DS_Store
-.sw[a-z]
 Thumbs.db
 tags
 tegra/
@@ -21,4 +22,3 @@ bin/
 *.log
 *.tlog
 build
-.cache
@@ -37,6 +37,4 @@ if(WITH_NEON)
    target_compile_definitions(carotene_objs PRIVATE "-DWITH_NEON")
 endif()

-set_target_properties(carotene_objs PROPERTIES POSITION_INDEPENDENT_CODE TRUE)
-
 add_library(carotene STATIC EXCLUDE_FROM_ALL "$<TARGET_OBJECTS:carotene_objs>")
@@ -3,8 +3,6 @@ cmake_minimum_required(VERSION 2.8.8 FATAL_ERROR)
 include(CheckCCompilerFlag)
 include(CheckCXXCompilerFlag)

-set(CMAKE_POSITION_INDEPENDENT_CODE ON)
-
 set(TEGRA_HAL_DIR "${CMAKE_CURRENT_SOURCE_DIR}")
 set(CAROTENE_DIR "${TEGRA_HAL_DIR}/../")

@@ -95,7 +93,6 @@ set_property(DIRECTORY APPEND PROPERTY COMPILE_DEFINITIONS ${carotene_defs})
  endif()

 add_library(tegra_hal STATIC $<TARGET_OBJECTS:carotene_objs>)
-set_target_properties(tegra_hal PROPERTIES POSITION_INDEPENDENT_CODE TRUE)
 set_target_properties(tegra_hal PROPERTIES ARCHIVE_OUTPUT_DIRECTORY ${3P_LIBRARY_OUTPUT_PATH})
 set(OPENCV_SRC_DIR "${CMAKE_SOURCE_DIR}")
 if(NOT BUILD_SHARED_LIBS)
@@ -67,7 +67,7 @@
                 size_t src2_step;

 #define DST_ARG1 DT * dst1_data_, size_t dst1_step_,
-#define DST_STORE1 dst1_data(dst1_data_), dst1_step(dst1_step_), 
+#define DST_STORE1 dst1_data(dst1_data_), dst1_step(dst1_step_),
 #define DST_VAR1 DT * dst1_data; \
                 size_t dst1_step;

@@ -1073,7 +1073,7 @@ struct FilterCtx
 inline int TEGRA_FILTERINIT(cvhalFilter2D **context, uchar *kernel_data, size_t kernel_step, int kernel_type, int kernel_width, int kernel_height,
                            int max_width, int max_height, int src_type, int dst_type, int borderType, double delta, int anchor_x, int anchor_y, bool allowSubmatrix, bool allowInplace)
 {
-    if(!context || !kernel_data || allowSubmatrix || allowInplace || 
+    if(!context || !kernel_data || allowSubmatrix || allowInplace ||
       src_type != CV_8UC1 || dst_type != CV_8UC1 ||
       delta != 0 || anchor_x != kernel_width / 2 || anchor_y != kernel_height / 2 )
        return CV_HAL_ERROR_NOT_IMPLEMENTED;
@@ -1105,7 +1105,7 @@ inline int TEGRA_FILTERINIT(cvhalFilter2D **context, uchar *kernel_data, size_t
        return CV_HAL_ERROR_NOT_IMPLEMENTED;
    }

-    if(!CAROTENE_NS::isConvolutionSupported(CAROTENE_NS::Size2D(max_width, max_height), ctx->ksize, ctx->border))    
+    if(!CAROTENE_NS::isConvolutionSupported(CAROTENE_NS::Size2D(max_width, max_height), ctx->ksize, ctx->border))
    {
        delete ctx;
        return CV_HAL_ERROR_NOT_IMPLEMENTED;
@@ -1212,7 +1212,7 @@ inline int TEGRA_SEPFILTERINIT(cvhalFilter2D **context, int src_type, int dst_ty
        return CV_HAL_ERROR_NOT_IMPLEMENTED;
    }

-    if(!CAROTENE_NS::isSeparableFilter3x3Supported(CAROTENE_NS::Size2D(16, 16), ctx->border, 3, 3))    
+    if(!CAROTENE_NS::isSeparableFilter3x3Supported(CAROTENE_NS::Size2D(16, 16), ctx->border, 3, 3))
    {
        delete ctx;
        return CV_HAL_ERROR_NOT_IMPLEMENTED;
@@ -1341,8 +1341,8 @@ inline int TEGRA_MORPHINIT(cvhalFilter2D **context, int operation, int src_type,
    ctx->anchor_y = anchor_y;
    switch(operation)
    {
-    case MORPH_ERODE:
-    case MORPH_DILATE:
+    case CV_HAL_MORPH_ERODE:
+    case CV_HAL_MORPH_DILATE:
        ctx->operation = operation;
        break;
    default:
@@ -1355,7 +1355,7 @@ inline int TEGRA_MORPHINIT(cvhalFilter2D **context, int operation, int src_type,
        ctx->border = CAROTENE_NS::BORDER_MODE_CONSTANT;
        if( borderValue[0] == DBL_MAX && borderValue[1] == DBL_MAX && borderValue[2] == DBL_MAX && borderValue[3] == DBL_MAX )
        {
-            if( operation == MORPH_ERODE )
+            if( operation == CV_HAL_MORPH_ERODE )
                for(int i = 0; i < ctx->channels; ++i)
                    ctx->borderValues[i] = (CAROTENE_NS::u8)UCHAR_MAX;
            else
@@ -1404,14 +1404,14 @@ inline int TEGRA_MORPHFREE(cvhalFilter2D *context)
 ( \
    (void)dst_full_width, (void)dst_full_height, (void)dst_roi_x, (void)dst_roi_y, \
    context && CAROTENE_NS::isSupportedConfiguration() ? \
-        ((MorphCtx*)context)->operation == MORPH_ERODE ? \
+        ((MorphCtx*)context)->operation == CV_HAL_MORPH_ERODE ? \
        CAROTENE_NS::erode(CAROTENE_NS::Size2D(width, height), ((MorphCtx*)context)->channels, \
                           src_data, src_step, dst_data, dst_step, \
                           ((MorphCtx*)context)->ksize, ((MorphCtx*)context)->anchor_x, ((MorphCtx*)context)->anchor_y, \
                           ((MorphCtx*)context)->border, ((MorphCtx*)context)->border, ((MorphCtx*)context)->borderValues, \
                           CAROTENE_NS::Margin(src_roi_x, src_full_width - width - src_roi_x, src_roi_y, src_full_height - height - src_roi_y)), \
        CV_HAL_ERROR_OK : \
-        ((MorphCtx*)context)->operation == MORPH_DILATE ? \
+        ((MorphCtx*)context)->operation == CV_HAL_MORPH_DILATE ? \
        CAROTENE_NS::dilate(CAROTENE_NS::Size2D(width, height), ((MorphCtx*)context)->channels, \
                            src_data, src_step, dst_data, dst_step, \
                            ((MorphCtx*)context)->ksize, ((MorphCtx*)context)->anchor_x, ((MorphCtx*)context)->anchor_y, \
@@ -231,7 +231,7 @@ void extract4(const Size2D &size,
                 srcStride == dst2Stride && \
                 srcStride == dst3Stride &&

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)

 #define SPLIT_ASM2(sgn, bits) __asm__ ( \
                                          "vld2." #bits " {d0, d2}, [%[in0]]            \n\t" \
@@ -351,7 +351,7 @@ void extract4(const Size2D &size,
    }                                                                                                   \
 }

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)

 #define ALPHA_QUAD(sgn, bits) { \
                                  internal::prefetch(src + sj); \
@@ -77,7 +77,7 @@ namespace CAROTENE_NS {
                 dstStride == src2Stride && \
                 dstStride == src3Stride &&

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)

 #define MERGE_ASM2(sgn, bits) __asm__ ( \
                                          "vld1." #bits " {d0-d1}, [%[in0]]             \n\t" \
@@ -97,7 +97,7 @@ void rgb2gray(const Size2D &size, COLOR_SPACE color_space,
    const u32 G2Y = color_space == COLOR_SPACE_BT601 ? G2Y_BT601 : G2Y_BT709;
    const u32 B2Y = color_space == COLOR_SPACE_BT601 ? B2Y_BT601 : B2Y_BT709;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register int16x4_t v_r2y asm ("d31") = vmov_n_s16(R2Y);
    register int16x4_t v_g2y asm ("d30") = vmov_n_s16(G2Y);
    register int16x4_t v_b2y asm ("d29") = vmov_n_s16(B2Y);
@@ -116,7 +116,7 @@ void rgb2gray(const Size2D &size, COLOR_SPACE color_space,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
        for (; dj < roiw8; sj += 24, dj += 8)
        {
            internal::prefetch(src + sj);
@@ -198,7 +198,7 @@ void rgbx2gray(const Size2D &size, COLOR_SPACE color_space,
    const u32 G2Y = color_space == COLOR_SPACE_BT601 ? G2Y_BT601 : G2Y_BT709;
    const u32 B2Y = color_space == COLOR_SPACE_BT601 ? B2Y_BT601 : B2Y_BT709;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register int16x4_t v_r2y asm ("d31") = vmov_n_s16(R2Y);
    register int16x4_t v_g2y asm ("d30") = vmov_n_s16(G2Y);
    register int16x4_t v_b2y asm ("d29") = vmov_n_s16(B2Y);
@@ -217,7 +217,7 @@ void rgbx2gray(const Size2D &size, COLOR_SPACE color_space,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
        for (; dj < roiw8; sj += 32, dj += 8)
        {
            internal::prefetch(src + sj);
@@ -300,7 +300,7 @@ void bgr2gray(const Size2D &size, COLOR_SPACE color_space,
    const u32 G2Y = color_space == COLOR_SPACE_BT601 ? G2Y_BT601 : G2Y_BT709;
    const u32 B2Y = color_space == COLOR_SPACE_BT601 ? B2Y_BT601 : B2Y_BT709;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register int16x4_t v_r2y asm ("d31") = vmov_n_s16(R2Y);
    register int16x4_t v_g2y asm ("d30") = vmov_n_s16(G2Y);
    register int16x4_t v_b2y asm ("d29") = vmov_n_s16(B2Y);
@@ -319,7 +319,7 @@ void bgr2gray(const Size2D &size, COLOR_SPACE color_space,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
        for (; dj < roiw8; sj += 24, dj += 8)
        {
            internal::prefetch(src + sj);
@@ -402,7 +402,7 @@ void bgrx2gray(const Size2D &size, COLOR_SPACE color_space,
    const u32 G2Y = color_space == COLOR_SPACE_BT601 ? G2Y_BT601 : G2Y_BT709;
    const u32 B2Y = color_space == COLOR_SPACE_BT601 ? B2Y_BT601 : B2Y_BT709;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register int16x4_t v_r2y asm ("d31") = vmov_n_s16(R2Y);
    register int16x4_t v_g2y asm ("d30") = vmov_n_s16(G2Y);
    register int16x4_t v_b2y asm ("d29") = vmov_n_s16(B2Y);
@@ -421,7 +421,7 @@ void bgrx2gray(const Size2D &size, COLOR_SPACE color_space,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
        for (; dj < roiw8; sj += 32, dj += 8)
        {
            internal::prefetch(src + sj);
@@ -512,7 +512,7 @@ void gray2rgb(const Size2D &size,
        for (; sj < roiw16; sj += 16, dj += 48)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
            "vld1.8 {d0-d1}, [%[in0]]            \n\t"
            "vmov.8 q1, q0                       \n\t"
@@ -538,7 +538,7 @@ void gray2rgb(const Size2D &size,

        if (sj < roiw8)
        {
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
            "vld1.8 {d0}, [%[in]]                \n\t"
            "vmov.8 d1, d0                       \n\t"
@@ -584,7 +584,7 @@ void gray2rgbx(const Size2D &size,
    size_t roiw16 = size.width >= 15 ? size.width - 15 : 0;
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register uint8x16_t vc255   asm ("q4") = vmovq_n_u8(255);
 #else
    uint8x16x4_t vRgba;
@@ -602,7 +602,7 @@ void gray2rgbx(const Size2D &size,
        for (; sj < roiw16; sj += 16, dj += 64)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
            "vld1.8 {d0-d1}, [%[in0]]            \n\t"
            "vmov.8 q1, q0                       \n\t"
@@ -628,7 +628,7 @@ void gray2rgbx(const Size2D &size,

        if (sj < roiw8)
        {
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
            "vld1.8 {d5}, [%[in]]                \n\t"
            "vmov.8 d6, d5                       \n\t"
@@ -672,7 +672,7 @@ void rgb2rgbx(const Size2D &size,
    internal::assertSupportedConfiguration();
 #ifdef CAROTENE_NEON
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
    register uint8x8_t vc255_0  asm ("d3") = vmov_n_u8(255);
 #else
    size_t roiw16 = size.width >= 15 ? size.width - 15 : 0;
@@ -688,7 +688,7 @@ void rgb2rgbx(const Size2D &size,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u, j = 0u;

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
        for (; j < roiw8; sj += 24, dj += 32, j += 8)
        {
            internal::prefetch(src + sj);
@@ -742,7 +742,7 @@ void rgbx2rgb(const Size2D &size,
    internal::assertSupportedConfiguration();
 #ifdef CAROTENE_NEON
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
-#if !defined(__GNUC__) || !defined(__arm__)
+#if !(!defined(__aarch64__) && defined(__GNUC__) && defined(__arm__))
    size_t roiw16 = size.width >= 15 ? size.width - 15 : 0;
    union { uint8x16x4_t v4; uint8x16x3_t v3; } v_dst0;
    union { uint8x8x4_t v4; uint8x8x3_t v3; } v_dst;
@@ -754,7 +754,7 @@ void rgbx2rgb(const Size2D &size,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u, j = 0u;

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
        for (; j < roiw8; sj += 32, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
@@ -805,7 +805,7 @@ void rgb2bgr(const Size2D &size,
 {
    internal::assertSupportedConfiguration();
 #ifdef CAROTENE_NEON
-#if !defined(__GNUC__) || !defined(__arm__)
+#if !(!defined(__aarch64__) && defined(__GNUC__) && defined(__arm__))
    size_t roiw16 = size.width >= 15 ? size.width - 15 : 0;
 #endif
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
@@ -817,7 +817,7 @@ void rgb2bgr(const Size2D &size,
        size_t sj = 0u, dj = 0u, j = 0u;


-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
        for (; j < roiw8; sj += 24, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
@@ -874,7 +874,7 @@ void rgbx2bgrx(const Size2D &size,
 {
    internal::assertSupportedConfiguration();
 #ifdef CAROTENE_NEON
-#if !defined(__GNUC__) || !defined(__arm__)
+#if !(!defined(__aarch64__) && defined(__GNUC__) && defined(__arm__))
    size_t roiw16 = size.width >= 15 ? size.width - 15 : 0;
 #endif
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
@@ -885,7 +885,7 @@ void rgbx2bgrx(const Size2D &size,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u, j = 0u;

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
        for (; j < roiw8; sj += 32, dj += 32, j += 8)
        {
            internal::prefetch(src + sj);
@@ -943,7 +943,7 @@ void rgbx2bgr(const Size2D &size,
 {
    internal::assertSupportedConfiguration();
 #ifdef CAROTENE_NEON
-#if !defined(__GNUC__) || !defined(__arm__)
+#if !(!defined(__aarch64__) && defined(__GNUC__) && defined(__arm__))
    size_t roiw16 = size.width >= 15 ? size.width - 15 : 0;
 #endif
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
@@ -954,7 +954,7 @@ void rgbx2bgr(const Size2D &size,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u, j = 0u;

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
        for (; j < roiw8; sj += 32, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
@@ -1010,7 +1010,7 @@ void rgb2bgrx(const Size2D &size,
 {
    internal::assertSupportedConfiguration();
 #ifdef CAROTENE_NEON
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
    register uint8x8_t vc255  asm ("d3") = vmov_n_u8(255);
 #else
    union { uint8x16x4_t v4; uint8x16x3_t v3; } vals0;
@@ -1019,7 +1019,7 @@ void rgb2bgrx(const Size2D &size,
    vals8.v4.val[3] = vmov_n_u8(255);
 #endif

-#if !defined(__GNUC__) || !defined(__arm__)
+#if !(!defined(__aarch64__) && defined(__GNUC__) && defined(__arm__))
    size_t roiw16 = size.width >= 15 ? size.width - 15 : 0;
 #endif
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
@@ -1030,7 +1030,7 @@ void rgb2bgrx(const Size2D &size,
        u8 * dst = internal::getRowPtr(dstBase, dstStride, i);
        size_t sj = 0u, dj = 0u, j = 0u;

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
        for (; j < roiw8; sj += 24, dj += 32, j += 8)
        {
            internal::prefetch(src + sj);
@@ -1087,12 +1087,12 @@ inline uint8x8x3_t convertToHSV(const uint8x8_t vR, const uint8x8_t vG, const ui
                                const s32 hrange )
 {
    const s32 hsv_shift = 12;
-    register const f32 vsdiv_table = f32(255 << hsv_shift);
-    register f32 vhdiv_table = f32(hrange << hsv_shift);
-    register const s32 vhrange = hrange;
-    register const s32 v0 = s32(0);
-    register const s32 vshift = s32(1 << (hsv_shift-1));
-    register const s32 v6 = s32(6);
+    const f32 vsdiv_table = f32(255 << hsv_shift);
+    f32 vhdiv_table = f32(hrange << hsv_shift);
+    const s32 vhrange = hrange;
+    const s32 v0 = s32(0);
+    const s32 vshift = s32(1 << (hsv_shift-1));
+    const s32 v6 = s32(6);

    uint8x8_t vMin = vmin_u8(vR, vG);
    uint8x8_t vMax = vmax_u8(vR, vG);
@@ -1409,7 +1409,7 @@ inline void convertToHSV(const s32 r, const s32 g, const s32 b,
              "d24","d25","d26","d27","d28","d29","d30","d31"         \
            );

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)

 #define YCRCB_CONSTS                                                        \
    register int16x4_t vcYR  asm ("d31") = vmov_n_s16(4899);                \
@@ -1555,7 +1555,7 @@ inline uint8x8x3_t convertToYCrCb( const int16x8_t& vR, const int16x8_t& vG, con
 #define COEFF_G   (  8663)
 #define COEFF_B   (-17705)

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 #define YUV420ALPHA3_CONST
 #define YUV420ALPHA4_CONST register uint8x16_t c255  asm ("q13") = vmovq_n_u8(255);
 #define YUV420ALPHA3_CONVERT
@@ -1852,7 +1852,7 @@ void rgb2hsv(const Size2D &size,
 #ifdef CAROTENE_NEON
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
    const s32 hsv_shift = 12;
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register const f32 vsdiv_table = f32(255 << hsv_shift);
    register f32 vhdiv_table = f32(hrange << hsv_shift);
    register const s32 vhrange = hrange;
@@ -1871,7 +1871,7 @@ void rgb2hsv(const Size2D &size,
        for (; j < roiw8; sj += 24, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERT_TO_HSV_ASM(vld3.8 {d0-d2}, d0, d2)
 #else
            uint8x8x3_t vRgb = vld3_u8(src + sj);
@@ -1904,7 +1904,7 @@ void rgbx2hsv(const Size2D &size,
 #ifdef CAROTENE_NEON
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
    const s32 hsv_shift = 12;
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register const f32 vsdiv_table = f32(255 << hsv_shift);
    register f32 vhdiv_table = f32(hrange << hsv_shift);
    register const s32 vhrange = hrange;
@@ -1923,7 +1923,7 @@ void rgbx2hsv(const Size2D &size,
        for (; j < roiw8; sj += 32, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERT_TO_HSV_ASM(vld4.8 {d0-d3}, d0, d2)
 #else
            uint8x8x4_t vRgb = vld4_u8(src + sj);
@@ -1956,7 +1956,7 @@ void bgr2hsv(const Size2D &size,
 #ifdef CAROTENE_NEON
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
    const s32 hsv_shift = 12;
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register const f32 vsdiv_table = f32(255 << hsv_shift);
    register f32 vhdiv_table = f32(hrange << hsv_shift);
    register const s32 vhrange = hrange;
@@ -1975,7 +1975,7 @@ void bgr2hsv(const Size2D &size,
        for (; j < roiw8; sj += 24, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERT_TO_HSV_ASM(vld3.8 {d0-d2}, d2, d0)
 #else
            uint8x8x3_t vRgb = vld3_u8(src + sj);
@@ -2008,7 +2008,7 @@ void bgrx2hsv(const Size2D &size,
 #ifdef CAROTENE_NEON
    size_t roiw8 = size.width >= 7 ? size.width - 7 : 0;
    const s32 hsv_shift = 12;
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register const f32 vsdiv_table = f32(255 << hsv_shift);
    register f32 vhdiv_table = f32(hrange << hsv_shift);
    register const s32 vhrange = hrange;
@@ -2027,7 +2027,7 @@ void bgrx2hsv(const Size2D &size,
        for (; j < roiw8; sj += 32, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERT_TO_HSV_ASM(vld4.8 {d0-d3}, d2, d0)
 #else
            uint8x8x4_t vRgb = vld4_u8(src + sj);
@@ -2068,7 +2068,7 @@ void rgbx2bgr565(const Size2D &size,
        for (; j < roiw16; sj += 64, dj += 32, j += 16)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
                "vld4.8 {d2, d4, d6, d8}, [%[in0]]        @  q0       q1       q2       q3       q4       \n\t"
                "vld4.8 {d3, d5, d7, d9}, [%[in1]]        @  xxxxxxxx rrrrRRRR ggggGGGG bbbbBBBB xxxxxxxx \n\t"
@@ -2122,7 +2122,7 @@ void rgb2bgr565(const Size2D &size,
        for (; j < roiw16; sj += 48, dj += 32, j += 16)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
                "vld3.8 {d2, d4, d6}, [%[in0]]       @  q0       q1       q2       q3       q4       \n\t"
                "vld3.8 {d3, d5, d7}, [%[in1]]       @  xxxxxxxx rrrrRRRR ggggGGGG bbbbBBBB xxxxxxxx \n\t"
@@ -2176,7 +2176,7 @@ void rgbx2rgb565(const Size2D &size,
        for (; j < roiw16; sj += 64, dj += 32, j += 16)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
                "vld4.8 {d0, d2, d4, d6}, [%[in0]]    @  q0       q1       q2       q3         \n\t"
                "vld4.8 {d1, d3, d5, d7}, [%[in1]]    @  rrrrRRRR ggggGGGG bbbbBBBB aaaaAAAA   \n\t"
@@ -2230,7 +2230,7 @@ void rgb2rgb565(const Size2D &size,
        for (; j < roiw16; sj += 48, dj += 32, j += 16)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
                "vld3.8 {d0, d2, d4}, [%[in0]]        @  q0       q1       q2       q3         \n\t"
                "vld3.8 {d1, d3, d5}, [%[in1]]        @  rrrrRRRR ggggGGGG bbbbBBBB xxxxxxxx   \n\t"
@@ -2285,7 +2285,7 @@ void rgb2ycrcb(const Size2D &size,
        for (; j < roiw8; sj += 24, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTTOYCRCB(vld3.8 {d0-d2}, d0, d1, d2)
 #else
            uint8x8x3_t vRgb = vld3_u8(src + sj);
@@ -2329,7 +2329,7 @@ void rgbx2ycrcb(const Size2D &size,
        for (; j < roiw8; sj += 32, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTTOYCRCB(vld4.8 {d0-d3}, d0, d1, d2)
 #else
            uint8x8x4_t vRgba = vld4_u8(src + sj);
@@ -2373,7 +2373,7 @@ void bgr2ycrcb(const Size2D &size,
        for (; j < roiw8; sj += 24, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTTOYCRCB(vld3.8 {d0-d2}, d2, d1, d0)
 #else
            uint8x8x3_t vBgr = vld3_u8(src + sj);
@@ -2417,7 +2417,7 @@ void bgrx2ycrcb(const Size2D &size,
        for (; j < roiw8; sj += 32, dj += 24, j += 8)
        {
            internal::prefetch(src + sj);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTTOYCRCB(vld4.8 {d0-d3}, d2, d1, d0)
 #else
            uint8x8x4_t vBgra = vld4_u8(src + sj);
@@ -2499,7 +2499,7 @@ void yuv420sp2rgb(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(3, d1, d0, q5, q6)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -2545,7 +2545,7 @@ void yuv420sp2rgbx(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(4, d1, d0, q5, q6)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -2591,7 +2591,7 @@ void yuv420i2rgb(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(3, d0, d1, q5, q6)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -2637,7 +2637,7 @@ void yuv420i2rgbx(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(4, d0, d1, q5, q6)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -2683,7 +2683,7 @@ void yuv420sp2bgr(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(3, d1, d0, q6, q5)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -2729,7 +2729,7 @@ void yuv420sp2bgrx(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(4, d1, d0, q6, q5)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -2775,7 +2775,7 @@ void yuv420i2bgr(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(3, d0, d1, q6, q5)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -2821,7 +2821,7 @@ void yuv420i2bgrx(const Size2D &size,
            internal::prefetch(uv + j);
            internal::prefetch(y1 + j);
            internal::prefetch(y2 + j);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            CONVERTYUV420TORGB(4, d0, d1, q6, q5)
 #else
            convertYUV420.ToRGB(y1 + j, y2 + j, uv + j, dst1 + dj, dst2 + dj);
@@ -101,7 +101,7 @@ CVT_FUNC(u8, s8, 16,
     }
 })

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVT_FUNC(u8, u16, 16,
     register uint8x16_t zero0 asm ("q1") = vmovq_n_u8(0);,
 {
@@ -135,7 +135,7 @@ CVT_FUNC(u8, u16, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVT_FUNC(u8, s32, 16,
     register uint8x16_t zero0 asm ("q1") = vmovq_n_u8(0);
     register uint8x16_t zero1 asm ("q2") = vmovq_n_u8(0);
@@ -173,7 +173,7 @@ CVT_FUNC(u8, s32, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(u8, f32, 16,
 ,
 {
@@ -248,7 +248,7 @@ CVT_FUNC(s8, u8, 16,
     }
 })

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVT_FUNC(s8, u16, 16,
     register uint8x16_t zero0 asm ("q1") = vmovq_n_u8(0);,
 {
@@ -284,7 +284,7 @@ CVT_FUNC(s8, u16, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s8, s16, 16,
 ,
 {
@@ -323,7 +323,7 @@ CVT_FUNC(s8, s16, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVT_FUNC(s8, s32, 16,
 ,
 {
@@ -377,7 +377,7 @@ CVT_FUNC(s8, s32, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s8, f32, 16,
 ,
 {
@@ -440,7 +440,7 @@ CVT_FUNC(s8, f32, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(u16, u8, 16,
 ,
 {
@@ -479,7 +479,7 @@ CVT_FUNC(u16, u8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(u16, s8, 16,
    register uint8x16_t v127 asm ("q4") = vmovq_n_u8(127);,
 {
@@ -522,7 +522,7 @@ CVT_FUNC(u16, s8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVT_FUNC(u16, s16, 8,
     register uint16x8_t v32767 asm ("q4") = vmovq_n_u16(0x7FFF);,
 {
@@ -555,7 +555,7 @@ CVT_FUNC(u16, s16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVT_FUNC(u16, s32, 8,
     register uint16x8_t zero0 asm ("q1") = vmovq_n_u16(0);,
 {
@@ -589,7 +589,7 @@ CVT_FUNC(u16, s32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(u16, f32, 8,
 ,
 {
@@ -633,7 +633,7 @@ CVT_FUNC(u16, f32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s16, u8, 16,
 ,
 {
@@ -672,7 +672,7 @@ CVT_FUNC(s16, u8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s16, s8, 16,
 ,
 {
@@ -711,7 +711,7 @@ CVT_FUNC(s16, s8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVT_FUNC(s16, u16, 8,
     register int16x8_t vZero asm ("q4") = vmovq_n_s16(0);,
 {
@@ -747,7 +747,7 @@ CVT_FUNC(s16, u16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s16, s32, 8,
 ,
 {
@@ -786,7 +786,7 @@ CVT_FUNC(s16, s32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s16, f32, 8,
 ,
 {
@@ -829,7 +829,7 @@ CVT_FUNC(s16, f32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s32, u8, 8,
 ,
 {
@@ -870,7 +870,7 @@ CVT_FUNC(s32, u8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s32, s8, 8,
 ,
 {
@@ -911,7 +911,7 @@ CVT_FUNC(s32, s8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s32, u16, 8,
 ,
 {
@@ -950,7 +950,7 @@ CVT_FUNC(s32, u16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s32, s16, 8,
 ,
 {
@@ -989,7 +989,7 @@ CVT_FUNC(s32, s16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(s32, f32, 8,
 ,
 {
@@ -1034,7 +1034,7 @@ CVT_FUNC(s32, f32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(f32, u8, 8,
    register float32x4_t vmult asm ("q0") = vdupq_n_f32((float)(1 << 16));
    register uint32x4_t  vmask asm ("q1") = vdupq_n_u32(1<<16);,
@@ -1101,7 +1101,7 @@ CVT_FUNC(f32, u8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(f32, s8, 8,
     register float32x4_t vhalf asm ("q0") = vdupq_n_f32(0.5f);,
 {
@@ -1153,7 +1153,7 @@ CVT_FUNC(f32, s8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(f32, u16, 8,
     register float32x4_t vhalf asm ("q0") = vdupq_n_f32(0.5f);,
 {
@@ -1212,7 +1212,7 @@ CVT_FUNC(f32, u16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(f32, s16, 8,
     register float32x4_t vhalf asm ("q0") = vdupq_n_f32(0.5f);,
 {
@@ -1271,7 +1271,7 @@ CVT_FUNC(f32, s16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
 CVT_FUNC(f32, s32, 8,
     register float32x4_t vhalf asm ("q0") = vdupq_n_f32(0.5f);,
 {
@@ -135,7 +135,7 @@ namespace CAROTENE_NS {

 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC1(u8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -220,7 +220,7 @@ CVTS_FUNC1(u8, 16,
 })
 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC(u8, s8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -305,7 +305,7 @@ CVTS_FUNC(u8, s8, 16,
 })
 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC(u8, u16, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -389,7 +389,7 @@ CVTS_FUNC(u8, u16, 16,
 })
 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC(u8, s16, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -473,7 +473,7 @@ CVTS_FUNC(u8, s16, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(u8, s32, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -562,7 +562,7 @@ CVTS_FUNC(u8, s32, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(u8, f32, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta);,
@@ -643,7 +643,7 @@ CVTS_FUNC(u8, f32, 16,
 })
 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC(s8, u8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -728,7 +728,7 @@ CVTS_FUNC(s8, u8, 16,
 })
 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC1(s8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -813,7 +813,7 @@ CVTS_FUNC1(s8, 16,
 })
 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC(s8, u16, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -899,7 +899,7 @@ CVTS_FUNC(s8, u16, 16,
 })
 #endif

-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
 CVTS_FUNC(s8, s16, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -985,7 +985,7 @@ CVTS_FUNC(s8, s16, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s8, s32, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1074,7 +1074,7 @@ CVTS_FUNC(s8, s32, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s8, f32, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta);,
@@ -1155,7 +1155,7 @@ CVTS_FUNC(s8, f32, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(u16, u8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1214,7 +1214,7 @@ CVTS_FUNC(u16, u8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(u16, s8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1273,7 +1273,7 @@ CVTS_FUNC(u16, s8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC1(u16, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1330,7 +1330,7 @@ CVTS_FUNC1(u16, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(u16, s16, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1387,7 +1387,7 @@ CVTS_FUNC(u16, s16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(u16, s32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1443,7 +1443,7 @@ CVTS_FUNC(u16, s32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(u16, f32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta);,
@@ -1495,7 +1495,7 @@ CVTS_FUNC(u16, f32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s16, u8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1554,7 +1554,7 @@ CVTS_FUNC(s16, u8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s16, s8, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1613,7 +1613,7 @@ CVTS_FUNC(s16, s8, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s16, u16, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1670,7 +1670,7 @@ CVTS_FUNC(s16, u16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC1(s16, 16,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1727,7 +1727,7 @@ CVTS_FUNC1(s16, 16,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s16, s32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1783,7 +1783,7 @@ CVTS_FUNC(s16, s32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s16, f32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta);,
@@ -1835,7 +1835,7 @@ CVTS_FUNC(s16, f32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s32, u8, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1893,7 +1893,7 @@ CVTS_FUNC(s32, u8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s32, s8, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -1951,7 +1951,7 @@ CVTS_FUNC(s32, s8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s32, u16, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -2007,7 +2007,7 @@ CVTS_FUNC(s32, u16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s32, s16, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -2063,7 +2063,7 @@ CVTS_FUNC(s32, s16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC1(s32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -2118,7 +2118,7 @@ CVTS_FUNC1(s32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(s32, f32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta);,
@@ -2169,7 +2169,7 @@ CVTS_FUNC(s32, f32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(f32, u8, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)((1 << 16)*alpha));
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)((1 << 16)*beta));
@@ -2239,7 +2239,7 @@ CVTS_FUNC(f32, u8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(f32, s8, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -2293,7 +2293,7 @@ CVTS_FUNC(f32, s8, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(f32, u16, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -2345,7 +2345,7 @@ CVTS_FUNC(f32, u16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(f32, s16, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -2397,7 +2397,7 @@ CVTS_FUNC(f32, s16, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC(f32, s32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta + 0.5f);,
@@ -2448,7 +2448,7 @@ CVTS_FUNC(f32, s32, 8,
 })
 #endif

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
 CVTS_FUNC1(f32, 8,
    register float32x4_t vscale asm ("q0") = vdupq_n_f32((f32)alpha);
    register float32x4_t vshift asm ("q1") = vdupq_n_f32((f32)beta);,
@@ -327,7 +327,7 @@ void gaussianBlur5x5(const Size2D &size, s32 cn,
                u16* lidx1 = lane + x - 1*2;
                u16* lidx3 = lane + x + 1*2;
                u16* lidx4 = lane + x + 2*2;
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
                __asm__ __volatile__ (
                    "vld2.16 {d0, d2}, [%[in0]]!                              \n\t"
                    "vld2.16 {d1, d3}, [%[in0]]                               \n\t"
@@ -398,7 +398,7 @@ void gaussianBlur5x5(const Size2D &size, s32 cn,
                u16* lidx1 = lane + x - 1*3;
                u16* lidx3 = lane + x + 1*3;
                u16* lidx4 = lane + x + 2*3;
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ __volatile__ (
                    "vld3.16 {d0, d2, d4}, [%[in0]]!                          \n\t"
                    "vld3.16 {d1, d3, d5}, [%[in0]]                           \n\t"
@@ -482,7 +482,7 @@ void gaussianBlur5x5(const Size2D &size, s32 cn,
                u16* lidx1 = lane + x - 1*4;
                u16* lidx3 = lane + x + 1*4;
                u16* lidx4 = lane + x + 2*4;
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ __volatile__ (
                    "vld4.16 {d0, d2, d4, d6}, [%[in0]]!                      \n\t"
                    "vld4.16 {d1, d3, d5, d7}, [%[in0]]                       \n\t"
@@ -331,7 +331,7 @@ void gaussianPyramidDown(const Size2D &srcSize,
            for (; x < roiw8; x += 8)
            {
                internal::prefetch(lane + 2 * x);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
                __asm__ (
                    "vld2.16 {d0-d3}, [%[in0]]                               \n\t"
                    "vld2.16 {d4-d7}, [%[in4]]                               \n\t"
@@ -538,7 +538,7 @@ void gaussianPyramidDown(const Size2D &srcSize,
            for (; x < roiw4; x += 4)
            {
                internal::prefetch(lane + 2 * x);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
                __asm__ (
                    "vld2.32 {d0-d3}, [%[in0]]                              \n\t"
                    "vld2.32 {d4-d7}, [%[in4]]                              \n\t"
@@ -672,7 +672,7 @@ void gaussianPyramidDown(const Size2D &srcSize,
    std::vector<f32> _buf(cn*(srcSize.width + 4) + 32/sizeof(f32));
    f32* lane = internal::alignPtr(&_buf[2*cn], 32);

-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
    register float32x4_t vc6d4f32  asm ("q11") = vmovq_n_f32(1.5f);  // 6/4
    register float32x4_t vc1d4f32  asm ("q12") = vmovq_n_f32(0.25f); // 1/4

@@ -680,12 +680,12 @@ void gaussianPyramidDown(const Size2D &srcSize,
    register float32x4_t vc4d64f32 asm ("q14") = vmovq_n_f32(0.0625f);   //4/4/16
    register float32x4_t vc6d64f32 asm ("q15") = vmovq_n_f32(0.09375f);  //6/4/16
 #else
-    register float32x4_t vc6d4f32  = vmovq_n_f32(1.5f);  // 6/4
-    register float32x4_t vc1d4f32  = vmovq_n_f32(0.25f); // 1/4
+    float32x4_t vc6d4f32  = vmovq_n_f32(1.5f);  // 6/4
+    float32x4_t vc1d4f32  = vmovq_n_f32(0.25f); // 1/4

-    register float32x4_t vc1d64f32 = vmovq_n_f32(0.015625f); //1/4/16
-    register float32x4_t vc4d64f32 = vmovq_n_f32(0.0625f);   //4/4/16
-    register float32x4_t vc6d64f32 = vmovq_n_f32(0.09375f);  //6/4/16
+    float32x4_t vc1d64f32 = vmovq_n_f32(0.015625f); //1/4/16
+    float32x4_t vc4d64f32 = vmovq_n_f32(0.0625f);   //4/4/16
+    float32x4_t vc6d64f32 = vmovq_n_f32(0.09375f);  //6/4/16
 #endif

    for (size_t i = 0; i < dstSize.height; ++i)
@@ -739,7 +739,7 @@ void gaussianPyramidDown(const Size2D &srcSize,
            for (; x < roiw4; x += 4)
            {
                internal::prefetch(lane + 2 * x);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
                __asm__ __volatile__ (
                    "vld2.32 {d0-d3}, [%[in0]]                              \n\t"
                    "vld2.32 {d8-d11}, [%[in4]]                             \n\t"
@@ -932,7 +932,7 @@ pyrUp8uHorizontalConvolution:
            for (; x < lim; x += 8)
            {
                internal::prefetch(lane + x);
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ (
                    "vld1.16 {d0-d1}, [%[in0]]       /*q0 = v0*/            \n\t"
                    "vld1.16 {d2-d3}, [%[in2]]       /*q1 = v2*/            \n\t"
@@ -973,7 +973,7 @@ pyrUp8uHorizontalConvolution:
            for (; x < lim; x += 24)
            {
                internal::prefetch(lane + x);
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ (
                    "vmov.u16 q9, #6                                           \n\t"
                    "vld3.16 {d0, d2, d4}, [%[in0]]        /*v0*/              \n\t"
@@ -1064,7 +1064,7 @@ pyrUp8uHorizontalConvolution:
            for (; x < lim; x += 8)
            {
                internal::prefetch(lane + x);
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ (
                    "vld1.16 {d0-d1}, [%[in0]]       /*q0 = v0*/            \n\t"
                    "vld1.16 {d2-d3}, [%[in2]]       /*q1 = v2*/            \n\t"
@@ -1210,7 +1210,7 @@ pyrUp16sHorizontalConvolution:
            for (; x < lim; x += 4)
            {
                internal::prefetch(lane + x);
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ (
                    "vld1.32 {d0-d1}, [%[in0]]       /*q0 = v0*/            \n\t"
                    "vld1.32 {d2-d3}, [%[in2]]       /*q1 = v2*/            \n\t"
@@ -1251,7 +1251,7 @@ pyrUp16sHorizontalConvolution:
            for (; x < lim; x += 12)
            {
                internal::prefetch(lane + x + 3);
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ (
                    "vmov.s32 q9, #6                                           \n\t"
                    "vld3.32 {d0, d2, d4}, [%[in0]]        /*v0*/              \n\t"
@@ -1343,7 +1343,7 @@ pyrUp16sHorizontalConvolution:
            for (; x < lim; x += 4)
            {
                internal::prefetch(lane + x);
-#if defined(__GNUC__) && defined(__arm__)
+#if !defined(__aarch64__) && defined(__GNUC__) && defined(__arm__)
                __asm__ (
                    "vld1.32 {d0-d1}, [%[in0]]       /*q0 = v0*/            \n\t"
                    "vld1.32 {d2-d3}, [%[in2]]       /*q1 = v2*/            \n\t"
@@ -106,7 +106,8 @@ bool isResizeLinearOpenCVSupported(const Size2D &ssize, const Size2D &dsize, u32
            && !(ssize.width > 0xffffFFFF || ssize.height > 0xffffFFFF)// Restrict image size since internal index evaluation
                                                                       // is performed with u32
 #endif
-            && dsize.width >= 2 && dsize.height >= 8)
+            && dsize.width >= 2 && dsize.height >= 8
+            && (2*dsize.width != ssize.width || 2*dsize.height != ssize.height)) // 2x downscaling is performed as area in OpenCV which differs from this implementation
            return isSupportedConfiguration();
    default:
        return false;
@@ -72,11 +72,11 @@ __declspec(naked) static void vcvtr_s32_f64_imp(f64 d)

 # if defined(__VFP_FP__) && !defined(__SOFTFP__) && !(defined _DEBUG || defined DEBUG) && !defined(__CUDACC__)
 #  define CAROTENE_ROUND_FLT(value) {                              \
-    register union { f32 f; s32 i; } result;                    \
+    union { f32 f; s32 i; } result; \
    asm ("ftosis  %0, %1 \n" : "=w" (result.f) : "w" (value) ); \
    return result.i; }
 #  define CAROTENE_ROUND_DBL(value) {                      \
-    register union {f32 f; s32 i;} __tegra_result;      \
+    union {f32 f; s32 i;} __tegra_result; \
    asm (                                               \
        "ftosid  %0, %P1\n"                             \
        : "=w" (__tegra_result.f)                       \
@@ -109,7 +109,7 @@ void ScharrDeriv(const Size2D &size, s32 cn,
            internal::prefetch(srow0 + x);
            internal::prefetch(srow1 + x);
            internal::prefetch(srow2 + x);
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 7
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 7 && !defined(__clang__)
            __asm__ (
                "vld1.8 {d0}, [%[src0]]                                \n\t"
                "vld1.8 {d2}, [%[src2]]                                \n\t"
@@ -161,7 +161,7 @@ void ScharrDeriv(const Size2D &size, s32 cn,
        x = 0;
        for( ; x < roiw8; x += 8 )
        {
-#if defined(__GNUC__) && __GNUC__ == 4 && __GNUC_MINOR__ < 6
+#if !defined(__aarch64__) && defined(__GNUC__) && __GNUC__ == 4 &&  __GNUC_MINOR__ < 6 && !defined(__clang__)
            __asm__ (
                "vld1.16 {d4-d5}, [%[s2ptr]]                           \n\t"
                "vld1.16 {d8-d9}, [%[s4ptr]]                           \n\t"
@@ -1,8 +1,8 @@
-# Binary branch name: ffmpeg/master_20170704
-# Binaries were created for OpenCV: f670a9927026629a4083e05a1612f0adcad7727e
-set(FFMPEG_BINARIES_COMMIT "a86e53eb35737a50e5100e26af3aa1d29e810890")
-set(FFMPEG_FILE_HASH_BIN32 "79c35cc654778e66237444bc562afbca")
-set(FFMPEG_FILE_HASH_BIN64 "0dc72775ec3c14d1e049f51dc1280dbb")
+# Binary branch name: ffmpeg/master_20171009
+# Binaries were created for OpenCV: 8ac2c5d620b467d3f22802e96c88ddde6da707af
+set(FFMPEG_BINARIES_COMMIT "66b1fed06cf3510235f367f96aa26da5cb234a15")
+set(FFMPEG_FILE_HASH_BIN32 "3ae76b105113d944984b2351c61e21c6")
+set(FFMPEG_FILE_HASH_BIN64 "cf3bb5bc9d393b022ea7a42eb63e794d")
 set(FFMPEG_FILE_HASH_CMAKE "ec59008da403fb18ab3c1ed66aed583b")

 function(download_win_ffmpeg script_var)
@@ -6,7 +6,7 @@ project(${IPP_IW_LIBRARY})

 ocv_include_directories(${IPP_INCLUDE_DIRS} ${IPP_IW_PATH}/include)
 add_definitions(-DIW_BUILD)
-if(HAVE_IPP_ICV_ONLY)
+if(HAVE_IPP_ICV)
  add_definitions(-DICV_BASE)
 endif()

@@ -21,7 +21,10 @@ add_library(${IPP_IW_LIBRARY} STATIC ${lib_srcs} ${lib_hdrs})

 if(UNIX)
  if(CMAKE_COMPILER_IS_GNUCXX OR CV_ICC)
-    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fPIC -Wno-unused-function")
+    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wno-unused-function -Wno-missing-braces -Wno-missing-field-initializers")
+  endif()
+  if (CMAKE_C_COMPILER_ID MATCHES "Clang")
+    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wno-self-assign")
  endif()
 endif()

@@ -2,37 +2,37 @@ function(download_ippicv root_var)
  set(${root_var} "" PARENT_SCOPE)

  # Commit SHA in the opencv_3rdparty repo
-  set(IPPICV_COMMIT "a62e20676a60ee0ad6581e217fe7e4bada3b95db")
+  set(IPPICV_COMMIT "dfe3162c237af211e98b8960018b564bc209261d")
  # Define actual ICV versions
  if(APPLE)
    set(OPENCV_ICV_PLATFORM "macosx")
    set(OPENCV_ICV_PACKAGE_SUBDIR "ippicv_mac")
    if(X86_64)
-      set(OPENCV_ICV_NAME "ippicv_2017u2_mac_intel64_20170418.tgz")
-      set(OPENCV_ICV_HASH "0c25953c99dbb499ff502485a9356d8d")
+      set(OPENCV_ICV_NAME "ippicv_2017u3_mac_intel64_general_20170822.tgz")
+      set(OPENCV_ICV_HASH "c1ebb5dfa5b7f54b0c44e1917805a463")
    else()
-      set(OPENCV_ICV_NAME "ippicv_2017u2_mac_ia32_20170418.tgz")
-      set(OPENCV_ICV_HASH "5f225948f3f64067c681293c098d50d8")
+      set(OPENCV_ICV_NAME "ippicv_2017u3_mac_ia32_general_20170822.tgz")
+      set(OPENCV_ICV_HASH "49b05a669042753ae75895a445ebd612")
    endif()
  elseif((UNIX AND NOT ANDROID) OR (UNIX AND ANDROID_ABI MATCHES "x86"))
    set(OPENCV_ICV_PLATFORM "linux")
    set(OPENCV_ICV_PACKAGE_SUBDIR "ippicv_lnx")
    if(X86_64)
-      set(OPENCV_ICV_NAME "ippicv_2017u2_lnx_intel64_20170418.tgz")
-      set(OPENCV_ICV_HASH "87cbdeb627415d8e4bc811156289fa3a")
+      set(OPENCV_ICV_NAME "ippicv_2017u3_lnx_intel64_general_20170822.tgz")
+      set(OPENCV_ICV_HASH "4e0352ce96473837b1d671ce87f17359")
    else()
-      set(OPENCV_ICV_NAME "ippicv_2017u2_lnx_ia32_20170418.tgz")
-      set(OPENCV_ICV_HASH "f2cece00d802d4dea86df52ed095257e")
+      set(OPENCV_ICV_NAME "ippicv_2017u3_lnx_ia32_general_20170822.tgz")
+      set(OPENCV_ICV_HASH "dcdb0ba4b123f240596db1840cd59a76")
    endif()
  elseif(WIN32 AND NOT ARM)
    set(OPENCV_ICV_PLATFORM "windows")
    set(OPENCV_ICV_PACKAGE_SUBDIR "ippicv_win")
    if(X86_64)
-      set(OPENCV_ICV_NAME "ippicv_2017u2_win_intel64_20170418.zip")
-      set(OPENCV_ICV_HASH "75060a0c662c0800f48995b7e9b085f6")
+      set(OPENCV_ICV_NAME "ippicv_2017u3_win_intel64_general_20170822.zip")
+      set(OPENCV_ICV_HASH "0421e642bc7ad741a2236d3ec4190bdd")
    else()
-      set(OPENCV_ICV_NAME "ippicv_2017u2_win_ia32_20170418.zip")
-      set(OPENCV_ICV_HASH "60fcf3ccd9a2ebc9e432ffb5cb91638b")
+      set(OPENCV_ICV_NAME "ippicv_2017u3_win_ia32_general_20170822.zip")
+      set(OPENCV_ICV_HASH "8a7680ae352c192de2e2e34936164bd0")
    endif()
  else()
    return()
@@ -43,12 +43,6 @@ if(NOT WIN32)
  endif()
 endif()

-if(UNIX)
-  if(CMAKE_COMPILER_IS_GNUCXX OR CV_ICC)
-     set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fPIC")
-  endif()
-endif()
-
 set_target_properties(${ITT_LIBRARY} PROPERTIES
        OUTPUT_NAME ${ITT_LIBRARY}
        DEBUG_POSTFIX "${OPENCV_DEBUG_POSTFIX}"
@@ -25,16 +25,12 @@ endif(WIN32 AND NOT MINGW)

 ocv_warnings_disable(CMAKE_C_FLAGS -Wno-implicit-function-declaration -Wno-uninitialized -Wmissing-prototypes
                                   -Wno-unused-but-set-parameter -Wmissing-declarations -Wunused -Wshadow
-                                   -Wsign-compare -Wstrict-overflow)
-ocv_warnings_disable(CMAKE_C_FLAGS -Wunused-parameter) # clang
+                                   -Wsign-compare -Wstrict-overflow -Wpointer-compare
+                                   -Wabsolute-value  # clang on Linux
+)
+ocv_warnings_disable(CMAKE_C_FLAGS -Wunused-parameter -Wstrict-prototypes) # clang
 ocv_warnings_disable(CMAKE_C_FLAGS /wd4013 /wd4018 /wd4101 /wd4244 /wd4267 /wd4715) # vs2005

-if(UNIX)
-  if(CMAKE_COMPILER_IS_GNUCXX OR CV_ICC)
-    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fPIC")
-  endif()
-endif()
-
 set_target_properties(${JASPER_LIBRARY}
  PROPERTIES
  OUTPUT_NAME ${JASPER_LIBRARY}
@@ -28,17 +28,11 @@ endif()

 add_library(${JPEG_LIBRARY} STATIC ${lib_srcs} ${lib_hdrs})

-if(UNIX)
-  if(CMAKE_COMPILER_IS_GNUCXX OR CV_ICC)
-    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fPIC")
-  endif()
-endif()
-
 if(CMAKE_COMPILER_IS_GNUCXX)
  set_source_files_properties(jcdctmgr.c PROPERTIES COMPILE_FLAGS "-O1")
 endif()

-ocv_warnings_disable(CMAKE_C_FLAGS -Wcast-align -Wshadow -Wunused)
+ocv_warnings_disable(CMAKE_C_FLAGS -Wcast-align -Wshadow -Wunused -Wshift-negative-value)
 ocv_warnings_disable(CMAKE_C_FLAGS -Wunused-parameter) # clang
 ocv_warnings_disable(CMAKE_C_FLAGS /wd4013 /wd4244 /wd4267) # vs2005

@@ -1,8 +1,8 @@
 The Independent JPEG Group's JPEG software
 ==========================================

-README for release 9 of 13-Jan-2013
-===================================
+README for release 9b of 17-Jan-2016
+====================================

 This distribution contains the ninth public release of the Independent JPEG
 Group's free JPEG software.  You are welcome to redistribute this software and
@@ -14,7 +14,7 @@ Julian Minguillon, Luis Ortiz, George Phillips, Davide Rossi, Ge' Weijers,
 and other members of the Independent JPEG Group.

 IJG is not affiliated with the ISO/IEC JTC1/SC29/WG1 standards committee
-(also known as JPEG, together with ITU-T SG16).
+(previously known as JPEG, together with ITU-T SG16).


 DOCUMENTATION ROADMAP
@@ -60,7 +60,7 @@ OVERVIEW

 This package contains C software to implement JPEG image encoding, decoding,
 and transcoding.  JPEG (pronounced "jay-peg") is a standardized compression
-method for full-color and gray-scale images.
+method for full-color and grayscale images.

 This software implements JPEG baseline, extended-sequential, and progressive
 compression processes.  Provision is made for supporting all variants of these
@@ -115,7 +115,7 @@ with respect to this software, its quality, accuracy, merchantability, or
 fitness for a particular purpose.  This software is provided "AS IS", and you,
 its user, assume the entire risk as to its quality and accuracy.

-This software is copyright (C) 1991-2013, Thomas G. Lane, Guido Vollbeding.
+This software is copyright (C) 1991-2016, Thomas G. Lane, Guido Vollbeding.
 All Rights Reserved except as specified below.

 Permission is hereby granted to use, copy, modify, and distribute this
@@ -153,16 +153,11 @@ ltmain.sh).  Another support script, install-sh, is copyright by X Consortium
 but is also freely distributable.

 The IJG distribution formerly included code to read and write GIF files.
-To avoid entanglement with the Unisys LZW patent, GIF reading support has
-been removed altogether, and the GIF writer has been simplified to produce
-"uncompressed GIFs".  This technique does not use the LZW algorithm; the
-resulting GIF files are larger than usual, but are readable by all standard
-GIF decoders.
-
-We are required to state that
-    "The Graphics Interchange Format(c) is the Copyright property of
-    CompuServe Incorporated.  GIF(sm) is a Service Mark property of
-    CompuServe Incorporated."
+To avoid entanglement with the Unisys LZW patent (now expired), GIF reading
+support has been removed altogether, and the GIF writer has been simplified
+to produce "uncompressed GIFs".  This technique does not use the LZW
+algorithm; the resulting GIF files are larger than usual, but are readable
+by all standard GIF decoders.


 REFERENCES
@@ -176,8 +171,8 @@ The best short technical introduction to the JPEG compression algorithm is
 	Communications of the ACM, April 1991 (vol. 34 no. 4), pp. 30-44.
 (Adjacent articles in that issue discuss MPEG motion picture compression,
 applications of JPEG, and related topics.)  If you don't have the CACM issue
-handy, a PostScript file containing a revised version of Wallace's article is
-available at http://www.ijg.org/files/wallace.ps.gz.  The file (actually
+handy, a PDF file containing a revised version of Wallace's article is
+available at http://www.ijg.org/files/Wallace.JPEG.pdf.  The file (actually
 a preprint for an article that appeared in IEEE Trans. Consumer Electronics)
 omits the sample images that appeared in CACM, but it includes corrections
 and some added material.  Note: the Wallace article is copyright ACM and IEEE,
@@ -225,14 +220,13 @@ WG1 N 6080 with title "JPEG 9 Lossless Coding", June/July 2012, Paris,
 France.

 The JPEG standard does not specify all details of an interchangeable file
-format.  For the omitted details we follow the "JFIF" conventions, revision
-1.02.  JFIF 1.02 has been adopted as an Ecma International Technical Report
-and thus received a formal publication status.  It is available as a free
-download in PDF format from
-http://www.ecma-international.org/publications/techreports/E-TR-098.htm.
-A PostScript version of the JFIF document is available at
-http://www.ijg.org/files/jfif.ps.gz.  There is also a plain text version at
-http://www.ijg.org/files/jfif.txt.gz, but it is missing the figures.
+format.  For the omitted details we follow the "JFIF" conventions, version 2.
+JFIF version 1 has been adopted as Recommendation ITU-T T.871 (05/2011) :
+Information technology - Digital compression and coding of continuous-tone
+still images: JPEG File Interchange Format (JFIF).  It is available as a
+free download in PDF file format from http://www.itu.int/rec/T-REC-T.871.
+A PDF file of the older JFIF document is available at
+http://www.w3.org/Graphics/JPEG/jfif3.pdf.

 The TIFF 6.0 file format specification can be obtained by FTP from
 ftp://ftp.sgi.com/graphics/tiff/TIFF6.ps.gz.  The JPEG incorporation scheme
@@ -252,8 +246,8 @@ ARCHIVE LOCATIONS
 The "official" archive site for this software is www.ijg.org.
 The most recent released version can always be found there in
 directory "files".  This particular version will be archived as
-http://www.ijg.org/files/jpegsrc.v9.tar.gz, and in Windows-compatible
-"zip" archive format as http://www.ijg.org/files/jpegsr9.zip.
+http://www.ijg.org/files/jpegsrc.v9b.tar.gz, and in Windows-compatible
+"zip" archive format as http://www.ijg.org/files/jpegsr9b.zip.

 The JPEG FAQ (Frequently Asked Questions) article is a source of some
 general information about JPEG.
@@ -280,7 +274,7 @@ Thank to Thomas Wiegand and Gary Sullivan for inviting me to the
 Joint Video Team (MPEG & ITU) meeting in Geneva, Switzerland.

 Thank to Thomas Richter and Daniel Lee for inviting me to the
-ISO/IEC JTC1/SC29/WG1 (also known as JPEG, together with ITU-T SG16)
+ISO/IEC JTC1/SC29/WG1 (previously known as JPEG, together with ITU-T SG16)
 meeting in Berlin, Germany.

 Thank to John Korejwa and Massimo Ballerini for inviting me to
@@ -306,10 +300,10 @@ design and development of this singular software package.
 FILE FORMAT WARS
 ================

-The ISO/IEC JTC1/SC29/WG1 standards committee (also known as JPEG, together
-with ITU-T SG16) currently promotes different formats containing the name
-"JPEG" which is misleading because these formats are incompatible with
-original DCT-based JPEG and are based on faulty technologies.
+The ISO/IEC JTC1/SC29/WG1 standards committee (previously known as JPEG,
+together with ITU-T SG16) currently promotes different formats containing
+the name "JPEG" which is misleading because these formats are incompatible
+with original DCT-based JPEG and are based on faulty technologies.
 IJG therefore does not and will not support such momentary mistakes
 (see REFERENCES).
 There exist also distributions under the name "OpenJPEG" promoting such
@@ -322,9 +316,13 @@ Don't use an incompatible file format!
 (In any case, our decoder will remain capable of reading existing JPEG
 image files indefinitely.)

-Furthermore, the ISO committee pretends to be "responsible for the popular
-JPEG" in their public reports which is not true because they don't respond to
-actual requirements for the maintenance of the original JPEG specification.
+The ISO committee pretends to be "responsible for the popular JPEG" in their
+public reports which is not true because they don't respond to actual
+requirements for the maintenance of the original JPEG specification.
+Furthermore, the ISO committee pretends to "ensure interoperability" with
+their standards which is not true because their "standards" support only
+application-specific and proprietary use cases and contain mathematically
+incorrect code.

 There are currently different distributions in circulation containing the
 name "libjpeg" which is misleading because they don't have the features and
@@ -332,19 +330,46 @@ are incompatible with formats supported by actual IJG libjpeg distributions.
 One of those fakes is released by members of the ISO committee and just uses
 the name of libjpeg for misdirection of people, similar to the abuse of the
 name JPEG as described above, while having nothing in common with actual IJG
-libjpeg distributions.
-The other one claims to be a "derivative" or "fork" of the original libjpeg
-and violates the license conditions as described under LEGAL ISSUES above.
-We have no sympathy for the release of misleading and illegal distributions
-derived from obsolete code bases.
+libjpeg distributions and containing mathematically incorrect code.
+The other one claims to be a "derivative" or "fork" of the original libjpeg,
+but violates the license conditions as described under LEGAL ISSUES above
+and violates basic C programming properties.
+We have no sympathy for the release of misleading, incorrect and illegal
+distributions derived from obsolete code bases.
 Don't use an obsolete code base!

+According to the UCC (Uniform Commercial Code) law, IJG has the lawful and
+legal right to foreclose on certain standardization bodies and other
+institutions or corporations that knowingly perform substantial and
+systematic deceptive acts and practices, fraud, theft, and damaging of the
+value of the people of this planet without their knowing, willing and
+intentional consent.
+The titles, ownership, and rights of these institutions and all their assets
+are now duly secured and held in trust for the free people of this planet.
+People of the planet, on every country, may have a financial interest in
+the assets of these former principals, agents, and beneficiaries of the
+foreclosed institutions and corporations.
+IJG asserts what is: that each man, woman, and child has unalienable value
+and rights granted and deposited in them by the Creator and not any one of
+the people is subordinate to any artificial principality, corporate fiction
+or the special interest of another without their appropriate knowing,
+willing and intentional consent made by contract or accommodation agreement.
+IJG expresses that which already was.
+The people have already determined and demanded that public administration
+entities, national governments, and their supporting judicial systems must
+be fully transparent, accountable, and liable.
+IJG has secured the value for all concerned free people of the planet.
+
+A partial list of foreclosed institutions and corporations ("Hall of Shame")
+is currently prepared and will be published later.
+

 TO DO
 =====

 Version 9 is the second release of a new generation JPEG standard
-to overcome the limitations of the original JPEG specification.
+to overcome the limitations of the original JPEG specification,
+and is the first true source reference JPEG codec.
 More features are being prepared for coming releases...

 Please send bug reports, offers of help, etc. to jpeg-info@jpegclub.org.
@@ -1,6 +1,56 @@
 CHANGE LOG for Independent JPEG Group's JPEG software


+Version 9b  17-Jan-2016
+-----------------------
+
+Improvements and optimizations in DCT and color calculations.
+Normalize range limit array composition and access pattern.
+Thank to Sia Furler and Maddie Ziegler for inspiration.
+
+Use merged upsample with scaled DCT sizes larger than 8.
+Thank to Taylor Hatala for inspiration.
+
+Check for excessive comment lengths in argument parsing in wrjpgcom.c.
+Thank to Julian Cohen for hint.
+
+Add makefile.b32 for use with Borland C++ 32-bit (bcc32).
+Thank to Joe Slater for contribution.
+
+Document 'f' specifier for jpegtran -crop specification.
+Thank to Michele Martone for suggestion.
+
+Use defined value from header instead of hardwired number in rdswitch.c.
+Thank to Robert Sprowson for hint.
+
+
+Version 9a  19-Jan-2014
+-----------------------
+
+Add support for wide gamut color spaces (JFIF version 2).
+Improve clarity and accuracy in color conversion modules.
+Note: Requires rebuild of test images.
+
+Extend the bit depth support to all values from 8 to 12
+(BITS_IN_JSAMPLE configuration option in jmorecfg.h).
+jpegtran now supports N bits sample data precision with all N from 8 to 12
+in a single instance.  Thank to Roland Fassauer for inspiration.
+
+Try to resolve issues with new boolean type definition.
+Thank also to v4hn for suggestion.
+
+Enable option to use default Huffman tables for lossless compression
+(for hardware solution), and in this case improve lossless RGB compression
+with reversible color transform.  Thank to Benny Alexandar for hint.
+
+Extend the entropy decoding structure, so that extraneous bytes between
+compressed scan data and following marker can be reported correctly.
+Thank to Nigel Tao for hint.
+
+Add jpegtran -wipe option and extension for -crop.
+Thank to Andrew Senior, David Clunie, and Josef Schmid for suggestion.
+
+
 Version 9  13-Jan-2013
 ----------------------

@@ -37,8 +37,8 @@ jpeg_CreateCompress (j_compress_ptr cinfo, int version, size_t structsize)
  if (version != JPEG_LIB_VERSION)
    ERREXIT2(cinfo, JERR_BAD_LIB_VERSION, JPEG_LIB_VERSION, version);
  if (structsize != SIZEOF(struct jpeg_compress_struct))
-    ERREXIT2(cinfo, JERR_BAD_STRUCT_SIZE,
-             (int) SIZEOF(struct jpeg_compress_struct), (int) structsize);
+    ERREXIT2(cinfo, JERR_BAD_STRUCT_SIZE, 
+	     (int) SIZEOF(struct jpeg_compress_struct), (int) structsize);

  /* For debugging purposes, we zero the whole master structure.
   * But the application has already set the err pointer, and may have set
@@ -169,15 +169,15 @@ jpeg_finish_compress (j_compress_ptr cinfo)
    (*cinfo->master->prepare_for_pass) (cinfo);
    for (iMCU_row = 0; iMCU_row < cinfo->total_iMCU_rows; iMCU_row++) {
      if (cinfo->progress != NULL) {
-        cinfo->progress->pass_counter = (long) iMCU_row;
-        cinfo->progress->pass_limit = (long) cinfo->total_iMCU_rows;
-        (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
+	cinfo->progress->pass_counter = (long) iMCU_row;
+	cinfo->progress->pass_limit = (long) cinfo->total_iMCU_rows;
+	(*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
      }
      /* We bypass the main controller and invoke coef controller directly;
       * all work is being done from the coefficient buffer.
       */
      if (! (*cinfo->coef->compress_data) (cinfo, (JSAMPIMAGE) NULL))
-        ERREXIT(cinfo, JERR_CANT_SUSPEND);
+	ERREXIT(cinfo, JERR_CANT_SUSPEND);
    }
    (*cinfo->master->finish_pass) (cinfo);
  }
@@ -198,7 +198,7 @@ jpeg_finish_compress (j_compress_ptr cinfo)

 GLOBAL(void)
 jpeg_write_marker (j_compress_ptr cinfo, int marker,
-                   const JOCTET *dataptr, unsigned int datalen)
+		   const JOCTET *dataptr, unsigned int datalen)
 {
  JMETHOD(void, write_marker_byte, (j_compress_ptr info, int val));

@@ -2,6 +2,7 @@
 * jcapistd.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
+ * Modified 2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -75,7 +76,7 @@ jpeg_start_compress (j_compress_ptr cinfo, boolean write_all_tables)

 GLOBAL(JDIMENSION)
 jpeg_write_scanlines (j_compress_ptr cinfo, JSAMPARRAY scanlines,
-                      JDIMENSION num_lines)
+		      JDIMENSION num_lines)
 {
  JDIMENSION row_ctr, rows_left;

@@ -118,7 +119,7 @@ jpeg_write_scanlines (j_compress_ptr cinfo, JSAMPARRAY scanlines,

 GLOBAL(JDIMENSION)
 jpeg_write_raw_data (j_compress_ptr cinfo, JSAMPIMAGE data,
-                     JDIMENSION num_lines)
+		     JDIMENSION num_lines)
 {
  JDIMENSION lines_per_iMCU_row;

@@ -145,7 +146,7 @@ jpeg_write_raw_data (j_compress_ptr cinfo, JSAMPIMAGE data,
    (*cinfo->master->pass_startup) (cinfo);

  /* Verify that at least one iMCU row has been passed. */
-  lines_per_iMCU_row = cinfo->max_v_samp_factor * DCTSIZE;
+  lines_per_iMCU_row = cinfo->max_v_samp_factor * cinfo->min_DCT_v_scaled_size;
  if (num_lines < lines_per_iMCU_row)
    ERREXIT(cinfo, JERR_BUFFER_SIZE);

@@ -1,7 +1,7 @@
 /*
 * jcarith.c
 *
- * Developed 1997-2012 by Guido Vollbeding.
+ * Developed 1997-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -103,9 +103,9 @@ typedef arith_entropy_encoder * arith_entropy_ptr;
 #ifdef RIGHT_SHIFT_IS_UNSIGNED
 #define ISHIFT_TEMPS	int ishift_temp;
 #define IRIGHT_SHIFT(x,shft)  \
-        ((ishift_temp = (x)) < 0 ? \
-         (ishift_temp >> (shft)) | ((~0) << (16-(shft))) : \
-         (ishift_temp >> (shft)))
+	((ishift_temp = (x)) < 0 ? \
+	 (ishift_temp >> (shft)) | ((~0) << (16-(shft))) : \
+	 (ishift_temp >> (shft)))
 #else
 #define ISHIFT_TEMPS
 #define IRIGHT_SHIFT(x,shft)	((x) >> (shft))
@@ -149,11 +149,11 @@ finish_pass (j_compress_ptr cinfo)
    /* One final overflow has to be handled */
    if (e->buffer >= 0) {
      if (e->zc)
-        do emit_byte(0x00, cinfo);
-        while (--e->zc);
+	do emit_byte(0x00, cinfo);
+	while (--e->zc);
      emit_byte(e->buffer + 1, cinfo);
      if (e->buffer + 1 == 0xFF)
-        emit_byte(0x00, cinfo);
+	emit_byte(0x00, cinfo);
    }
    e->zc += e->sc;  /* carry-over converts stacked 0xFF bytes to 0x00 */
    e->sc = 0;
@@ -162,17 +162,17 @@ finish_pass (j_compress_ptr cinfo)
      ++e->zc;
    else if (e->buffer >= 0) {
      if (e->zc)
-        do emit_byte(0x00, cinfo);
-        while (--e->zc);
+	do emit_byte(0x00, cinfo);
+	while (--e->zc);
      emit_byte(e->buffer, cinfo);
    }
    if (e->sc) {
      if (e->zc)
-        do emit_byte(0x00, cinfo);
-        while (--e->zc);
+	do emit_byte(0x00, cinfo);
+	while (--e->zc);
      do {
-        emit_byte(0xFF, cinfo);
-        emit_byte(0x00, cinfo);
+	emit_byte(0xFF, cinfo);
+	emit_byte(0x00, cinfo);
      } while (--e->sc);
    }
  }
@@ -187,7 +187,7 @@ finish_pass (j_compress_ptr cinfo)
    if (e->c & 0x7F800L) {
      emit_byte((e->c >> 11) & 0xFF, cinfo);
      if (((e->c >> 11) & 0xFF) == 0xFF)
-        emit_byte(0x00, cinfo);
+	emit_byte(0x00, cinfo);
    }
  }
 }
@@ -216,7 +216,7 @@ finish_pass (j_compress_ptr cinfo)
 */

 LOCAL(void)
-arith_encode (j_compress_ptr cinfo, unsigned char *st, int val)
+arith_encode (j_compress_ptr cinfo, unsigned char *st, int val) 
 {
  register arith_entropy_ptr e = (arith_entropy_ptr) cinfo->entropy;
  register unsigned char nl, nm;
@@ -266,43 +266,43 @@ arith_encode (j_compress_ptr cinfo, unsigned char *st, int val)
      /* Another byte is ready for output */
      temp = e->c >> 19;
      if (temp > 0xFF) {
-        /* Handle overflow over all stacked 0xFF bytes */
-        if (e->buffer >= 0) {
-          if (e->zc)
-            do emit_byte(0x00, cinfo);
-            while (--e->zc);
-          emit_byte(e->buffer + 1, cinfo);
-          if (e->buffer + 1 == 0xFF)
-            emit_byte(0x00, cinfo);
-        }
-        e->zc += e->sc;  /* carry-over converts stacked 0xFF bytes to 0x00 */
-        e->sc = 0;
-        /* Note: The 3 spacer bits in the C register guarantee
-         * that the new buffer byte can't be 0xFF here
-         * (see page 160 in the P&M JPEG book). */
-        e->buffer = temp & 0xFF;  /* new output byte, might overflow later */
+	/* Handle overflow over all stacked 0xFF bytes */
+	if (e->buffer >= 0) {
+	  if (e->zc)
+	    do emit_byte(0x00, cinfo);
+	    while (--e->zc);
+	  emit_byte(e->buffer + 1, cinfo);
+	  if (e->buffer + 1 == 0xFF)
+	    emit_byte(0x00, cinfo);
+	}
+	e->zc += e->sc;  /* carry-over converts stacked 0xFF bytes to 0x00 */
+	e->sc = 0;
+	/* Note: The 3 spacer bits in the C register guarantee
+	 * that the new buffer byte can't be 0xFF here
+	 * (see page 160 in the P&M JPEG book). */
+	e->buffer = temp & 0xFF;  /* new output byte, might overflow later */
      } else if (temp == 0xFF) {
-        ++e->sc;  /* stack 0xFF byte (which might overflow later) */
+	++e->sc;  /* stack 0xFF byte (which might overflow later) */
      } else {
-        /* Output all stacked 0xFF bytes, they will not overflow any more */
-        if (e->buffer == 0)
-          ++e->zc;
-        else if (e->buffer >= 0) {
-          if (e->zc)
-            do emit_byte(0x00, cinfo);
-            while (--e->zc);
-          emit_byte(e->buffer, cinfo);
-        }
-        if (e->sc) {
-          if (e->zc)
-            do emit_byte(0x00, cinfo);
-            while (--e->zc);
-          do {
-            emit_byte(0xFF, cinfo);
-            emit_byte(0x00, cinfo);
-          } while (--e->sc);
-        }
-        e->buffer = temp & 0xFF;  /* new output byte (can still overflow) */
+	/* Output all stacked 0xFF bytes, they will not overflow any more */
+	if (e->buffer == 0)
+	  ++e->zc;
+	else if (e->buffer >= 0) {
+	  if (e->zc)
+	    do emit_byte(0x00, cinfo);
+	    while (--e->zc);
+	  emit_byte(e->buffer, cinfo);
+	}
+	if (e->sc) {
+	  if (e->zc)
+	    do emit_byte(0x00, cinfo);
+	    while (--e->zc);
+	  do {
+	    emit_byte(0xFF, cinfo);
+	    emit_byte(0x00, cinfo);
+	  } while (--e->sc);
+	}
+	e->buffer = temp & 0xFF;  /* new output byte (can still overflow) */
      }
      e->c &= 0x7FFFFL;
      e->ct += 8;
@@ -362,7 +362,6 @@ METHODDEF(boolean)
 encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
 {
  arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy;
-  JBLOCKROW block;
  unsigned char *st;
  int blkn, ci, tbl;
  int v, v2, m;
@@ -381,14 +380,13 @@ encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)

  /* Encode the MCU data blocks */
  for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) {
-    block = MCU_data[blkn];
    ci = cinfo->MCU_membership[blkn];
    tbl = cinfo->cur_comp_info[ci]->dc_tbl_no;

    /* Compute the DC value after the required point transform by Al.
     * This is simply an arithmetic right shift.
     */
-    m = IRIGHT_SHIFT((int) ((*block)[0]), cinfo->Al);
+    m = IRIGHT_SHIFT((int) (MCU_data[blkn][0][0]), cinfo->Al);

    /* Sections F.1.4.1 & F.1.4.4.1: Encoding of DC coefficients */

@@ -405,38 +403,38 @@ encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
      /* Figure F.6: Encoding nonzero value v */
      /* Figure F.7: Encoding the sign of v */
      if (v > 0) {
-        arith_encode(cinfo, st + 1, 0);	/* Table F.4: SS = S0 + 1 */
-        st += 2;			/* Table F.4: SP = S0 + 2 */
-        entropy->dc_context[ci] = 4;	/* small positive diff category */
+	arith_encode(cinfo, st + 1, 0);	/* Table F.4: SS = S0 + 1 */
+	st += 2;			/* Table F.4: SP = S0 + 2 */
+	entropy->dc_context[ci] = 4;	/* small positive diff category */
      } else {
-        v = -v;
-        arith_encode(cinfo, st + 1, 1);	/* Table F.4: SS = S0 + 1 */
-        st += 3;			/* Table F.4: SN = S0 + 3 */
-        entropy->dc_context[ci] = 8;	/* small negative diff category */
+	v = -v;
+	arith_encode(cinfo, st + 1, 1);	/* Table F.4: SS = S0 + 1 */
+	st += 3;			/* Table F.4: SN = S0 + 3 */
+	entropy->dc_context[ci] = 8;	/* small negative diff category */
      }
      /* Figure F.8: Encoding the magnitude category of v */
      m = 0;
      if (v -= 1) {
-        arith_encode(cinfo, st, 1);
-        m = 1;
-        v2 = v;
-        st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */
-        while (v2 >>= 1) {
-          arith_encode(cinfo, st, 1);
-          m <<= 1;
-          st += 1;
-        }
+	arith_encode(cinfo, st, 1);
+	m = 1;
+	v2 = v;
+	st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */
+	while (v2 >>= 1) {
+	  arith_encode(cinfo, st, 1);
+	  m <<= 1;
+	  st += 1;
+	}
      }
      arith_encode(cinfo, st, 0);
      /* Section F.1.4.4.1.2: Establish dc_context conditioning category */
      if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1))
-        entropy->dc_context[ci] = 0;	/* zero diff category */
+	entropy->dc_context[ci] = 0;	/* zero diff category */
      else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1))
-        entropy->dc_context[ci] += 8;	/* large diff category */
+	entropy->dc_context[ci] += 8;	/* large diff category */
      /* Figure F.9: Encoding the magnitude bit pattern of v */
      st += 14;
      while (m >>= 1)
-        arith_encode(cinfo, st, (m & v) ? 1 : 0);
+	arith_encode(cinfo, st, (m & v) ? 1 : 0);
    }
  }

@@ -453,11 +451,11 @@ METHODDEF(boolean)
 encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
 {
  arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy;
+  const int * natural_order;
  JBLOCKROW block;
  unsigned char *st;
  int tbl, k, ke;
  int v, v2, m;
-  const int * natural_order;

  /* Emit restart marker if needed */
  if (cinfo->restart_interval) {
@@ -499,18 +497,18 @@ encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
    arith_encode(cinfo, st, 0);		/* EOB decision */
    for (;;) {
      if ((v = (*block)[natural_order[++k]]) >= 0) {
-        if (v >>= cinfo->Al) {
-          arith_encode(cinfo, st + 1, 1);
-          arith_encode(cinfo, entropy->fixed_bin, 0);
-          break;
-        }
+	if (v >>= cinfo->Al) {
+	  arith_encode(cinfo, st + 1, 1);
+	  arith_encode(cinfo, entropy->fixed_bin, 0);
+	  break;
+	}
      } else {
-        v = -v;
-        if (v >>= cinfo->Al) {
-          arith_encode(cinfo, st + 1, 1);
-          arith_encode(cinfo, entropy->fixed_bin, 1);
-          break;
-        }
+	v = -v;
+	if (v >>= cinfo->Al) {
+	  arith_encode(cinfo, st + 1, 1);
+	  arith_encode(cinfo, entropy->fixed_bin, 1);
+	  break;
+	}
      }
      arith_encode(cinfo, st + 1, 0);
      st += 3;
@@ -523,15 +521,15 @@ encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
      m = 1;
      v2 = v;
      if (v2 >>= 1) {
-        arith_encode(cinfo, st, 1);
-        m <<= 1;
-        st = entropy->ac_stats[tbl] +
-             (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
-        while (v2 >>= 1) {
-          arith_encode(cinfo, st, 1);
-          m <<= 1;
-          st += 1;
-        }
+	arith_encode(cinfo, st, 1);
+	m <<= 1;
+	st = entropy->ac_stats[tbl] +
+	     (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
+	while (v2 >>= 1) {
+	  arith_encode(cinfo, st, 1);
+	  m <<= 1;
+	  st += 1;
+	}
      }
    }
    arith_encode(cinfo, st, 0);
@@ -552,6 +550,8 @@ encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)

 /*
 * MCU encoding for DC successive approximation refinement scan.
+ * Note: we assume such scans can be multi-component,
+ * although the spec is not very clear on the point.
 */

 METHODDEF(boolean)
@@ -593,11 +593,11 @@ METHODDEF(boolean)
 encode_mcu_AC_refine (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
 {
  arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy;
+  const int * natural_order;
  JBLOCKROW block;
  unsigned char *st;
  int tbl, k, ke, kex;
  int v;
-  const int * natural_order;

  /* Emit restart marker if needed */
  if (cinfo->restart_interval) {
@@ -649,26 +649,26 @@ encode_mcu_AC_refine (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
      arith_encode(cinfo, st, 0);	/* EOB decision */
    for (;;) {
      if ((v = (*block)[natural_order[++k]]) >= 0) {
-        if (v >>= cinfo->Al) {
-          if (v >> 1)			/* previously nonzero coef */
-            arith_encode(cinfo, st + 2, (v & 1));
-          else {			/* newly nonzero coef */
-            arith_encode(cinfo, st + 1, 1);
-            arith_encode(cinfo, entropy->fixed_bin, 0);
-          }
-          break;
-        }
+	if (v >>= cinfo->Al) {
+	  if (v >> 1)			/* previously nonzero coef */
+	    arith_encode(cinfo, st + 2, (v & 1));
+	  else {			/* newly nonzero coef */
+	    arith_encode(cinfo, st + 1, 1);
+	    arith_encode(cinfo, entropy->fixed_bin, 0);
+	  }
+	  break;
+	}
      } else {
-        v = -v;
-        if (v >>= cinfo->Al) {
-          if (v >> 1)			/* previously nonzero coef */
-            arith_encode(cinfo, st + 2, (v & 1));
-          else {			/* newly nonzero coef */
-            arith_encode(cinfo, st + 1, 1);
-            arith_encode(cinfo, entropy->fixed_bin, 1);
-          }
-          break;
-        }
+	v = -v;
+	if (v >>= cinfo->Al) {
+	  if (v >> 1)			/* previously nonzero coef */
+	    arith_encode(cinfo, st + 2, (v & 1));
+	  else {			/* newly nonzero coef */
+	    arith_encode(cinfo, st + 1, 1);
+	    arith_encode(cinfo, entropy->fixed_bin, 1);
+	  }
+	  break;
+	}
      }
      arith_encode(cinfo, st + 1, 0);
      st += 3;
@@ -692,12 +692,13 @@ METHODDEF(boolean)
 encode_mcu (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
 {
  arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy;
-  jpeg_component_info * compptr;
+  const int * natural_order;
  JBLOCKROW block;
  unsigned char *st;
-  int blkn, ci, tbl, k, ke;
+  int tbl, k, ke;
  int v, v2, m;
-  const int * natural_order;
+  int blkn, ci;
+  jpeg_component_info * compptr;

  /* Emit restart marker if needed */
  if (cinfo->restart_interval) {
@@ -735,38 +736,38 @@ encode_mcu (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
      /* Figure F.6: Encoding nonzero value v */
      /* Figure F.7: Encoding the sign of v */
      if (v > 0) {
-        arith_encode(cinfo, st + 1, 0);	/* Table F.4: SS = S0 + 1 */
-        st += 2;			/* Table F.4: SP = S0 + 2 */
-        entropy->dc_context[ci] = 4;	/* small positive diff category */
+	arith_encode(cinfo, st + 1, 0);	/* Table F.4: SS = S0 + 1 */
+	st += 2;			/* Table F.4: SP = S0 + 2 */
+	entropy->dc_context[ci] = 4;	/* small positive diff category */
      } else {
-        v = -v;
-        arith_encode(cinfo, st + 1, 1);	/* Table F.4: SS = S0 + 1 */
-        st += 3;			/* Table F.4: SN = S0 + 3 */
-        entropy->dc_context[ci] = 8;	/* small negative diff category */
+	v = -v;
+	arith_encode(cinfo, st + 1, 1);	/* Table F.4: SS = S0 + 1 */
+	st += 3;			/* Table F.4: SN = S0 + 3 */
+	entropy->dc_context[ci] = 8;	/* small negative diff category */
      }
      /* Figure F.8: Encoding the magnitude category of v */
      m = 0;
      if (v -= 1) {
-        arith_encode(cinfo, st, 1);
-        m = 1;
-        v2 = v;
-        st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */
-        while (v2 >>= 1) {
-          arith_encode(cinfo, st, 1);
-          m <<= 1;
-          st += 1;
-        }
+	arith_encode(cinfo, st, 1);
+	m = 1;
+	v2 = v;
+	st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */
+	while (v2 >>= 1) {
+	  arith_encode(cinfo, st, 1);
+	  m <<= 1;
+	  st += 1;
+	}
      }
      arith_encode(cinfo, st, 0);
      /* Section F.1.4.4.1.2: Establish dc_context conditioning category */
      if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1))
-        entropy->dc_context[ci] = 0;	/* zero diff category */
+	entropy->dc_context[ci] = 0;	/* zero diff category */
      else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1))
-        entropy->dc_context[ci] += 8;	/* large diff category */
+	entropy->dc_context[ci] += 8;	/* large diff category */
      /* Figure F.9: Encoding the magnitude bit pattern of v */
      st += 14;
      while (m >>= 1)
-        arith_encode(cinfo, st, (m & v) ? 1 : 0);
+	arith_encode(cinfo, st, (m & v) ? 1 : 0);
    }

    /* Sections F.1.4.2 & F.1.4.4.2: Encoding of AC coefficients */
@@ -784,42 +785,42 @@ encode_mcu (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
      st = entropy->ac_stats[tbl] + 3 * k;
      arith_encode(cinfo, st, 0);	/* EOB decision */
      while ((v = (*block)[natural_order[++k]]) == 0) {
-        arith_encode(cinfo, st + 1, 0);
-        st += 3;
+	arith_encode(cinfo, st + 1, 0);
+	st += 3;
      }
      arith_encode(cinfo, st + 1, 1);
      /* Figure F.6: Encoding nonzero value v */
      /* Figure F.7: Encoding the sign of v */
      if (v > 0) {
-        arith_encode(cinfo, entropy->fixed_bin, 0);
+	arith_encode(cinfo, entropy->fixed_bin, 0);
      } else {
-        v = -v;
-        arith_encode(cinfo, entropy->fixed_bin, 1);
+	v = -v;
+	arith_encode(cinfo, entropy->fixed_bin, 1);
      }
      st += 2;
      /* Figure F.8: Encoding the magnitude category of v */
      m = 0;
      if (v -= 1) {
-        arith_encode(cinfo, st, 1);
-        m = 1;
-        v2 = v;
-        if (v2 >>= 1) {
-          arith_encode(cinfo, st, 1);
-          m <<= 1;
-          st = entropy->ac_stats[tbl] +
-               (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
-          while (v2 >>= 1) {
-            arith_encode(cinfo, st, 1);
-            m <<= 1;
-            st += 1;
-          }
-        }
+	arith_encode(cinfo, st, 1);
+	m = 1;
+	v2 = v;
+	if (v2 >>= 1) {
+	  arith_encode(cinfo, st, 1);
+	  m <<= 1;
+	  st = entropy->ac_stats[tbl] +
+	       (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
+	  while (v2 >>= 1) {
+	    arith_encode(cinfo, st, 1);
+	    m <<= 1;
+	    st += 1;
+	  }
+	}
      }
      arith_encode(cinfo, st, 0);
      /* Figure F.9: Encoding the magnitude bit pattern of v */
      st += 14;
      while (m >>= 1)
-        arith_encode(cinfo, st, (m & v) ? 1 : 0);
+	arith_encode(cinfo, st, (m & v) ? 1 : 0);
    }
    /* Encode EOB decision only if k < cinfo->lim_Se */
    if (k < cinfo->lim_Se) {
@@ -856,14 +857,14 @@ start_pass (j_compress_ptr cinfo, boolean gather_statistics)
  if (cinfo->progressive_mode) {
    if (cinfo->Ah == 0) {
      if (cinfo->Ss == 0)
-        entropy->pub.encode_mcu = encode_mcu_DC_first;
+	entropy->pub.encode_mcu = encode_mcu_DC_first;
      else
-        entropy->pub.encode_mcu = encode_mcu_AC_first;
+	entropy->pub.encode_mcu = encode_mcu_AC_first;
    } else {
      if (cinfo->Ss == 0)
-        entropy->pub.encode_mcu = encode_mcu_DC_refine;
+	entropy->pub.encode_mcu = encode_mcu_DC_refine;
      else
-        entropy->pub.encode_mcu = encode_mcu_AC_refine;
+	entropy->pub.encode_mcu = encode_mcu_AC_refine;
    }
  } else
    entropy->pub.encode_mcu = encode_mcu;
@@ -875,10 +876,10 @@ start_pass (j_compress_ptr cinfo, boolean gather_statistics)
    if (cinfo->Ss == 0 && cinfo->Ah == 0) {
      tbl = compptr->dc_tbl_no;
      if (tbl < 0 || tbl >= NUM_ARITH_TBLS)
-        ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
+	ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
      if (entropy->dc_stats[tbl] == NULL)
-        entropy->dc_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
-          ((j_common_ptr) cinfo, JPOOL_IMAGE, DC_STAT_BINS);
+	entropy->dc_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
+	  ((j_common_ptr) cinfo, JPOOL_IMAGE, DC_STAT_BINS);
      MEMZERO(entropy->dc_stats[tbl], DC_STAT_BINS);
      /* Initialize DC predictions to 0 */
      entropy->last_dc_val[ci] = 0;
@@ -888,15 +889,15 @@ start_pass (j_compress_ptr cinfo, boolean gather_statistics)
    if (cinfo->Se) {
      tbl = compptr->ac_tbl_no;
      if (tbl < 0 || tbl >= NUM_ARITH_TBLS)
-        ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
+	ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
      if (entropy->ac_stats[tbl] == NULL)
-        entropy->ac_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
-          ((j_common_ptr) cinfo, JPOOL_IMAGE, AC_STAT_BINS);
+	entropy->ac_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
+	  ((j_common_ptr) cinfo, JPOOL_IMAGE, AC_STAT_BINS);
      MEMZERO(entropy->ac_stats[tbl], AC_STAT_BINS);
 #ifdef CALCULATE_SPECTRAL_CONDITIONING
      if (cinfo->progressive_mode)
-        /* Section G.1.3.2: Set appropriate arithmetic conditioning value Kx */
-        cinfo->arith_ac_K[tbl] = cinfo->Ss + ((8 + cinfo->Se - cinfo->Ss) >> 4);
+	/* Section G.1.3.2: Set appropriate arithmetic conditioning value Kx */
+	cinfo->arith_ac_K[tbl] = cinfo->Ss + ((8 + cinfo->Se - cinfo->Ss) >> 4);
 #endif
    }
  }
@@ -927,7 +928,7 @@ jinit_arith_encoder (j_compress_ptr cinfo)

  entropy = (arith_entropy_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(arith_entropy_encoder));
+				SIZEOF(arith_entropy_encoder));
  cinfo->entropy = &entropy->pub;
  entropy->pub.start_pass = start_pass;
  entropy->pub.finish_pass = finish_pass;
@@ -156,7 +156,7 @@ compress_data (j_compress_ptr cinfo, JSAMPIMAGE input_buf)
  for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row;
       yoffset++) {
    for (MCU_col_num = coef->mcu_ctr; MCU_col_num <= last_MCU_col;
-         MCU_col_num++) {
+	 MCU_col_num++) {
      /* Determine where data comes from in input_buf and do the DCT thing.
       * Each call on forward_DCT processes a horizontal row of DCT blocks
       * as wide as an MCU; we rely on having allocated the MCU_buffer[] blocks
@@ -168,48 +168,48 @@ compress_data (j_compress_ptr cinfo, JSAMPIMAGE input_buf)
       */
      blkn = 0;
      for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
-        compptr = cinfo->cur_comp_info[ci];
-        forward_DCT = cinfo->fdct->forward_DCT[compptr->component_index];
-        blockcnt = (MCU_col_num < last_MCU_col) ? compptr->MCU_width
-                                                : compptr->last_col_width;
-        xpos = MCU_col_num * compptr->MCU_sample_width;
-        ypos = yoffset * compptr->DCT_v_scaled_size;
-        /* ypos == (yoffset+yindex) * DCTSIZE */
-        for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
-          if (coef->iMCU_row_num < last_iMCU_row ||
-              yoffset+yindex < compptr->last_row_height) {
-            (*forward_DCT) (cinfo, compptr,
-                            input_buf[compptr->component_index],
-                            coef->MCU_buffer[blkn],
-                            ypos, xpos, (JDIMENSION) blockcnt);
-            if (blockcnt < compptr->MCU_width) {
-              /* Create some dummy blocks at the right edge of the image. */
-              FMEMZERO((void FAR *) coef->MCU_buffer[blkn + blockcnt],
-                       (compptr->MCU_width - blockcnt) * SIZEOF(JBLOCK));
-              for (bi = blockcnt; bi < compptr->MCU_width; bi++) {
-                coef->MCU_buffer[blkn+bi][0][0] = coef->MCU_buffer[blkn+bi-1][0][0];
-              }
-            }
-          } else {
-            /* Create a row of dummy blocks at the bottom of the image. */
-            FMEMZERO((void FAR *) coef->MCU_buffer[blkn],
-                     compptr->MCU_width * SIZEOF(JBLOCK));
-            for (bi = 0; bi < compptr->MCU_width; bi++) {
-              coef->MCU_buffer[blkn+bi][0][0] = coef->MCU_buffer[blkn-1][0][0];
-            }
-          }
-          blkn += compptr->MCU_width;
-          ypos += compptr->DCT_v_scaled_size;
-        }
+	compptr = cinfo->cur_comp_info[ci];
+	forward_DCT = cinfo->fdct->forward_DCT[compptr->component_index];
+	blockcnt = (MCU_col_num < last_MCU_col) ? compptr->MCU_width
+						: compptr->last_col_width;
+	xpos = MCU_col_num * compptr->MCU_sample_width;
+	ypos = yoffset * compptr->DCT_v_scaled_size;
+	/* ypos == (yoffset+yindex) * DCTSIZE */
+	for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
+	  if (coef->iMCU_row_num < last_iMCU_row ||
+	      yoffset+yindex < compptr->last_row_height) {
+	    (*forward_DCT) (cinfo, compptr,
+			    input_buf[compptr->component_index],
+			    coef->MCU_buffer[blkn],
+			    ypos, xpos, (JDIMENSION) blockcnt);
+	    if (blockcnt < compptr->MCU_width) {
+	      /* Create some dummy blocks at the right edge of the image. */
+	      FMEMZERO((void FAR *) coef->MCU_buffer[blkn + blockcnt],
+		       (compptr->MCU_width - blockcnt) * SIZEOF(JBLOCK));
+	      for (bi = blockcnt; bi < compptr->MCU_width; bi++) {
+		coef->MCU_buffer[blkn+bi][0][0] = coef->MCU_buffer[blkn+bi-1][0][0];
+	      }
+	    }
+	  } else {
+	    /* Create a row of dummy blocks at the bottom of the image. */
+	    FMEMZERO((void FAR *) coef->MCU_buffer[blkn],
+		     compptr->MCU_width * SIZEOF(JBLOCK));
+	    for (bi = 0; bi < compptr->MCU_width; bi++) {
+	      coef->MCU_buffer[blkn+bi][0][0] = coef->MCU_buffer[blkn-1][0][0];
+	    }
+	  }
+	  blkn += compptr->MCU_width;
+	  ypos += compptr->DCT_v_scaled_size;
+	}
      }
      /* Try to write the MCU.  In event of a suspension failure, we will
       * re-DCT the MCU on restart (a bit inefficient, could be fixed...)
       */
      if (! (*cinfo->entropy->encode_mcu) (cinfo, coef->MCU_buffer)) {
-        /* Suspension forced; update state counters and exit */
-        coef->MCU_vert_offset = yoffset;
-        coef->mcu_ctr = MCU_col_num;
-        return FALSE;
+	/* Suspension forced; update state counters and exit */
+	coef->MCU_vert_offset = yoffset;
+	coef->mcu_ctr = MCU_col_num;
+	return FALSE;
      }
    }
    /* Completed an MCU row, but perhaps not an iMCU row */
@@ -286,16 +286,16 @@ compress_first_pass (j_compress_ptr cinfo, JSAMPIMAGE input_buf)
    for (block_row = 0; block_row < block_rows; block_row++) {
      thisblockrow = buffer[block_row];
      (*forward_DCT) (cinfo, compptr, input_buf[ci], thisblockrow,
-                      (JDIMENSION) (block_row * compptr->DCT_v_scaled_size),
-                      (JDIMENSION) 0, blocks_across);
+		      (JDIMENSION) (block_row * compptr->DCT_v_scaled_size),
+		      (JDIMENSION) 0, blocks_across);
      if (ndummy > 0) {
-        /* Create dummy blocks at the right edge of the image. */
-        thisblockrow += blocks_across; /* => first dummy block */
-        FMEMZERO((void FAR *) thisblockrow, ndummy * SIZEOF(JBLOCK));
-        lastDC = thisblockrow[-1][0];
-        for (bi = 0; bi < ndummy; bi++) {
-          thisblockrow[bi][0] = lastDC;
-        }
+	/* Create dummy blocks at the right edge of the image. */
+	thisblockrow += blocks_across; /* => first dummy block */
+	FMEMZERO((void FAR *) thisblockrow, ndummy * SIZEOF(JBLOCK));
+	lastDC = thisblockrow[-1][0];
+	for (bi = 0; bi < ndummy; bi++) {
+	  thisblockrow[bi][0] = lastDC;
+	}
      }
    }
    /* If at end of image, create dummy block rows as needed.
@@ -307,19 +307,19 @@ compress_first_pass (j_compress_ptr cinfo, JSAMPIMAGE input_buf)
      blocks_across += ndummy;	/* include lower right corner */
      MCUs_across = blocks_across / h_samp_factor;
      for (block_row = block_rows; block_row < compptr->v_samp_factor;
-           block_row++) {
-        thisblockrow = buffer[block_row];
-        lastblockrow = buffer[block_row-1];
-        FMEMZERO((void FAR *) thisblockrow,
-                 (size_t) (blocks_across * SIZEOF(JBLOCK)));
-        for (MCUindex = 0; MCUindex < MCUs_across; MCUindex++) {
-          lastDC = lastblockrow[h_samp_factor-1][0];
-          for (bi = 0; bi < h_samp_factor; bi++) {
-            thisblockrow[bi][0] = lastDC;
-          }
-          thisblockrow += h_samp_factor; /* advance to next MCU in row */
-          lastblockrow += h_samp_factor;
-        }
+	   block_row++) {
+	thisblockrow = buffer[block_row];
+	lastblockrow = buffer[block_row-1];
+	FMEMZERO((void FAR *) thisblockrow,
+		 (size_t) (blocks_across * SIZEOF(JBLOCK)));
+	for (MCUindex = 0; MCUindex < MCUs_across; MCUindex++) {
+	  lastDC = lastblockrow[h_samp_factor-1][0];
+	  for (bi = 0; bi < h_samp_factor; bi++) {
+	    thisblockrow[bi][0] = lastDC;
+	  }
+	  thisblockrow += h_samp_factor; /* advance to next MCU in row */
+	  lastblockrow += h_samp_factor;
+	}
      }
    }
  }
@@ -369,25 +369,25 @@ compress_output (j_compress_ptr cinfo, JSAMPIMAGE input_buf)
  for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row;
       yoffset++) {
    for (MCU_col_num = coef->mcu_ctr; MCU_col_num < cinfo->MCUs_per_row;
-         MCU_col_num++) {
+	 MCU_col_num++) {
      /* Construct list of pointers to DCT blocks belonging to this MCU */
      blkn = 0;			/* index of current DCT block within MCU */
      for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
-        compptr = cinfo->cur_comp_info[ci];
-        start_col = MCU_col_num * compptr->MCU_width;
-        for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
-          buffer_ptr = buffer[ci][yindex+yoffset] + start_col;
-          for (xindex = 0; xindex < compptr->MCU_width; xindex++) {
-            coef->MCU_buffer[blkn++] = buffer_ptr++;
-          }
-        }
+	compptr = cinfo->cur_comp_info[ci];
+	start_col = MCU_col_num * compptr->MCU_width;
+	for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
+	  buffer_ptr = buffer[ci][yindex+yoffset] + start_col;
+	  for (xindex = 0; xindex < compptr->MCU_width; xindex++) {
+	    coef->MCU_buffer[blkn++] = buffer_ptr++;
+	  }
+	}
      }
      /* Try to write the MCU. */
      if (! (*cinfo->entropy->encode_mcu) (cinfo, coef->MCU_buffer)) {
-        /* Suspension forced; update state counters and exit */
-        coef->MCU_vert_offset = yoffset;
-        coef->mcu_ctr = MCU_col_num;
-        return FALSE;
+	/* Suspension forced; update state counters and exit */
+	coef->MCU_vert_offset = yoffset;
+	coef->mcu_ctr = MCU_col_num;
+	return FALSE;
      }
    }
    /* Completed an MCU row, but perhaps not an iMCU row */
@@ -413,7 +413,7 @@ jinit_c_coef_controller (j_compress_ptr cinfo, boolean need_full_buffer)

  coef = (my_coef_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_coef_controller));
+				SIZEOF(my_coef_controller));
  cinfo->coef = (struct jpeg_c_coef_controller *) coef;
  coef->pub.start_pass = start_pass_coef;

@@ -426,14 +426,14 @@ jinit_c_coef_controller (j_compress_ptr cinfo, boolean need_full_buffer)
    jpeg_component_info *compptr;

    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      coef->whole_image[ci] = (*cinfo->mem->request_virt_barray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE,
-         (JDIMENSION) jround_up((long) compptr->width_in_blocks,
-                                (long) compptr->h_samp_factor),
-         (JDIMENSION) jround_up((long) compptr->height_in_blocks,
-                                (long) compptr->v_samp_factor),
-         (JDIMENSION) compptr->v_samp_factor);
+	((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE,
+	 (JDIMENSION) jround_up((long) compptr->width_in_blocks,
+				(long) compptr->h_samp_factor),
+	 (JDIMENSION) jround_up((long) compptr->height_in_blocks,
+				(long) compptr->v_samp_factor),
+	 (JDIMENSION) compptr->v_samp_factor);
    }
 #else
    ERREXIT(cinfo, JERR_BAD_BUFFER_MODE);
@@ -445,7 +445,7 @@ jinit_c_coef_controller (j_compress_ptr cinfo, boolean need_full_buffer)

    buffer = (JBLOCKROW)
      (*cinfo->mem->alloc_large) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  C_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK));
+				  C_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK));
    for (i = 0; i < C_MAX_BLOCKS_IN_MCU; i++) {
      coef->MCU_buffer[i] = buffer + i;
    }
@@ -2,7 +2,7 @@
 * jccolor.c
 *
 * Copyright (C) 1991-1996, Thomas G. Lane.
- * Modified 2011-2012 by Guido Vollbeding.
+ * Modified 2011-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -29,13 +29,25 @@ typedef my_color_converter * my_cconvert_ptr;
 /**************** RGB -> YCbCr conversion: most common case **************/

 /*
- * YCbCr is defined per CCIR 601-1, except that Cb and Cr are
- * normalized to the range 0..MAXJSAMPLE rather than -0.5 .. 0.5.
- * The conversion equations to be implemented are therefore
- *	Y  =  0.29900 * R + 0.58700 * G + 0.11400 * B
- *	Cb = -0.16874 * R - 0.33126 * G + 0.50000 * B  + CENTERJSAMPLE
- *	Cr =  0.50000 * R - 0.41869 * G - 0.08131 * B  + CENTERJSAMPLE
- * (These numbers are derived from TIFF 6.0 section 21, dated 3-June-92.)
+ * YCbCr is defined per Recommendation ITU-R BT.601-7 (03/2011),
+ * previously known as Recommendation CCIR 601-1, except that Cb and Cr
+ * are normalized to the range 0..MAXJSAMPLE rather than -0.5 .. 0.5.
+ * sRGB (standard RGB color space) is defined per IEC 61966-2-1:1999.
+ * sYCC (standard luma-chroma-chroma color space with extended gamut)
+ * is defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex F.
+ * bg-sRGB and bg-sYCC (big gamut standard color spaces)
+ * are defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex G.
+ * Note that the derived conversion coefficients given in some of these
+ * documents are imprecise.  The general conversion equations are
+ *	Y  = Kr * R + (1 - Kr - Kb) * G + Kb * B
+ *	Cb = 0.5 * (B - Y) / (1 - Kb)
+ *	Cr = 0.5 * (R - Y) / (1 - Kr)
+ * With Kr = 0.299 and Kb = 0.114 (derived according to SMPTE RP 177-1993
+ * from the 1953 FCC NTSC primaries and CIE Illuminant C),
+ * the conversion equations to be implemented are therefore
+ *	Y  =  0.299 * R + 0.587 * G + 0.114 * B
+ *	Cb = -0.168735892 * R - 0.331264108 * G + 0.5 * B + CENTERJSAMPLE
+ *	Cr =  0.5 * R - 0.418687589 * G - 0.081312411 * B + CENTERJSAMPLE
 * Note: older versions of the IJG code used a zero offset of MAXJSAMPLE/2,
 * rather than CENTERJSAMPLE, for Cb and Cr.  This gave equal positive and
 * negative swings for Cb/Cr, but meant that grayscale values (Cb=Cr=0)
@@ -49,9 +61,9 @@ typedef my_color_converter * my_cconvert_ptr;
 * For even more speed, we avoid doing any multiplications in the inner loop
 * by precalculating the constants times R,G,B for all possible values.
 * For 8-bit JSAMPLEs this is very reasonable (only 256 entries per table);
- * for 12-bit samples it is still acceptable.  It's not very reasonable for
- * 16-bit samples, but if you want lossless storage you shouldn't be changing
- * colorspace anyway.
+ * for 9-bit to 12-bit samples it is still acceptable.  It's not very
+ * reasonable for 16-bit samples, but if you want lossless storage you
+ * shouldn't be changing colorspace anyway.
 * The CENTERJSAMPLE offsets and the rounding fudge-factor of 0.5 are included
 * in the tables to save adding them separately in the inner loop.
 */
@@ -93,24 +105,24 @@ rgb_ycc_start (j_compress_ptr cinfo)
  /* Allocate and fill in the conversion tables. */
  cconvert->rgb_ycc_tab = rgb_ycc_tab = (INT32 *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (TABLE_SIZE * SIZEOF(INT32)));
+				(TABLE_SIZE * SIZEOF(INT32)));

  for (i = 0; i <= MAXJSAMPLE; i++) {
-    rgb_ycc_tab[i+R_Y_OFF] = FIX(0.29900) * i;
-    rgb_ycc_tab[i+G_Y_OFF] = FIX(0.58700) * i;
-    rgb_ycc_tab[i+B_Y_OFF] = FIX(0.11400) * i     + ONE_HALF;
-    rgb_ycc_tab[i+R_CB_OFF] = (-FIX(0.16874)) * i;
-    rgb_ycc_tab[i+G_CB_OFF] = (-FIX(0.33126)) * i;
+    rgb_ycc_tab[i+R_Y_OFF] = FIX(0.299) * i;
+    rgb_ycc_tab[i+G_Y_OFF] = FIX(0.587) * i;
+    rgb_ycc_tab[i+B_Y_OFF] = FIX(0.114) * i   + ONE_HALF;
+    rgb_ycc_tab[i+R_CB_OFF] = (-FIX(0.168735892)) * i;
+    rgb_ycc_tab[i+G_CB_OFF] = (-FIX(0.331264108)) * i;
    /* We use a rounding fudge-factor of 0.5-epsilon for Cb and Cr.
     * This ensures that the maximum output will round to MAXJSAMPLE
     * not MAXJSAMPLE+1, and thus that we don't have to range-limit.
     */
-    rgb_ycc_tab[i+B_CB_OFF] = FIX(0.50000) * i    + CBCR_OFFSET + ONE_HALF-1;
+    rgb_ycc_tab[i+B_CB_OFF] = FIX(0.5) * i    + CBCR_OFFSET + ONE_HALF-1;
 /*  B=>Cb and R=>Cr tables are the same
-    rgb_ycc_tab[i+R_CR_OFF] = FIX(0.50000) * i    + CBCR_OFFSET + ONE_HALF-1;
+    rgb_ycc_tab[i+R_CR_OFF] = FIX(0.5) * i    + CBCR_OFFSET + ONE_HALF-1;
 */
-    rgb_ycc_tab[i+G_CR_OFF] = (-FIX(0.41869)) * i;
-    rgb_ycc_tab[i+B_CR_OFF] = (-FIX(0.08131)) * i;
+    rgb_ycc_tab[i+G_CR_OFF] = (-FIX(0.418687589)) * i;
+    rgb_ycc_tab[i+B_CR_OFF] = (-FIX(0.081312411)) * i;
  }
 }

@@ -129,8 +141,8 @@ rgb_ycc_start (j_compress_ptr cinfo)

 METHODDEF(void)
 rgb_ycc_convert (j_compress_ptr cinfo,
-                 JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-                 JDIMENSION output_row, int num_rows)
+		 JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+		 JDIMENSION output_row, int num_rows)
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  register INT32 * ctab = cconvert->rgb_ycc_tab;
@@ -157,16 +169,16 @@ rgb_ycc_convert (j_compress_ptr cinfo,
       */
      /* Y */
      outptr0[col] = (JSAMPLE)
-                ((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
+		 >> SCALEBITS);
      /* Cb */
      outptr1[col] = (JSAMPLE)
-                ((ctab[r+R_CB_OFF] + ctab[g+G_CB_OFF] + ctab[b+B_CB_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_CB_OFF] + ctab[g+G_CB_OFF] + ctab[b+B_CB_OFF])
+		 >> SCALEBITS);
      /* Cr */
      outptr2[col] = (JSAMPLE)
-                ((ctab[r+R_CR_OFF] + ctab[g+G_CR_OFF] + ctab[b+B_CR_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_CR_OFF] + ctab[g+G_CR_OFF] + ctab[b+B_CR_OFF])
+		 >> SCALEBITS);
      inptr += RGB_PIXELSIZE;
    }
  }
@@ -185,8 +197,8 @@ rgb_ycc_convert (j_compress_ptr cinfo,

 METHODDEF(void)
 rgb_gray_convert (j_compress_ptr cinfo,
-                  JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-                  JDIMENSION output_row, int num_rows)
+		  JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+		  JDIMENSION output_row, int num_rows)
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  register INT32 * ctab = cconvert->rgb_ycc_tab;
@@ -205,8 +217,8 @@ rgb_gray_convert (j_compress_ptr cinfo,
      b = GETJSAMPLE(inptr[RGB_BLUE]);
      /* Y */
      outptr[col] = (JSAMPLE)
-                ((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
+		 >> SCALEBITS);
      inptr += RGB_PIXELSIZE;
    }
  }
@@ -223,8 +235,8 @@ rgb_gray_convert (j_compress_ptr cinfo,

 METHODDEF(void)
 cmyk_ycck_convert (j_compress_ptr cinfo,
-                   JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-                   JDIMENSION output_row, int num_rows)
+		   JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+		   JDIMENSION output_row, int num_rows)
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  register INT32 * ctab = cconvert->rgb_ycc_tab;
@@ -254,16 +266,16 @@ cmyk_ycck_convert (j_compress_ptr cinfo,
       */
      /* Y */
      outptr0[col] = (JSAMPLE)
-                ((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
+		 >> SCALEBITS);
      /* Cb */
      outptr1[col] = (JSAMPLE)
-                ((ctab[r+R_CB_OFF] + ctab[g+G_CB_OFF] + ctab[b+B_CB_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_CB_OFF] + ctab[g+G_CB_OFF] + ctab[b+B_CB_OFF])
+		 >> SCALEBITS);
      /* Cr */
      outptr2[col] = (JSAMPLE)
-                ((ctab[r+R_CR_OFF] + ctab[g+G_CR_OFF] + ctab[b+B_CR_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_CR_OFF] + ctab[g+G_CR_OFF] + ctab[b+B_CR_OFF])
+		 >> SCALEBITS);
      inptr += 4;
    }
  }
@@ -274,12 +286,15 @@ cmyk_ycck_convert (j_compress_ptr cinfo,
 * Convert some rows of samples to the JPEG colorspace.
 * [R,G,B] to [R-G,G,B-G] conversion with modulo calculation
 * (forward reversible color transform).
+ * This can be seen as an adaption of the general RGB->YCbCr
+ * conversion equation with Kr = Kb = 0, while replacing the
+ * normalization by modulo calculation.
 */

 METHODDEF(void)
 rgb_rgb1_convert (j_compress_ptr cinfo,
-                  JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-                  JDIMENSION output_row, int num_rows)
+		  JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+		  JDIMENSION output_row, int num_rows)
 {
  register int r, g, b;
  register JSAMPROW inptr;
@@ -312,13 +327,13 @@ rgb_rgb1_convert (j_compress_ptr cinfo,
 /*
 * Convert some rows of samples to the JPEG colorspace.
 * This version handles grayscale output with no conversion.
- * The source can be either plain grayscale or YCbCr (since Y == gray).
+ * The source can be either plain grayscale or YCC (since Y == gray).
 */

 METHODDEF(void)
 grayscale_convert (j_compress_ptr cinfo,
-                   JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-                   JDIMENSION output_row, int num_rows)
+		   JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+		   JDIMENSION output_row, int num_rows)
 {
  int instride = cinfo->input_components;
  register JSAMPROW inptr;
@@ -345,8 +360,8 @@ grayscale_convert (j_compress_ptr cinfo,

 METHODDEF(void)
 rgb_convert (j_compress_ptr cinfo,
-             JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-             JDIMENSION output_row, int num_rows)
+	     JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+	     JDIMENSION output_row, int num_rows)
 {
  register JSAMPROW inptr;
  register JSAMPROW outptr0, outptr1, outptr2;
@@ -378,8 +393,8 @@ rgb_convert (j_compress_ptr cinfo,

 METHODDEF(void)
 null_convert (j_compress_ptr cinfo,
-              JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-              JDIMENSION output_row, int num_rows)
+	      JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+	      JDIMENSION output_row, int num_rows)
 {
  int ci;
  register int nc = cinfo->num_components;
@@ -394,8 +409,8 @@ null_convert (j_compress_ptr cinfo,
      inptr = input_buf[0] + ci;
      outptr = output_buf[ci][output_row];
      for (col = 0; col < num_cols; col++) {
-        *outptr++ = *inptr;	/* don't need GETJSAMPLE() here */
-        inptr += nc;
+	*outptr++ = *inptr;	/* don't need GETJSAMPLE() here */
+	inptr += nc;
      }
    }
    input_buf++;
@@ -426,7 +441,7 @@ jinit_color_converter (j_compress_ptr cinfo)

  cconvert = (my_cconvert_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_color_converter));
+				SIZEOF(my_color_converter));
  cinfo->cconvert = &cconvert->pub;
  /* set start_pass to null method until we find out differently */
  cconvert->pub.start_pass = null_method;
@@ -439,11 +454,13 @@ jinit_color_converter (j_compress_ptr cinfo)
    break;

  case JCS_RGB:
+  case JCS_BG_RGB:
    if (cinfo->input_components != RGB_PIXELSIZE)
      ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE);
    break;

  case JCS_YCbCr:
+  case JCS_BG_YCC:
    if (cinfo->input_components != 3)
      ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE);
    break;
@@ -460,8 +477,10 @@ jinit_color_converter (j_compress_ptr cinfo)
    break;
  }

-  /* Support color transform only for RGB colorspace */
-  if (cinfo->color_transform && cinfo->jpeg_color_space != JCS_RGB)
+  /* Support color transform only for RGB colorspaces */
+  if (cinfo->color_transform &&
+      cinfo->jpeg_color_space != JCS_RGB &&
+      cinfo->jpeg_color_space != JCS_BG_RGB)
    ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);

  /* Check num_components, set conversion method based on requested space */
@@ -469,30 +488,35 @@ jinit_color_converter (j_compress_ptr cinfo)
  case JCS_GRAYSCALE:
    if (cinfo->num_components != 1)
      ERREXIT(cinfo, JERR_BAD_J_COLORSPACE);
-    if (cinfo->in_color_space == JCS_GRAYSCALE ||
-        cinfo->in_color_space == JCS_YCbCr)
+    switch (cinfo->in_color_space) {
+    case JCS_GRAYSCALE:
+    case JCS_YCbCr:
+    case JCS_BG_YCC:
      cconvert->pub.color_convert = grayscale_convert;
-    else if (cinfo->in_color_space == JCS_RGB) {
+      break;
+    case JCS_RGB:
      cconvert->pub.start_pass = rgb_ycc_start;
      cconvert->pub.color_convert = rgb_gray_convert;
-    } else
+      break;
+    default:
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+    }
    break;

  case JCS_RGB:
+  case JCS_BG_RGB:
    if (cinfo->num_components != 3)
      ERREXIT(cinfo, JERR_BAD_J_COLORSPACE);
-    if (cinfo->in_color_space == JCS_RGB) {
+    if (cinfo->in_color_space == cinfo->jpeg_color_space) {
      switch (cinfo->color_transform) {
      case JCT_NONE:
-        cconvert->pub.color_convert = rgb_convert;
-        break;
+	cconvert->pub.color_convert = rgb_convert;
+	break;
      case JCT_SUBTRACT_GREEN:
-        cconvert->pub.color_convert = rgb_rgb1_convert;
-        break;
+	cconvert->pub.color_convert = rgb_rgb1_convert;
+	break;
      default:
-        ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
-        break;
+	ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
      }
    } else
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
@@ -501,13 +525,48 @@ jinit_color_converter (j_compress_ptr cinfo)
  case JCS_YCbCr:
    if (cinfo->num_components != 3)
      ERREXIT(cinfo, JERR_BAD_J_COLORSPACE);
-    if (cinfo->in_color_space == JCS_RGB) {
+    switch (cinfo->in_color_space) {
+    case JCS_RGB:
      cconvert->pub.start_pass = rgb_ycc_start;
      cconvert->pub.color_convert = rgb_ycc_convert;
-    } else if (cinfo->in_color_space == JCS_YCbCr)
+      break;
+    case JCS_YCbCr:
      cconvert->pub.color_convert = null_convert;
-    else
+      break;
+    default:
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+    }
+    break;
+
+  case JCS_BG_YCC:
+    if (cinfo->num_components != 3)
+      ERREXIT(cinfo, JERR_BAD_J_COLORSPACE);
+    switch (cinfo->in_color_space) {
+    case JCS_RGB:
+      /* For conversion from normal RGB input to BG_YCC representation,
+       * the Cb/Cr values are first computed as usual, and then
+       * quantized further after DCT processing by a factor of
+       * 2 in reference to the nominal quantization factor.
+       */
+      /* need quantization scale by factor of 2 after DCT */
+      cinfo->comp_info[1].component_needed = TRUE;
+      cinfo->comp_info[2].component_needed = TRUE;
+      /* compute normal YCC first */
+      cconvert->pub.start_pass = rgb_ycc_start;
+      cconvert->pub.color_convert = rgb_ycc_convert;
+      break;
+    case JCS_YCbCr:
+      /* need quantization scale by factor of 2 after DCT */
+      cinfo->comp_info[1].component_needed = TRUE;
+      cinfo->comp_info[2].component_needed = TRUE;
+      /*FALLTHROUGH*/
+    case JCS_BG_YCC:
+      /* Pass through for BG_YCC input */
+      cconvert->pub.color_convert = null_convert;
+      break;
+    default:
+      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+    }
    break;

  case JCS_CMYK:
@@ -522,18 +581,22 @@ jinit_color_converter (j_compress_ptr cinfo)
  case JCS_YCCK:
    if (cinfo->num_components != 4)
      ERREXIT(cinfo, JERR_BAD_J_COLORSPACE);
-    if (cinfo->in_color_space == JCS_CMYK) {
+    switch (cinfo->in_color_space) {
+    case JCS_CMYK:
      cconvert->pub.start_pass = rgb_ycc_start;
      cconvert->pub.color_convert = cmyk_ycck_convert;
-    } else if (cinfo->in_color_space == JCS_YCCK)
+      break;
+    case JCS_YCCK:
      cconvert->pub.color_convert = null_convert;
-    else
+      break;
+    default:
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+    }
    break;

  default:			/* allow null conversion of JCS_UNKNOWN */
    if (cinfo->jpeg_color_space != cinfo->in_color_space ||
-        cinfo->num_components != cinfo->input_components)
+	cinfo->num_components != cinfo->input_components)
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
    cconvert->pub.color_convert = null_convert;
    break;
@@ -2,6 +2,7 @@
 * jcdctmgr.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
+ * Modified 2003-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -25,22 +26,30 @@ typedef struct {
  /* Pointer to the DCT routine actually in use */
  forward_DCT_method_ptr do_dct[MAX_COMPONENTS];

-  /* The actual post-DCT divisors --- not identical to the quant table
-   * entries, because of scaling (especially for an unnormalized DCT).
-   * Each table is given in normal array order.
-   */
-  DCTELEM * divisors[NUM_QUANT_TBLS];
-
 #ifdef DCT_FLOAT_SUPPORTED
  /* Same as above for the floating-point case. */
  float_DCT_method_ptr do_float_dct[MAX_COMPONENTS];
-  FAST_FLOAT * float_divisors[NUM_QUANT_TBLS];
 #endif
 } my_fdct_controller;

 typedef my_fdct_controller * my_fdct_ptr;


+/* The allocated post-DCT divisor tables -- big enough for any
+ * supported variant and not identical to the quant table entries,
+ * because of scaling (especially for an unnormalized DCT) --
+ * are pointed to by dct_table in the per-component comp_info
+ * structures.  Each table is given in normal array order.
+ */
+
+typedef union {
+  DCTELEM int_array[DCTSIZE2];
+#ifdef DCT_FLOAT_SUPPORTED
+  FAST_FLOAT float_array[DCTSIZE2];
+#endif
+} divisor_table;
+
+
 /* The current scaled-DCT routines require ISLOW-style divisor tables,
 * so be sure to compile that code if either ISLOW or SCALING is requested.
 */
@@ -63,15 +72,15 @@ typedef my_fdct_controller * my_fdct_ptr;

 METHODDEF(void)
 forward_DCT (j_compress_ptr cinfo, jpeg_component_info * compptr,
-             JSAMPARRAY sample_data, JBLOCKROW coef_blocks,
-             JDIMENSION start_row, JDIMENSION start_col,
-             JDIMENSION num_blocks)
+	     JSAMPARRAY sample_data, JBLOCKROW coef_blocks,
+	     JDIMENSION start_row, JDIMENSION start_col,
+	     JDIMENSION num_blocks)
 /* This version is used for integer DCT implementations. */
 {
  /* This routine is heavily used, so it's worth coding it tightly. */
  my_fdct_ptr fdct = (my_fdct_ptr) cinfo->fdct;
  forward_DCT_method_ptr do_dct = fdct->do_dct[compptr->component_index];
-  DCTELEM * divisors = fdct->divisors[compptr->quant_tbl_no];
+  DCTELEM * divisors = (DCTELEM *) compptr->dct_table;
  DCTELEM workspace[DCTSIZE2];	/* work area for FDCT subroutine */
  JDIMENSION bi;

@@ -87,35 +96,35 @@ forward_DCT (j_compress_ptr cinfo, jpeg_component_info * compptr,
      register JCOEFPTR output_ptr = coef_blocks[bi];

      for (i = 0; i < DCTSIZE2; i++) {
-        qval = divisors[i];
-        temp = workspace[i];
-        /* Divide the coefficient value by qval, ensuring proper rounding.
-         * Since C does not specify the direction of rounding for negative
-         * quotients, we have to force the dividend positive for portability.
-         *
-         * In most files, at least half of the output values will be zero
-         * (at default quantization settings, more like three-quarters...)
-         * so we should ensure that this case is fast.  On many machines,
-         * a comparison is enough cheaper than a divide to make a special test
-         * a win.  Since both inputs will be nonnegative, we need only test
-         * for a < b to discover whether a/b is 0.
-         * If your machine's division is fast enough, define FAST_DIVIDE.
-         */
+	qval = divisors[i];
+	temp = workspace[i];
+	/* Divide the coefficient value by qval, ensuring proper rounding.
+	 * Since C does not specify the direction of rounding for negative
+	 * quotients, we have to force the dividend positive for portability.
+	 *
+	 * In most files, at least half of the output values will be zero
+	 * (at default quantization settings, more like three-quarters...)
+	 * so we should ensure that this case is fast.  On many machines,
+	 * a comparison is enough cheaper than a divide to make a special test
+	 * a win.  Since both inputs will be nonnegative, we need only test
+	 * for a < b to discover whether a/b is 0.
+	 * If your machine's division is fast enough, define FAST_DIVIDE.
+	 */
 #ifdef FAST_DIVIDE
 #define DIVIDE_BY(a,b)	a /= b
 #else
 #define DIVIDE_BY(a,b)	if (a >= b) a /= b; else a = 0
 #endif
-        if (temp < 0) {
-          temp = -temp;
-          temp += qval>>1;	/* for rounding */
-          DIVIDE_BY(temp, qval);
-          temp = -temp;
-        } else {
-          temp += qval>>1;	/* for rounding */
-          DIVIDE_BY(temp, qval);
-        }
-        output_ptr[i] = (JCOEF) temp;
+	if (temp < 0) {
+	  temp = -temp;
+	  temp += qval>>1;	/* for rounding */
+	  DIVIDE_BY(temp, qval);
+	  temp = -temp;
+	} else {
+	  temp += qval>>1;	/* for rounding */
+	  DIVIDE_BY(temp, qval);
+	}
+	output_ptr[i] = (JCOEF) temp;
      }
    }
  }
@@ -126,15 +135,15 @@ forward_DCT (j_compress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 forward_DCT_float (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                   JSAMPARRAY sample_data, JBLOCKROW coef_blocks,
-                   JDIMENSION start_row, JDIMENSION start_col,
-                   JDIMENSION num_blocks)
+		   JSAMPARRAY sample_data, JBLOCKROW coef_blocks,
+		   JDIMENSION start_row, JDIMENSION start_col,
+		   JDIMENSION num_blocks)
 /* This version is used for floating-point DCT implementations. */
 {
  /* This routine is heavily used, so it's worth coding it tightly. */
  my_fdct_ptr fdct = (my_fdct_ptr) cinfo->fdct;
  float_DCT_method_ptr do_dct = fdct->do_float_dct[compptr->component_index];
-  FAST_FLOAT * divisors = fdct->float_divisors[compptr->quant_tbl_no];
+  FAST_FLOAT * divisors = (FAST_FLOAT *) compptr->dct_table;
  FAST_FLOAT workspace[DCTSIZE2]; /* work area for FDCT subroutine */
  JDIMENSION bi;

@@ -150,15 +159,15 @@ forward_DCT_float (j_compress_ptr cinfo, jpeg_component_info * compptr,
      register JCOEFPTR output_ptr = coef_blocks[bi];

      for (i = 0; i < DCTSIZE2; i++) {
-        /* Apply the quantization and scaling factor */
-        temp = workspace[i] * divisors[i];
-        /* Round to nearest integer.
-         * Since C does not specify the direction of rounding for negative
-         * quotients, we have to force the dividend positive for portability.
-         * The maximum coefficient size is +-16K (for 12-bit data), so this
-         * code should work for either 16-bit or 32-bit ints.
-         */
-        output_ptr[i] = (JCOEF) ((int) (temp + (FAST_FLOAT) 16384.5) - 16384);
+	/* Apply the quantization and scaling factor */
+	temp = workspace[i] * divisors[i];
+	/* Round to nearest integer.
+	 * Since C does not specify the direction of rounding for negative
+	 * quotients, we have to force the dividend positive for portability.
+	 * The maximum coefficient size is +-16K (for 12-bit data), so this
+	 * code should work for either 16-bit or 32-bit ints.
+	 */
+	output_ptr[i] = (JCOEF) ((int) (temp + (FAST_FLOAT) 16384.5) - 16384);
      }
    }
  }
@@ -320,54 +329,49 @@ start_pass_fdctmgr (j_compress_ptr cinfo)
      switch (cinfo->dct_method) {
 #ifdef DCT_ISLOW_SUPPORTED
      case JDCT_ISLOW:
-        fdct->do_dct[ci] = jpeg_fdct_islow;
-        method = JDCT_ISLOW;
-        break;
+	fdct->do_dct[ci] = jpeg_fdct_islow;
+	method = JDCT_ISLOW;
+	break;
 #endif
 #ifdef DCT_IFAST_SUPPORTED
      case JDCT_IFAST:
-        fdct->do_dct[ci] = jpeg_fdct_ifast;
-        method = JDCT_IFAST;
-        break;
+	fdct->do_dct[ci] = jpeg_fdct_ifast;
+	method = JDCT_IFAST;
+	break;
 #endif
 #ifdef DCT_FLOAT_SUPPORTED
      case JDCT_FLOAT:
-        fdct->do_float_dct[ci] = jpeg_fdct_float;
-        method = JDCT_FLOAT;
-        break;
+	fdct->do_float_dct[ci] = jpeg_fdct_float;
+	method = JDCT_FLOAT;
+	break;
 #endif
      default:
-        ERREXIT(cinfo, JERR_NOT_COMPILED);
-        break;
+	ERREXIT(cinfo, JERR_NOT_COMPILED);
+	break;
      }
      break;
    default:
      ERREXIT2(cinfo, JERR_BAD_DCTSIZE,
-               compptr->DCT_h_scaled_size, compptr->DCT_v_scaled_size);
+	       compptr->DCT_h_scaled_size, compptr->DCT_v_scaled_size);
      break;
    }
    qtblno = compptr->quant_tbl_no;
    /* Make sure specified quantization table is present */
    if (qtblno < 0 || qtblno >= NUM_QUANT_TBLS ||
-        cinfo->quant_tbl_ptrs[qtblno] == NULL)
+	cinfo->quant_tbl_ptrs[qtblno] == NULL)
      ERREXIT1(cinfo, JERR_NO_QUANT_TABLE, qtblno);
    qtbl = cinfo->quant_tbl_ptrs[qtblno];
-    /* Compute divisors for this quant table */
-    /* We may do this more than once for same table, but it's not a big deal */
+    /* Create divisor table from quant table */
    switch (method) {
 #ifdef PROVIDE_ISLOW_TABLES
    case JDCT_ISLOW:
      /* For LL&M IDCT method, divisors are equal to raw quantization
       * coefficients multiplied by 8 (to counteract scaling).
       */
-      if (fdct->divisors[qtblno] == NULL) {
-        fdct->divisors[qtblno] = (DCTELEM *)
-          (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                      DCTSIZE2 * SIZEOF(DCTELEM));
-      }
-      dtbl = fdct->divisors[qtblno];
+      dtbl = (DCTELEM *) compptr->dct_table;
      for (i = 0; i < DCTSIZE2; i++) {
-        dtbl[i] = ((DCTELEM) qtbl->quantval[i]) << 3;
+	dtbl[i] =
+	  ((DCTELEM) qtbl->quantval[i]) << (compptr->component_needed ? 4 : 3);
      }
      fdct->pub.forward_DCT[ci] = forward_DCT;
      break;
@@ -375,38 +379,33 @@ start_pass_fdctmgr (j_compress_ptr cinfo)
 #ifdef DCT_IFAST_SUPPORTED
    case JDCT_IFAST:
      {
-        /* For AA&N IDCT method, divisors are equal to quantization
-         * coefficients scaled by scalefactor[row]*scalefactor[col], where
-         *   scalefactor[0] = 1
-         *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
-         * We apply a further scale factor of 8.
-         */
+	/* For AA&N IDCT method, divisors are equal to quantization
+	 * coefficients scaled by scalefactor[row]*scalefactor[col], where
+	 *   scalefactor[0] = 1
+	 *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
+	 * We apply a further scale factor of 8.
+	 */
 #define CONST_BITS 14
-        static const INT16 aanscales[DCTSIZE2] = {
-          /* precomputed values scaled up by 14 bits */
-          16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
-          22725, 31521, 29692, 26722, 22725, 17855, 12299,  6270,
-          21407, 29692, 27969, 25172, 21407, 16819, 11585,  5906,
-          19266, 26722, 25172, 22654, 19266, 15137, 10426,  5315,
-          16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
-          12873, 17855, 16819, 15137, 12873, 10114,  6967,  3552,
-           8867, 12299, 11585, 10426,  8867,  6967,  4799,  2446,
-           4520,  6270,  5906,  5315,  4520,  3552,  2446,  1247
-        };
-        SHIFT_TEMPS
+	static const INT16 aanscales[DCTSIZE2] = {
+	  /* precomputed values scaled up by 14 bits */
+	  16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
+	  22725, 31521, 29692, 26722, 22725, 17855, 12299,  6270,
+	  21407, 29692, 27969, 25172, 21407, 16819, 11585,  5906,
+	  19266, 26722, 25172, 22654, 19266, 15137, 10426,  5315,
+	  16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
+	  12873, 17855, 16819, 15137, 12873, 10114,  6967,  3552,
+	   8867, 12299, 11585, 10426,  8867,  6967,  4799,  2446,
+	   4520,  6270,  5906,  5315,  4520,  3552,  2446,  1247
+	};
+	SHIFT_TEMPS

-        if (fdct->divisors[qtblno] == NULL) {
-          fdct->divisors[qtblno] = (DCTELEM *)
-            (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                        DCTSIZE2 * SIZEOF(DCTELEM));
-        }
-        dtbl = fdct->divisors[qtblno];
-        for (i = 0; i < DCTSIZE2; i++) {
-          dtbl[i] = (DCTELEM)
-            DESCALE(MULTIPLY16V16((INT32) qtbl->quantval[i],
-                                  (INT32) aanscales[i]),
-                    CONST_BITS-3);
-        }
+	dtbl = (DCTELEM *) compptr->dct_table;
+	for (i = 0; i < DCTSIZE2; i++) {
+	  dtbl[i] = (DCTELEM)
+	    DESCALE(MULTIPLY16V16((INT32) qtbl->quantval[i],
+				  (INT32) aanscales[i]),
+		    compptr->component_needed ? CONST_BITS-4 : CONST_BITS-3);
+	}
      }
      fdct->pub.forward_DCT[ci] = forward_DCT;
      break;
@@ -414,36 +413,31 @@ start_pass_fdctmgr (j_compress_ptr cinfo)
 #ifdef DCT_FLOAT_SUPPORTED
    case JDCT_FLOAT:
      {
-        /* For float AA&N IDCT method, divisors are equal to quantization
-         * coefficients scaled by scalefactor[row]*scalefactor[col], where
-         *   scalefactor[0] = 1
-         *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
-         * We apply a further scale factor of 8.
-         * What's actually stored is 1/divisor so that the inner loop can
-         * use a multiplication rather than a division.
-         */
-        FAST_FLOAT * fdtbl;
-        int row, col;
-        static const double aanscalefactor[DCTSIZE] = {
-          1.0, 1.387039845, 1.306562965, 1.175875602,
-          1.0, 0.785694958, 0.541196100, 0.275899379
-        };
+	/* For float AA&N IDCT method, divisors are equal to quantization
+	 * coefficients scaled by scalefactor[row]*scalefactor[col], where
+	 *   scalefactor[0] = 1
+	 *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
+	 * We apply a further scale factor of 8.
+	 * What's actually stored is 1/divisor so that the inner loop can
+	 * use a multiplication rather than a division.
+	 */
+	FAST_FLOAT * fdtbl = (FAST_FLOAT *) compptr->dct_table;
+	int row, col;
+	static const double aanscalefactor[DCTSIZE] = {
+	  1.0, 1.387039845, 1.306562965, 1.175875602,
+	  1.0, 0.785694958, 0.541196100, 0.275899379
+	};

-        if (fdct->float_divisors[qtblno] == NULL) {
-          fdct->float_divisors[qtblno] = (FAST_FLOAT *)
-            (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                        DCTSIZE2 * SIZEOF(FAST_FLOAT));
-        }
-        fdtbl = fdct->float_divisors[qtblno];
-        i = 0;
-        for (row = 0; row < DCTSIZE; row++) {
-          for (col = 0; col < DCTSIZE; col++) {
-            fdtbl[i] = (FAST_FLOAT)
-              (1.0 / (((double) qtbl->quantval[i] *
-                       aanscalefactor[row] * aanscalefactor[col] * 8.0)));
-            i++;
-          }
-        }
+	i = 0;
+	for (row = 0; row < DCTSIZE; row++) {
+	  for (col = 0; col < DCTSIZE; col++) {
+	    fdtbl[i] = (FAST_FLOAT)
+	      (1.0 / ((double) qtbl->quantval[i] *
+		      aanscalefactor[row] * aanscalefactor[col] *
+		      (compptr->component_needed ? 16.0 : 8.0)));
+	    i++;
+	  }
+	}
      }
      fdct->pub.forward_DCT[ci] = forward_DCT_float;
      break;
@@ -464,19 +458,20 @@ GLOBAL(void)
 jinit_forward_dct (j_compress_ptr cinfo)
 {
  my_fdct_ptr fdct;
-  int i;
+  int ci;
+  jpeg_component_info *compptr;

  fdct = (my_fdct_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_fdct_controller));
-  cinfo->fdct = (struct jpeg_forward_dct *) fdct;
+				SIZEOF(my_fdct_controller));
+  cinfo->fdct = &fdct->pub;
  fdct->pub.start_pass = start_pass_fdctmgr;

-  /* Mark divisor tables unallocated */
-  for (i = 0; i < NUM_QUANT_TBLS; i++) {
-    fdct->divisors[i] = NULL;
-#ifdef DCT_FLOAT_SUPPORTED
-    fdct->float_divisors[i] = NULL;
-#endif
+  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
+       ci++, compptr++) {
+    /* Allocate a divisor table for each component */
+    compptr->dct_table =
+      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				  SIZEOF(divisor_table));
  }
 }
@@ -2,7 +2,7 @@
 * jchuff.c
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
- * Modified 2006-2009 by Guido Vollbeding.
+ * Modified 2006-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -68,12 +68,12 @@ typedef struct {
 #else
 #if MAX_COMPS_IN_SCAN == 4
 #define ASSIGN_STATE(dest,src)  \
-        ((dest).put_buffer = (src).put_buffer, \
-         (dest).put_bits = (src).put_bits, \
-         (dest).last_dc_val[0] = (src).last_dc_val[0], \
-         (dest).last_dc_val[1] = (src).last_dc_val[1], \
-         (dest).last_dc_val[2] = (src).last_dc_val[2], \
-         (dest).last_dc_val[3] = (src).last_dc_val[3])
+	((dest).put_buffer = (src).put_buffer, \
+	 (dest).put_bits = (src).put_bits, \
+	 (dest).last_dc_val[0] = (src).last_dc_val[0], \
+	 (dest).last_dc_val[1] = (src).last_dc_val[1], \
+	 (dest).last_dc_val[2] = (src).last_dc_val[2], \
+	 (dest).last_dc_val[3] = (src).last_dc_val[3])
 #endif
 #endif

@@ -143,9 +143,9 @@ typedef struct {
 #ifdef RIGHT_SHIFT_IS_UNSIGNED
 #define ISHIFT_TEMPS	int ishift_temp;
 #define IRIGHT_SHIFT(x,shft)  \
-        ((ishift_temp = (x)) < 0 ? \
-         (ishift_temp >> (shft)) | ((~0) << (16-(shft))) : \
-         (ishift_temp >> (shft)))
+	((ishift_temp = (x)) < 0 ? \
+	 (ishift_temp >> (shft)) | ((~0) << (16-(shft))) : \
+	 (ishift_temp >> (shft)))
 #else
 #define ISHIFT_TEMPS
 #define IRIGHT_SHIFT(x,shft)	((x) >> (shft))
@@ -159,7 +159,7 @@ typedef struct {

 LOCAL(void)
 jpeg_make_c_derived_tbl (j_compress_ptr cinfo, boolean isDC, int tblno,
-                         c_derived_tbl ** pdtbl)
+			 c_derived_tbl ** pdtbl)
 {
  JHUFF_TBL *htbl;
  c_derived_tbl *dtbl;
@@ -184,9 +184,9 @@ jpeg_make_c_derived_tbl (j_compress_ptr cinfo, boolean isDC, int tblno,
  if (*pdtbl == NULL)
    *pdtbl = (c_derived_tbl *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  SIZEOF(c_derived_tbl));
+				  SIZEOF(c_derived_tbl));
  dtbl = *pdtbl;
-
+  
  /* Figure C.1: make table of Huffman code length for each symbol */

  p = 0;
@@ -199,7 +199,7 @@ jpeg_make_c_derived_tbl (j_compress_ptr cinfo, boolean isDC, int tblno,
  }
  huffsize[p] = 0;
  lastp = p;
-
+  
  /* Figure C.2: generate the codes themselves */
  /* We also validate that the counts represent a legal Huffman code tree. */

@@ -219,7 +219,7 @@ jpeg_make_c_derived_tbl (j_compress_ptr cinfo, boolean isDC, int tblno,
    code <<= 1;
    si++;
  }
-
+  
  /* Figure C.3: generate encoding tables */
  /* These are code and size indexed by symbol value */

@@ -253,16 +253,16 @@ jpeg_make_c_derived_tbl (j_compress_ptr cinfo, boolean isDC, int tblno,

 /* Emit a byte, taking 'action' if must suspend. */
 #define emit_byte_s(state,val,action)  \
-        { *(state)->next_output_byte++ = (JOCTET) (val);  \
-          if (--(state)->free_in_buffer == 0)  \
-            if (! dump_buffer_s(state))  \
-              { action; } }
+	{ *(state)->next_output_byte++ = (JOCTET) (val);  \
+	  if (--(state)->free_in_buffer == 0)  \
+	    if (! dump_buffer_s(state))  \
+	      { action; } }

 /* Emit a byte */
 #define emit_byte_e(entropy,val)  \
-        { *(entropy)->next_output_byte++ = (JOCTET) (val);  \
-          if (--(entropy)->free_in_buffer == 0)  \
-            dump_buffer_e(entropy); }
+	{ *(entropy)->next_output_byte++ = (JOCTET) (val);  \
+	  if (--(entropy)->free_in_buffer == 0)  \
+	    dump_buffer_e(entropy); }


 LOCAL(boolean)
@@ -308,20 +308,23 @@ emit_bits_s (working_state * state, unsigned int code, int size)
 /* Emit some bits; return TRUE if successful, FALSE if must suspend */
 {
  /* This routine is heavily used, so it's worth coding tightly. */
-  register INT32 put_buffer = (INT32) code;
-  register int put_bits = state->cur.put_bits;
+  register INT32 put_buffer;
+  register int put_bits;

  /* if size is 0, caller used an invalid Huffman table entry */
  if (size == 0)
    ERREXIT(state->cinfo, JERR_HUFF_MISSING_CODE);

-  put_buffer &= (((INT32) 1)<<size) - 1; /* mask off any extra bits in code */
+  /* mask off any extra bits in code */
+  put_buffer = ((INT32) code) & ((((INT32) 1) << size) - 1);

-  put_bits += size;		/* new number of bits in buffer */
+  /* new number of bits in buffer */
+  put_bits = size + state->cur.put_bits;

  put_buffer <<= 24 - put_bits; /* align incoming bits */

-  put_buffer |= state->cur.put_buffer; /* and merge with old buffer contents */
+  /* and merge with old buffer contents */
+  put_buffer |= state->cur.put_buffer;

  while (put_bits >= 8) {
    int c = (int) ((put_buffer >> 16) & 0xFF);
@@ -347,8 +350,8 @@ emit_bits_e (huff_entropy_ptr entropy, unsigned int code, int size)
 /* Emit some bits, unless we are in gather mode */
 {
  /* This routine is heavily used, so it's worth coding tightly. */
-  register INT32 put_buffer = (INT32) code;
-  register int put_bits = entropy->saved.put_bits;
+  register INT32 put_buffer;
+  register int put_bits;

  /* if size is 0, caller used an invalid Huffman table entry */
  if (size == 0)
@@ -357,9 +360,11 @@ emit_bits_e (huff_entropy_ptr entropy, unsigned int code, int size)
  if (entropy->gather_statistics)
    return;			/* do nothing if we're only getting stats */

-  put_buffer &= (((INT32) 1)<<size) - 1; /* mask off any extra bits in code */
+  /* mask off any extra bits in code */
+  put_buffer = ((INT32) code) & ((((INT32) 1) << size) - 1);

-  put_bits += size;		/* new number of bits in buffer */
+  /* new number of bits in buffer */
+  put_bits = size + entropy->saved.put_bits;

  put_buffer <<= 24 - put_bits; /* align incoming bits */

@@ -438,7 +443,7 @@ emit_ac_symbol (huff_entropy_ptr entropy, int tbl_no, int symbol)

 LOCAL(void)
 emit_buffered_bits (huff_entropy_ptr entropy, char * bufstart,
-                    unsigned int nbits)
+		    unsigned int nbits)
 {
  if (entropy->gather_statistics)
    return;			/* no real work */
@@ -543,10 +548,7 @@ encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
  huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy;
  register int temp, temp2;
  register int nbits;
-  int blkn, ci;
-  int Al = cinfo->Al;
-  JBLOCKROW block;
-  jpeg_component_info * compptr;
+  int blkn, ci, tbl;
  ISHIFT_TEMPS

  entropy->next_output_byte = cinfo->dest->next_output_byte;
@@ -559,21 +561,20 @@ encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)

  /* Encode the MCU data blocks */
  for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) {
-    block = MCU_data[blkn];
    ci = cinfo->MCU_membership[blkn];
-    compptr = cinfo->cur_comp_info[ci];
+    tbl = cinfo->cur_comp_info[ci]->dc_tbl_no;

    /* Compute the DC value after the required point transform by Al.
     * This is simply an arithmetic right shift.
     */
-    temp2 = IRIGHT_SHIFT((int) ((*block)[0]), Al);
+    temp = IRIGHT_SHIFT((int) (MCU_data[blkn][0][0]), cinfo->Al);

    /* DC differences are figured on the point-transformed values. */
-    temp = temp2 - entropy->saved.last_dc_val[ci];
-    entropy->saved.last_dc_val[ci] = temp2;
+    temp2 = temp - entropy->saved.last_dc_val[ci];
+    entropy->saved.last_dc_val[ci] = temp;

    /* Encode the DC coefficient difference per section G.1.2.1 */
-    temp2 = temp;
+    temp = temp2;
    if (temp < 0) {
      temp = -temp;		/* temp is abs value of input */
      /* For a negative input, want temp2 = bitwise complement of abs(input) */
@@ -594,7 +595,7 @@ encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
      ERREXIT(cinfo, JERR_BAD_DCT_COEF);

    /* Count/emit the Huffman-coded symbol for the number of bits */
-    emit_dc_symbol(entropy, compptr->dc_tbl_no, nbits);
+    emit_dc_symbol(entropy, tbl, nbits);

    /* Emit that number of bits of the value, if positive, */
    /* or the complement of its magnitude, if negative. */
@@ -628,12 +629,12 @@ METHODDEF(boolean)
 encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
 {
  huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy;
+  const int * natural_order;
+  JBLOCKROW block;
  register int temp, temp2;
  register int nbits;
  register int r, k;
  int Se, Al;
-  const int * natural_order;
-  JBLOCKROW block;

  entropy->next_output_byte = cinfo->dest->next_output_byte;
  entropy->free_in_buffer = cinfo->dest->free_in_buffer;
@@ -651,9 +652,9 @@ encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
  block = MCU_data[0];

  /* Encode the AC coefficients per section G.1.2.2, fig. G.3 */
-
+  
  r = 0;			/* r = run length of zeros */
-
+   
  for (k = cinfo->Ss; k <= Se; k++) {
    if ((temp = (*block)[natural_order[k]]) == 0) {
      r++;
@@ -731,18 +732,15 @@ encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKROW *MCU_data)

 /*
 * MCU encoding for DC successive approximation refinement scan.
- * Note: we assume such scans can be multi-component, although the spec
- * is not very clear on the point.
+ * Note: we assume such scans can be multi-component,
+ * although the spec is not very clear on the point.
 */

 METHODDEF(boolean)
 encode_mcu_DC_refine (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
 {
  huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy;
-  register int temp;
-  int blkn;
-  int Al = cinfo->Al;
-  JBLOCKROW block;
+  int Al, blkn;

  entropy->next_output_byte = cinfo->dest->next_output_byte;
  entropy->free_in_buffer = cinfo->dest->free_in_buffer;
@@ -752,13 +750,12 @@ encode_mcu_DC_refine (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
    if (entropy->restarts_to_go == 0)
      emit_restart_e(entropy, entropy->next_restart_num);

+  Al = cinfo->Al;
+
  /* Encode the MCU data blocks */
  for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) {
-    block = MCU_data[blkn];
-
    /* We simply emit the Al'th bit of the DC coefficient value. */
-    temp = (*block)[0];
-    emit_bits_e(entropy, (unsigned int) (temp >> Al), 1);
+    emit_bits_e(entropy, (unsigned int) (MCU_data[blkn][0][0] >> Al), 1);
  }

  cinfo->dest->next_output_byte = entropy->next_output_byte;
@@ -786,14 +783,14 @@ METHODDEF(boolean)
 encode_mcu_AC_refine (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
 {
  huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy;
+  const int * natural_order;
+  JBLOCKROW block;
  register int temp;
  register int r, k;
+  int Se, Al;
  int EOB;
  char *BR_buffer;
  unsigned int BR;
-  int Se, Al;
-  const int * natural_order;
-  JBLOCKROW block;
  int absvalues[DCTSIZE2];

  entropy->next_output_byte = cinfo->dest->next_output_byte;
@@ -830,7 +827,7 @@ encode_mcu_AC_refine (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
  }

  /* Encode the AC coefficients per section G.1.2.3, fig. G.7 */
-
+  
  r = 0;			/* r = run length of zeros */
  BR = 0;			/* BR = count of buffered bits added now */
  BR_buffer = entropy->bit_buffer + entropy->BE; /* Append bits to buffer */
@@ -914,11 +911,11 @@ encode_mcu_AC_refine (j_compress_ptr cinfo, JBLOCKROW *MCU_data)

 LOCAL(boolean)
 encode_one_block (working_state * state, JCOEFPTR block, int last_dc_val,
-                  c_derived_tbl *dctbl, c_derived_tbl *actbl)
+		  c_derived_tbl *dctbl, c_derived_tbl *actbl)
 {
  register int temp, temp2;
  register int nbits;
-  register int k, r, i;
+  register int r, k;
  int Se = state->cinfo->lim_Se;
  const int * natural_order = state->cinfo->natural_order;

@@ -960,40 +957,40 @@ encode_one_block (working_state * state, JCOEFPTR block, int last_dc_val,
  r = 0;			/* r = run length of zeros */

  for (k = 1; k <= Se; k++) {
-    if ((temp = block[natural_order[k]]) == 0) {
+    if ((temp2 = block[natural_order[k]]) == 0) {
      r++;
    } else {
      /* if run length > 15, must emit special run-length-16 codes (0xF0) */
      while (r > 15) {
-        if (! emit_bits_s(state, actbl->ehufco[0xF0], actbl->ehufsi[0xF0]))
-          return FALSE;
-        r -= 16;
+	if (! emit_bits_s(state, actbl->ehufco[0xF0], actbl->ehufsi[0xF0]))
+	  return FALSE;
+	r -= 16;
      }

-      temp2 = temp;
+      temp = temp2;
      if (temp < 0) {
-        temp = -temp;		/* temp is abs value of input */
-        /* This code assumes we are on a two's complement machine */
-        temp2--;
+	temp = -temp;		/* temp is abs value of input */
+	/* This code assumes we are on a two's complement machine */
+	temp2--;
      }

      /* Find the number of bits needed for the magnitude of the coefficient */
      nbits = 1;		/* there must be at least one 1 bit */
      while ((temp >>= 1))
-        nbits++;
+	nbits++;
      /* Check for out-of-range coefficient values */
      if (nbits > MAX_COEF_BITS)
-        ERREXIT(state->cinfo, JERR_BAD_DCT_COEF);
+	ERREXIT(state->cinfo, JERR_BAD_DCT_COEF);

      /* Emit Huffman symbol for run length / number of bits */
-      i = (r << 4) + nbits;
-      if (! emit_bits_s(state, actbl->ehufco[i], actbl->ehufsi[i]))
-        return FALSE;
+      temp = (r << 4) + nbits;
+      if (! emit_bits_s(state, actbl->ehufco[temp], actbl->ehufsi[temp]))
+	return FALSE;

      /* Emit that number of bits of the value, if positive, */
      /* or the complement of its magnitude, if negative. */
      if (! emit_bits_s(state, (unsigned int) temp2, nbits))
-        return FALSE;
+	return FALSE;

      r = 0;
    }
@@ -1030,7 +1027,7 @@ encode_mcu_huff (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
  if (cinfo->restart_interval) {
    if (entropy->restarts_to_go == 0)
      if (! emit_restart_s(&state, entropy->next_restart_num))
-        return FALSE;
+	return FALSE;
  }

  /* Encode the MCU data blocks */
@@ -1038,9 +1035,9 @@ encode_mcu_huff (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
    ci = cinfo->MCU_membership[blkn];
    compptr = cinfo->cur_comp_info[ci];
    if (! encode_one_block(&state,
-                           MCU_data[blkn][0], state.cur.last_dc_val[ci],
-                           entropy->dc_derived_tbls[compptr->dc_tbl_no],
-                           entropy->ac_derived_tbls[compptr->ac_tbl_no]))
+			   MCU_data[blkn][0], state.cur.last_dc_val[ci],
+			   entropy->dc_derived_tbls[compptr->dc_tbl_no],
+			   entropy->ac_derived_tbls[compptr->ac_tbl_no]))
      return FALSE;
    /* Update last_dc_val */
    state.cur.last_dc_val[ci] = MCU_data[blkn][0][0];
@@ -1120,11 +1117,11 @@ finish_pass_huff (j_compress_ptr cinfo)

 LOCAL(void)
 htest_one_block (j_compress_ptr cinfo, JCOEFPTR block, int last_dc_val,
-                 long dc_counts[], long ac_counts[])
+		 long dc_counts[], long ac_counts[])
 {
  register int temp;
  register int nbits;
-  register int k, r;
+  register int r, k;
  int Se = cinfo->lim_Se;
  const int * natural_order = cinfo->natural_order;

@@ -1159,21 +1156,21 @@ htest_one_block (j_compress_ptr cinfo, JCOEFPTR block, int last_dc_val,
    } else {
      /* if run length > 15, must emit special run-length-16 codes (0xF0) */
      while (r > 15) {
-        ac_counts[0xF0]++;
-        r -= 16;
+	ac_counts[0xF0]++;
+	r -= 16;
      }

      /* Find the number of bits needed for the magnitude of the coefficient */
      if (temp < 0)
-        temp = -temp;
+	temp = -temp;

      /* Find the number of bits needed for the magnitude of the coefficient */
      nbits = 1;		/* there must be at least one 1 bit */
      while ((temp >>= 1))
-        nbits++;
+	nbits++;
      /* Check for out-of-range coefficient values */
      if (nbits > MAX_COEF_BITS)
-        ERREXIT(cinfo, JERR_BAD_DCT_COEF);
+	ERREXIT(cinfo, JERR_BAD_DCT_COEF);

      /* Count Huffman symbol for run length / number of bits */
      ac_counts[(r << 4) + nbits]++;
@@ -1205,7 +1202,7 @@ encode_mcu_gather (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
    if (entropy->restarts_to_go == 0) {
      /* Re-initialize DC predictions to 0 */
      for (ci = 0; ci < cinfo->comps_in_scan; ci++)
-        entropy->saved.last_dc_val[ci] = 0;
+	entropy->saved.last_dc_val[ci] = 0;
      /* Update restart state */
      entropy->restarts_to_go = cinfo->restart_interval;
    }
@@ -1216,8 +1213,8 @@ encode_mcu_gather (j_compress_ptr cinfo, JBLOCKROW *MCU_data)
    ci = cinfo->MCU_membership[blkn];
    compptr = cinfo->cur_comp_info[ci];
    htest_one_block(cinfo, MCU_data[blkn][0], entropy->saved.last_dc_val[ci],
-                    entropy->dc_count_ptrs[compptr->dc_tbl_no],
-                    entropy->ac_count_ptrs[compptr->ac_tbl_no]);
+		    entropy->dc_count_ptrs[compptr->dc_tbl_no],
+		    entropy->ac_count_ptrs[compptr->ac_tbl_no]);
    entropy->saved.last_dc_val[ci] = MCU_data[blkn][0][0];
  }

@@ -1269,7 +1266,7 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
  MEMZERO(codesize, SIZEOF(codesize));
  for (i = 0; i < 257; i++)
    others[i] = -1;		/* init links to empty */
-
+  
  freq[256] = 1;		/* make sure 256 has a nonzero count */
  /* Including the pseudo-symbol 256 in the Huffman procedure guarantees
   * that no real symbol is given code-value of all ones, because 256
@@ -1285,8 +1282,8 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
    v = 1000000000L;
    for (i = 0; i <= 256; i++) {
      if (freq[i] && freq[i] <= v) {
-        v = freq[i];
-        c1 = i;
+	v = freq[i];
+	c1 = i;
      }
    }

@@ -1296,15 +1293,15 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
    v = 1000000000L;
    for (i = 0; i <= 256; i++) {
      if (freq[i] && freq[i] <= v && i != c1) {
-        v = freq[i];
-        c2 = i;
+	v = freq[i];
+	c2 = i;
      }
    }

    /* Done if we've merged everything into one frequency */
    if (c2 < 0)
      break;
-
+    
    /* Else merge the two counts/trees */
    freq[c1] += freq[c2];
    freq[c2] = 0;
@@ -1315,9 +1312,9 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
      c1 = others[c1];
      codesize[c1]++;
    }
-
+    
    others[c1] = c2;		/* chain c2 onto c1's tree branch */
-
+    
    /* Increment the codesize of everything in c2's tree branch */
    codesize[c2]++;
    while (others[c2] >= 0) {
@@ -1332,7 +1329,7 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
      /* The JPEG standard seems to think that this can't happen, */
      /* but I'm paranoid... */
      if (codesize[i] > MAX_CLEN)
-        ERREXIT(cinfo, JERR_HUFF_CLEN_OVERFLOW);
+	ERREXIT(cinfo, JERR_HUFF_CLEN_OVERFLOW);

      bits[codesize[i]]++;
    }
@@ -1348,13 +1345,13 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
   * shortest nonzero BITS entry is converted into a prefix for two code words
   * one bit longer.
   */
-
+  
  for (i = MAX_CLEN; i > 16; i--) {
    while (bits[i] > 0) {
      j = i - 2;		/* find length of new prefix to be used */
      while (bits[j] == 0)
-        j--;
-
+	j--;
+      
      bits[i] -= 2;		/* remove two symbols */
      bits[i-1]++;		/* one goes in this length */
      bits[j+1] += 2;		/* two new symbols in this length */
@@ -1366,10 +1363,10 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
  while (bits[i] == 0)		/* find largest codelength still in use */
    i--;
  bits[i]--;
-
+  
  /* Return final symbol counts (only for lengths 0..16) */
  MEMCOPY(htbl->bits, bits, SIZEOF(htbl->bits));
-
+  
  /* Return a list of the symbols sorted by code length */
  /* It's not real clear to me why we don't need to consider the codelength
   * changes made above, but the JPEG spec seems to think this works.
@@ -1378,8 +1375,8 @@ jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[])
  for (i = 1; i <= MAX_CLEN; i++) {
    for (j = 0; j <= 255; j++) {
      if (codesize[j] == i) {
-        htbl->huffval[p] = (UINT8) j;
-        p++;
+	htbl->huffval[p] = (UINT8) j;
+	p++;
      }
    }
  }
@@ -1419,22 +1416,22 @@ finish_pass_gather (j_compress_ptr cinfo)
    if (cinfo->Ss == 0 && cinfo->Ah == 0) {
      tbl = compptr->dc_tbl_no;
      if (! did_dc[tbl]) {
-        htblptr = & cinfo->dc_huff_tbl_ptrs[tbl];
-        if (*htblptr == NULL)
-          *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo);
-        jpeg_gen_optimal_table(cinfo, *htblptr, entropy->dc_count_ptrs[tbl]);
-        did_dc[tbl] = TRUE;
+	htblptr = & cinfo->dc_huff_tbl_ptrs[tbl];
+	if (*htblptr == NULL)
+	  *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo);
+	jpeg_gen_optimal_table(cinfo, *htblptr, entropy->dc_count_ptrs[tbl]);
+	did_dc[tbl] = TRUE;
      }
    }
    /* AC needs no table when not present */
    if (cinfo->Se) {
      tbl = compptr->ac_tbl_no;
      if (! did_ac[tbl]) {
-        htblptr = & cinfo->ac_huff_tbl_ptrs[tbl];
-        if (*htblptr == NULL)
-          *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo);
-        jpeg_gen_optimal_table(cinfo, *htblptr, entropy->ac_count_ptrs[tbl]);
-        did_ac[tbl] = TRUE;
+	htblptr = & cinfo->ac_huff_tbl_ptrs[tbl];
+	if (*htblptr == NULL)
+	  *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo);
+	jpeg_gen_optimal_table(cinfo, *htblptr, entropy->ac_count_ptrs[tbl]);
+	did_ac[tbl] = TRUE;
      }
    }
  }
@@ -1468,19 +1465,19 @@ start_pass_huff (j_compress_ptr cinfo, boolean gather_statistics)
    /* Select execution routine */
    if (cinfo->Ah == 0) {
      if (cinfo->Ss == 0)
-        entropy->pub.encode_mcu = encode_mcu_DC_first;
+	entropy->pub.encode_mcu = encode_mcu_DC_first;
      else
-        entropy->pub.encode_mcu = encode_mcu_AC_first;
+	entropy->pub.encode_mcu = encode_mcu_AC_first;
    } else {
      if (cinfo->Ss == 0)
-        entropy->pub.encode_mcu = encode_mcu_DC_refine;
+	entropy->pub.encode_mcu = encode_mcu_DC_refine;
      else {
-        entropy->pub.encode_mcu = encode_mcu_AC_refine;
-        /* AC refinement needs a correction bit buffer */
-        if (entropy->bit_buffer == NULL)
-          entropy->bit_buffer = (char *)
-            (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                        MAX_CORR_BITS * SIZEOF(char));
+	entropy->pub.encode_mcu = encode_mcu_AC_refine;
+	/* AC refinement needs a correction bit buffer */
+	if (entropy->bit_buffer == NULL)
+	  entropy->bit_buffer = (char *)
+	    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+					MAX_CORR_BITS * SIZEOF(char));
      }
    }

@@ -1501,22 +1498,22 @@ start_pass_huff (j_compress_ptr cinfo, boolean gather_statistics)
    if (cinfo->Ss == 0 && cinfo->Ah == 0) {
      tbl = compptr->dc_tbl_no;
      if (gather_statistics) {
-        /* Check for invalid table index */
-        /* (make_c_derived_tbl does this in the other path) */
-        if (tbl < 0 || tbl >= NUM_HUFF_TBLS)
-          ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tbl);
-        /* Allocate and zero the statistics tables */
-        /* Note that jpeg_gen_optimal_table expects 257 entries in each table! */
-        if (entropy->dc_count_ptrs[tbl] == NULL)
-          entropy->dc_count_ptrs[tbl] = (long *)
-            (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                        257 * SIZEOF(long));
-        MEMZERO(entropy->dc_count_ptrs[tbl], 257 * SIZEOF(long));
+	/* Check for invalid table index */
+	/* (make_c_derived_tbl does this in the other path) */
+	if (tbl < 0 || tbl >= NUM_HUFF_TBLS)
+	  ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tbl);
+	/* Allocate and zero the statistics tables */
+	/* Note that jpeg_gen_optimal_table expects 257 entries in each table! */
+	if (entropy->dc_count_ptrs[tbl] == NULL)
+	  entropy->dc_count_ptrs[tbl] = (long *)
+	    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+					257 * SIZEOF(long));
+	MEMZERO(entropy->dc_count_ptrs[tbl], 257 * SIZEOF(long));
      } else {
-        /* Compute derived values for Huffman tables */
-        /* We may do this more than once for a table, but it's not expensive */
-        jpeg_make_c_derived_tbl(cinfo, TRUE, tbl,
-                                & entropy->dc_derived_tbls[tbl]);
+	/* Compute derived values for Huffman tables */
+	/* We may do this more than once for a table, but it's not expensive */
+	jpeg_make_c_derived_tbl(cinfo, TRUE, tbl,
+				& entropy->dc_derived_tbls[tbl]);
      }
      /* Initialize DC predictions to 0 */
      entropy->saved.last_dc_val[ci] = 0;
@@ -1525,16 +1522,16 @@ start_pass_huff (j_compress_ptr cinfo, boolean gather_statistics)
    if (cinfo->Se) {
      tbl = compptr->ac_tbl_no;
      if (gather_statistics) {
-        if (tbl < 0 || tbl >= NUM_HUFF_TBLS)
-          ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tbl);
-        if (entropy->ac_count_ptrs[tbl] == NULL)
-          entropy->ac_count_ptrs[tbl] = (long *)
-            (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                        257 * SIZEOF(long));
-        MEMZERO(entropy->ac_count_ptrs[tbl], 257 * SIZEOF(long));
+	if (tbl < 0 || tbl >= NUM_HUFF_TBLS)
+	  ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tbl);
+	if (entropy->ac_count_ptrs[tbl] == NULL)
+	  entropy->ac_count_ptrs[tbl] = (long *)
+	    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+					257 * SIZEOF(long));
+	MEMZERO(entropy->ac_count_ptrs[tbl], 257 * SIZEOF(long));
      } else {
-        jpeg_make_c_derived_tbl(cinfo, FALSE, tbl,
-                                & entropy->ac_derived_tbls[tbl]);
+	jpeg_make_c_derived_tbl(cinfo, FALSE, tbl,
+				& entropy->ac_derived_tbls[tbl]);
      }
    }
  }
@@ -1561,8 +1558,8 @@ jinit_huff_encoder (j_compress_ptr cinfo)

  entropy = (huff_entropy_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(huff_entropy_encoder));
-  cinfo->entropy = (struct jpeg_entropy_encoder *) entropy;
+				SIZEOF(huff_entropy_encoder));
+  cinfo->entropy = &entropy->pub;
  entropy->pub.start_pass = start_pass_huff;

  /* Mark tables unallocated */
@@ -2,6 +2,7 @@
 * jcinit.c
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
+ * Modified 2003-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -29,6 +30,24 @@
 GLOBAL(void)
 jinit_compress_master (j_compress_ptr cinfo)
 {
+  long samplesperrow;
+  JDIMENSION jd_samplesperrow;
+
+  /* For now, precision must match compiled-in value... */
+  if (cinfo->data_precision != BITS_IN_JSAMPLE)
+    ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision);
+
+  /* Sanity check on image dimensions */
+  if (cinfo->image_height <= 0 || cinfo->image_width <= 0 ||
+      cinfo->input_components <= 0)
+    ERREXIT(cinfo, JERR_EMPTY_IMAGE);
+
+  /* Width of an input scanline must be representable as JDIMENSION. */
+  samplesperrow = (long) cinfo->image_width * (long) cinfo->input_components;
+  jd_samplesperrow = (JDIMENSION) samplesperrow;
+  if ((long) jd_samplesperrow != samplesperrow)
+    ERREXIT(cinfo, JERR_WIDTH_OVERFLOW);
+
  /* Initialize master control (includes parameter checking/processing) */
  jinit_c_master_control(cinfo, FALSE /* full compression */);

@@ -49,7 +68,7 @@ jinit_compress_master (j_compress_ptr cinfo)

  /* Need a full-image coefficient buffer in any multi-pass mode. */
  jinit_c_coef_controller(cinfo,
-                (boolean) (cinfo->num_scans > 1 || cinfo->optimize_coding));
+		(boolean) (cinfo->num_scans > 1 || cinfo->optimize_coding));
  jinit_c_main_controller(cinfo, FALSE /* never need full buffer here */);

  jinit_marker_writer(cinfo);
@@ -53,12 +53,12 @@ typedef my_main_controller * my_main_ptr;

 /* Forward declarations */
 METHODDEF(void) process_data_simple_main
-        JPP((j_compress_ptr cinfo, JSAMPARRAY input_buf,
-             JDIMENSION *in_row_ctr, JDIMENSION in_rows_avail));
+	JPP((j_compress_ptr cinfo, JSAMPARRAY input_buf,
+	     JDIMENSION *in_row_ctr, JDIMENSION in_rows_avail));
 #ifdef FULL_MAIN_BUFFER_SUPPORTED
 METHODDEF(void) process_data_buffer_main
-        JPP((j_compress_ptr cinfo, JSAMPARRAY input_buf,
-             JDIMENSION *in_row_ctr, JDIMENSION in_rows_avail));
+	JPP((j_compress_ptr cinfo, JSAMPARRAY input_buf,
+	     JDIMENSION *in_row_ctr, JDIMENSION in_rows_avail));
 #endif


@@ -112,8 +112,8 @@ start_pass_main (j_compress_ptr cinfo, J_BUF_MODE pass_mode)

 METHODDEF(void)
 process_data_simple_main (j_compress_ptr cinfo,
-                          JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
-                          JDIMENSION in_rows_avail)
+			  JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
+			  JDIMENSION in_rows_avail)
 {
  my_main_ptr mainp = (my_main_ptr) cinfo->main;

@@ -121,9 +121,9 @@ process_data_simple_main (j_compress_ptr cinfo,
    /* Read input data if we haven't filled the main buffer yet */
    if (mainp->rowgroup_ctr < (JDIMENSION) cinfo->min_DCT_v_scaled_size)
      (*cinfo->prep->pre_process_data) (cinfo,
-                                        input_buf, in_row_ctr, in_rows_avail,
-                                        mainp->buffer, &mainp->rowgroup_ctr,
-                                        (JDIMENSION) cinfo->min_DCT_v_scaled_size);
+					input_buf, in_row_ctr, in_rows_avail,
+					mainp->buffer, &mainp->rowgroup_ctr,
+					(JDIMENSION) cinfo->min_DCT_v_scaled_size);

    /* If we don't have a full iMCU row buffered, return to application for
     * more data.  Note that preprocessor will always pad to fill the iMCU row
@@ -141,8 +141,8 @@ process_data_simple_main (j_compress_ptr cinfo,
       * think we were done.
       */
      if (! mainp->suspended) {
-        (*in_row_ctr)--;
-        mainp->suspended = TRUE;
+	(*in_row_ctr)--;
+	mainp->suspended = TRUE;
      }
      return;
    }
@@ -168,8 +168,8 @@ process_data_simple_main (j_compress_ptr cinfo,

 METHODDEF(void)
 process_data_buffer_main (j_compress_ptr cinfo,
-                          JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
-                          JDIMENSION in_rows_avail)
+			  JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
+			  JDIMENSION in_rows_avail)
 {
  my_main_ptr mainp = (my_main_ptr) cinfo->main;
  int ci;
@@ -180,18 +180,18 @@ process_data_buffer_main (j_compress_ptr cinfo,
    /* Realign the virtual buffers if at the start of an iMCU row. */
    if (mainp->rowgroup_ctr == 0) {
      for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-           ci++, compptr++) {
-        mainp->buffer[ci] = (*cinfo->mem->access_virt_sarray)
-          ((j_common_ptr) cinfo, mainp->whole_image[ci], mainp->cur_iMCU_row *
-           ((JDIMENSION) (compptr->v_samp_factor * cinfo->min_DCT_v_scaled_size)),
-           (JDIMENSION) (compptr->v_samp_factor * cinfo->min_DCT_v_scaled_size),
-           writing);
+	   ci++, compptr++) {
+	mainp->buffer[ci] = (*cinfo->mem->access_virt_sarray)
+	  ((j_common_ptr) cinfo, mainp->whole_image[ci], mainp->cur_iMCU_row *
+	   ((JDIMENSION) (compptr->v_samp_factor * cinfo->min_DCT_v_scaled_size)),
+	   (JDIMENSION) (compptr->v_samp_factor * cinfo->min_DCT_v_scaled_size),
+	   writing);
      }
      /* In a read pass, pretend we just read some source data. */
      if (! writing) {
-        *in_row_ctr += (JDIMENSION)
-          (cinfo->max_v_samp_factor * cinfo->min_DCT_v_scaled_size);
-        mainp->rowgroup_ctr = (JDIMENSION) cinfo->min_DCT_v_scaled_size;
+	*in_row_ctr += (JDIMENSION)
+	  (cinfo->max_v_samp_factor * cinfo->min_DCT_v_scaled_size);
+	mainp->rowgroup_ctr = (JDIMENSION) cinfo->min_DCT_v_scaled_size;
      }
    }

@@ -199,35 +199,35 @@ process_data_buffer_main (j_compress_ptr cinfo,
    /* Note: preprocessor will pad if necessary to fill the last iMCU row. */
    if (writing) {
      (*cinfo->prep->pre_process_data) (cinfo,
-                                        input_buf, in_row_ctr, in_rows_avail,
-                                        mainp->buffer, &mainp->rowgroup_ctr,
-                                        (JDIMENSION) cinfo->min_DCT_v_scaled_size);
+					input_buf, in_row_ctr, in_rows_avail,
+					mainp->buffer, &mainp->rowgroup_ctr,
+					(JDIMENSION) cinfo->min_DCT_v_scaled_size);
      /* Return to application if we need more data to fill the iMCU row. */
      if (mainp->rowgroup_ctr < (JDIMENSION) cinfo->min_DCT_v_scaled_size)
-        return;
+	return;
    }

    /* Emit data, unless this is a sink-only pass. */
    if (mainp->pass_mode != JBUF_SAVE_SOURCE) {
      if (! (*cinfo->coef->compress_data) (cinfo, mainp->buffer)) {
-        /* If compressor did not consume the whole row, then we must need to
-         * suspend processing and return to the application.  In this situation
-         * we pretend we didn't yet consume the last input row; otherwise, if
-         * it happened to be the last row of the image, the application would
-         * think we were done.
-         */
-        if (! mainp->suspended) {
-          (*in_row_ctr)--;
-          mainp->suspended = TRUE;
-        }
-        return;
+	/* If compressor did not consume the whole row, then we must need to
+	 * suspend processing and return to the application.  In this situation
+	 * we pretend we didn't yet consume the last input row; otherwise, if
+	 * it happened to be the last row of the image, the application would
+	 * think we were done.
+	 */
+	if (! mainp->suspended) {
+	  (*in_row_ctr)--;
+	  mainp->suspended = TRUE;
+	}
+	return;
      }
      /* We did finish the row.  Undo our little suspension hack if a previous
       * call suspended; then mark the main buffer empty.
       */
      if (mainp->suspended) {
-        (*in_row_ctr)++;
-        mainp->suspended = FALSE;
+	(*in_row_ctr)++;
+	mainp->suspended = FALSE;
      }
    }

@@ -253,7 +253,7 @@ jinit_c_main_controller (j_compress_ptr cinfo, boolean need_full_buffer)

  mainp = (my_main_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_main_controller));
+				SIZEOF(my_main_controller));
  cinfo->main = &mainp->pub;
  mainp->pub.start_pass = start_pass_main;

@@ -269,14 +269,14 @@ jinit_c_main_controller (j_compress_ptr cinfo, boolean need_full_buffer)
    /* Allocate a full-image virtual array for each component */
    /* Note we pad the bottom to a multiple of the iMCU height */
    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      mainp->whole_image[ci] = (*cinfo->mem->request_virt_sarray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE,
-         compptr->width_in_blocks * ((JDIMENSION) compptr->DCT_h_scaled_size),
-         ((JDIMENSION) jround_up((long) compptr->height_in_blocks,
-                                 (long) compptr->v_samp_factor)) *
-         ((JDIMENSION) cinfo->min_DCT_v_scaled_size),
-         (JDIMENSION) (compptr->v_samp_factor * compptr->DCT_v_scaled_size));
+	((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE,
+	 compptr->width_in_blocks * ((JDIMENSION) compptr->DCT_h_scaled_size),
+	 ((JDIMENSION) jround_up((long) compptr->height_in_blocks,
+				 (long) compptr->v_samp_factor)) *
+	 ((JDIMENSION) cinfo->min_DCT_v_scaled_size),
+	 (JDIMENSION) (compptr->v_samp_factor * compptr->DCT_v_scaled_size));
    }
 #else
    ERREXIT(cinfo, JERR_BAD_BUFFER_MODE);
@@ -287,11 +287,11 @@ jinit_c_main_controller (j_compress_ptr cinfo, boolean need_full_buffer)
 #endif
    /* Allocate a strip buffer for each component */
    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      mainp->buffer[ci] = (*cinfo->mem->alloc_sarray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE,
-         compptr->width_in_blocks * ((JDIMENSION) compptr->DCT_h_scaled_size),
-         (JDIMENSION) (compptr->v_samp_factor * compptr->DCT_v_scaled_size));
+	((j_common_ptr) cinfo, JPOOL_IMAGE,
+	 compptr->width_in_blocks * ((JDIMENSION) compptr->DCT_h_scaled_size),
+	 (JDIMENSION) (compptr->v_samp_factor * compptr->DCT_v_scaled_size));
    }
  }
 }
@@ -2,7 +2,7 @@
 * jcmarker.c
 *
 * Copyright (C) 1991-1998, Thomas G. Lane.
- * Modified 2003-2012 by Guido Vollbeding.
+ * Modified 2003-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -172,7 +172,7 @@ emit_dqt (j_compress_ptr cinfo, int index)
      /* The table entries must be emitted in zigzag order. */
      unsigned int qval = qtbl->quantval[cinfo->natural_order[i]];
      if (prec)
-        emit_byte(cinfo, (int) (qval >> 8));
+	emit_byte(cinfo, (int) (qval >> 8));
      emit_byte(cinfo, (int) (qval & 0xFF));
    }

@@ -189,7 +189,7 @@ emit_dht (j_compress_ptr cinfo, int index, boolean is_ac)
 {
  JHUFF_TBL * htbl;
  int length, i;
-
+  
  if (is_ac) {
    htbl = cinfo->ac_huff_tbl_ptrs[index];
    index += 0x10;		/* output index has AC bit set */
@@ -199,23 +199,23 @@ emit_dht (j_compress_ptr cinfo, int index, boolean is_ac)

  if (htbl == NULL)
    ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, index);
-
+  
  if (! htbl->sent_table) {
    emit_marker(cinfo, M_DHT);
-
+    
    length = 0;
    for (i = 1; i <= 16; i++)
      length += htbl->bits[i];
-
+    
    emit_2bytes(cinfo, length + 2 + 1 + 16);
    emit_byte(cinfo, index);
-
+    
    for (i = 1; i <= 16; i++)
      emit_byte(cinfo, htbl->bits[i]);
-
+    
    for (i = 0; i < length; i++)
      emit_byte(cinfo, htbl->huffval[i]);
-
+    
    htbl->sent_table = TRUE;
  }
 }
@@ -257,12 +257,12 @@ emit_dac (j_compress_ptr cinfo)

    for (i = 0; i < NUM_ARITH_TBLS; i++) {
      if (dc_in_use[i]) {
-        emit_byte(cinfo, i);
-        emit_byte(cinfo, cinfo->arith_dc_L[i] + (cinfo->arith_dc_U[i]<<4));
+	emit_byte(cinfo, i);
+	emit_byte(cinfo, cinfo->arith_dc_L[i] + (cinfo->arith_dc_U[i]<<4));
      }
      if (ac_in_use[i]) {
-        emit_byte(cinfo, i + 0x10);
-        emit_byte(cinfo, cinfo->arith_ac_K[i]);
+	emit_byte(cinfo, i + 0x10);
+	emit_byte(cinfo, cinfo->arith_ac_K[i]);
      }
    }
  }
@@ -275,7 +275,7 @@ emit_dri (j_compress_ptr cinfo)
 /* Emit a DRI marker */
 {
  emit_marker(cinfo, M_DRI);
-
+  
  emit_2bytes(cinfo, 4);	/* fixed length */

  emit_2bytes(cinfo, (int) cinfo->restart_interval);
@@ -292,7 +292,7 @@ emit_lse_ict (j_compress_ptr cinfo)
    ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);

  emit_marker(cinfo, M_JPG8);
-
+  
  emit_2bytes(cinfo, 24);	/* fixed length */

  emit_byte(cinfo, 0x0D);	/* ID inverse transform specification */
@@ -319,9 +319,9 @@ emit_sof (j_compress_ptr cinfo, JPEG_MARKER code)
 {
  int ci;
  jpeg_component_info *compptr;
-
+  
  emit_marker(cinfo, code);
-
+  
  emit_2bytes(cinfo, 3 * cinfo->num_components + 2 + 5 + 1); /* length */

  /* Make sure image isn't bigger than SOF field can handle */
@@ -350,13 +350,13 @@ emit_sos (j_compress_ptr cinfo)
 {
  int i, td, ta;
  jpeg_component_info *compptr;
-
+  
  emit_marker(cinfo, M_SOS);
-
+  
  emit_2bytes(cinfo, 2 * cinfo->comps_in_scan + 2 + 1 + 3); /* length */
-
+  
  emit_byte(cinfo, cinfo->comps_in_scan);
-
+  
  for (i = 0; i < cinfo->comps_in_scan; i++) {
    compptr = cinfo->cur_comp_info[i];
    emit_byte(cinfo, compptr->component_id);
@@ -384,9 +384,9 @@ emit_pseudo_sos (j_compress_ptr cinfo)
 /* Emit a pseudo SOS marker */
 {
  emit_marker(cinfo, M_SOS);
-
+  
  emit_2bytes(cinfo, 2 + 1 + 3); /* length */
-
+  
  emit_byte(cinfo, 0); /* Ns */

  emit_byte(cinfo, 0); /* Ss */
@@ -410,9 +410,9 @@ emit_jfif_app0 (j_compress_ptr cinfo)
   * Thumbnail X size		(1 byte)
   * Thumbnail Y size		(1 byte)
   */
-
+  
  emit_marker(cinfo, M_APP0);
-
+  
  emit_2bytes(cinfo, 2 + 4 + 1 + 2 + 1 + 2 + 2 + 1 + 1); /* length */

  emit_byte(cinfo, 0x4A);	/* Identifier: ASCII "JFIF" */
@@ -449,9 +449,9 @@ emit_adobe_app14 (j_compress_ptr cinfo)
   * YCbCr, 2 if it's YCCK, 0 otherwise.  Adobe's definition has to do with
   * whether the encoder performed a transformation, which is pretty useless.
   */
-
+  
  emit_marker(cinfo, M_APP14);
-
+  
  emit_2bytes(cinfo, 2 + 5 + 2 + 2 + 2 + 1); /* length */

  emit_byte(cinfo, 0x41);	/* Identifier: ASCII "Adobe" */
@@ -508,8 +508,8 @@ write_marker_byte (j_compress_ptr cinfo, int val)
 * Write datastream header.
 * This consists of an SOI and optional APPn markers.
 * We recommend use of the JFIF marker, but not the Adobe marker,
- * when using YCbCr or grayscale data.  The JFIF marker should NOT
- * be used for any other JPEG colorspace.  The Adobe marker is helpful
+ * when using YCbCr or grayscale data.  The JFIF marker is also used
+ * for other standard JPEG colorspaces.  The Adobe marker is helpful
 * to distinguish RGB, CMYK, and YCCK colorspaces.
 * Note that an application can write additional header markers after
 * jpeg_start_compress returns.
@@ -547,7 +547,7 @@ write_frame_header (j_compress_ptr cinfo)
  int ci, prec;
  boolean is_baseline;
  jpeg_component_info *compptr;
-
+  
  /* Emit DQT for each quantization table.
   * Note that emit_dqt() suppresses any duplicate tables.
   */
@@ -567,9 +567,9 @@ write_frame_header (j_compress_ptr cinfo)
  } else {
    is_baseline = TRUE;
    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      if (compptr->dc_tbl_no > 1 || compptr->ac_tbl_no > 1)
-        is_baseline = FALSE;
+	is_baseline = FALSE;
    }
    if (prec && is_baseline) {
      is_baseline = FALSE;
@@ -630,10 +630,10 @@ write_scan_header (j_compress_ptr cinfo)
      compptr = cinfo->cur_comp_info[i];
      /* DC needs no table for refinement scan */
      if (cinfo->Ss == 0 && cinfo->Ah == 0)
-        emit_dht(cinfo, compptr->dc_tbl_no, FALSE);
+	emit_dht(cinfo, compptr->dc_tbl_no, FALSE);
      /* AC needs no table when not present */
      if (cinfo->Se)
-        emit_dht(cinfo, compptr->ac_tbl_no, TRUE);
+	emit_dht(cinfo, compptr->ac_tbl_no, TRUE);
    }
  }

@@ -682,9 +682,9 @@ write_tables_only (j_compress_ptr cinfo)
  if (! cinfo->arith_code) {
    for (i = 0; i < NUM_HUFF_TBLS; i++) {
      if (cinfo->dc_huff_tbl_ptrs[i] != NULL)
-        emit_dht(cinfo, i, FALSE);
+	emit_dht(cinfo, i, FALSE);
      if (cinfo->ac_huff_tbl_ptrs[i] != NULL)
-        emit_dht(cinfo, i, TRUE);
+	emit_dht(cinfo, i, TRUE);
    }
  }

@@ -704,7 +704,7 @@ jinit_marker_writer (j_compress_ptr cinfo)
  /* Create the subobject */
  marker = (my_marker_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_marker_writer));
+				SIZEOF(my_marker_writer));
  cinfo->marker = &marker->pub;
  /* Initialize method pointers */
  marker->pub.write_file_header = write_file_header;
@@ -2,13 +2,13 @@
 * jcmaster.c
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
- * Modified 2003-2011 by Guido Vollbeding.
+ * Modified 2003-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
 * This file contains master control logic for the JPEG compressor.
 * These routines are concerned with parameter validation, initial setup,
- * and inter-pass control (determining the number of passes and the work
+ * and inter-pass control (determining the number of passes and the work 
 * to be done in each pass).
 */

@@ -20,9 +20,9 @@
 /* Private state */

 typedef enum {
-        main_pass,		/* input data, also do first output step */
-        huff_opt_pass,		/* Huffman code optimization pass */
-        output_pass		/* data output pass */
+	main_pass,		/* input data, also do first output step */
+	huff_opt_pass,		/* Huffman code optimization pass */
+	output_pass		/* data output pass */
 } c_pass_type;

 typedef struct {
@@ -210,7 +210,7 @@ jpeg_calc_trans_dimensions (j_compress_ptr cinfo)
 {
  if (cinfo->min_DCT_h_scaled_size != cinfo->min_DCT_v_scaled_size)
    ERREXIT2(cinfo, JERR_BAD_DCTSIZE,
-             cinfo->min_DCT_h_scaled_size, cinfo->min_DCT_v_scaled_size);
+	     cinfo->min_DCT_h_scaled_size, cinfo->min_DCT_v_scaled_size);

  cinfo->block_size = cinfo->min_DCT_h_scaled_size;
 }
@@ -222,8 +222,6 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)
 {
  int ci, ssize;
  jpeg_component_info *compptr;
-  long samplesperrow;
-  JDIMENSION jd_samplesperrow;

  if (transcode_only)
    jpeg_calc_trans_dimensions(cinfo);
@@ -251,7 +249,7 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)

  /* Sanity check on image dimensions */
  if (cinfo->jpeg_height <= 0 || cinfo->jpeg_width <= 0 ||
-      cinfo->num_components <= 0 || cinfo->input_components <= 0)
+      cinfo->num_components <= 0)
    ERREXIT(cinfo, JERR_EMPTY_IMAGE);

  /* Make sure image isn't bigger than I can handle */
@@ -259,20 +257,14 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)
      (long) cinfo->jpeg_width > (long) JPEG_MAX_DIMENSION)
    ERREXIT1(cinfo, JERR_IMAGE_TOO_BIG, (unsigned int) JPEG_MAX_DIMENSION);

-  /* Width of an input scanline must be representable as JDIMENSION. */
-  samplesperrow = (long) cinfo->image_width * (long) cinfo->input_components;
-  jd_samplesperrow = (JDIMENSION) samplesperrow;
-  if ((long) jd_samplesperrow != samplesperrow)
-    ERREXIT(cinfo, JERR_WIDTH_OVERFLOW);
-
-  /* For now, precision must match compiled-in value... */
-  if (cinfo->data_precision != BITS_IN_JSAMPLE)
+  /* Only 8 to 12 bits data precision are supported for DCT based JPEG */
+  if (cinfo->data_precision < 8 || cinfo->data_precision > 12)
    ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision);

  /* Check that number of components won't exceed internal array sizes */
  if (cinfo->num_components > MAX_COMPONENTS)
    ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components,
-             MAX_COMPONENTS);
+	     MAX_COMPONENTS);

  /* Compute maximum sampling factors; check factor validity */
  cinfo->max_h_samp_factor = 1;
@@ -280,12 +272,12 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)
  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
       ci++, compptr++) {
    if (compptr->h_samp_factor<=0 || compptr->h_samp_factor>MAX_SAMP_FACTOR ||
-        compptr->v_samp_factor<=0 || compptr->v_samp_factor>MAX_SAMP_FACTOR)
+	compptr->v_samp_factor<=0 || compptr->v_samp_factor>MAX_SAMP_FACTOR)
      ERREXIT(cinfo, JERR_BAD_SAMPLING);
    cinfo->max_h_samp_factor = MAX(cinfo->max_h_samp_factor,
-                                   compptr->h_samp_factor);
+				   compptr->h_samp_factor);
    cinfo->max_v_samp_factor = MAX(cinfo->max_v_samp_factor,
-                                   compptr->v_samp_factor);
+				   compptr->v_samp_factor);
  }

  /* Compute dimensions of components */
@@ -301,8 +293,8 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)
    ssize = 1;
 #ifdef DCT_SCALING_SUPPORTED
    while (cinfo->min_DCT_h_scaled_size * ssize <=
-           (cinfo->do_fancy_downsampling ? DCTSIZE : DCTSIZE / 2) &&
-           (cinfo->max_h_samp_factor % (compptr->h_samp_factor * ssize * 2)) == 0) {
+	   (cinfo->do_fancy_downsampling ? DCTSIZE : DCTSIZE / 2) &&
+	   (cinfo->max_h_samp_factor % (compptr->h_samp_factor * ssize * 2)) == 0) {
      ssize = ssize * 2;
    }
 #endif
@@ -310,8 +302,8 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)
    ssize = 1;
 #ifdef DCT_SCALING_SUPPORTED
    while (cinfo->min_DCT_v_scaled_size * ssize <=
-           (cinfo->do_fancy_downsampling ? DCTSIZE : DCTSIZE / 2) &&
-           (cinfo->max_v_samp_factor % (compptr->v_samp_factor * ssize * 2)) == 0) {
+	   (cinfo->do_fancy_downsampling ? DCTSIZE : DCTSIZE / 2) &&
+	   (cinfo->max_v_samp_factor % (compptr->v_samp_factor * ssize * 2)) == 0) {
      ssize = ssize * 2;
    }
 #endif
@@ -319,28 +311,30 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)

    /* We don't support DCT ratios larger than 2. */
    if (compptr->DCT_h_scaled_size > compptr->DCT_v_scaled_size * 2)
-        compptr->DCT_h_scaled_size = compptr->DCT_v_scaled_size * 2;
+	compptr->DCT_h_scaled_size = compptr->DCT_v_scaled_size * 2;
    else if (compptr->DCT_v_scaled_size > compptr->DCT_h_scaled_size * 2)
-        compptr->DCT_v_scaled_size = compptr->DCT_h_scaled_size * 2;
+	compptr->DCT_v_scaled_size = compptr->DCT_h_scaled_size * 2;

    /* Size in DCT blocks */
    compptr->width_in_blocks = (JDIMENSION)
      jdiv_round_up((long) cinfo->jpeg_width * (long) compptr->h_samp_factor,
-                    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
+		    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
    compptr->height_in_blocks = (JDIMENSION)
      jdiv_round_up((long) cinfo->jpeg_height * (long) compptr->v_samp_factor,
-                    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+		    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
    /* Size in samples */
    compptr->downsampled_width = (JDIMENSION)
      jdiv_round_up((long) cinfo->jpeg_width *
-                    (long) (compptr->h_samp_factor * compptr->DCT_h_scaled_size),
-                    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
+		    (long) (compptr->h_samp_factor * compptr->DCT_h_scaled_size),
+		    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
    compptr->downsampled_height = (JDIMENSION)
      jdiv_round_up((long) cinfo->jpeg_height *
-                    (long) (compptr->v_samp_factor * compptr->DCT_v_scaled_size),
-                    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
-    /* Mark component needed (this flag isn't actually used for compression) */
-    compptr->component_needed = TRUE;
+		    (long) (compptr->v_samp_factor * compptr->DCT_v_scaled_size),
+		    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+    /* Don't need quantization scale after DCT,
+     * until color conversion says otherwise.
+     */
+    compptr->component_needed = FALSE;
  }

  /* Compute number of fully interleaved MCU rows (number of times that
@@ -348,7 +342,7 @@ initial_setup (j_compress_ptr cinfo, boolean transcode_only)
   */
  cinfo->total_iMCU_rows = (JDIMENSION)
    jdiv_round_up((long) cinfo->jpeg_height,
-                  (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+		  (long) (cinfo->max_v_samp_factor * cinfo->block_size));
 }


@@ -381,15 +375,15 @@ validate_script (j_compress_ptr cinfo)
 #ifdef C_PROGRESSIVE_SUPPORTED
    cinfo->progressive_mode = TRUE;
    last_bitpos_ptr = & last_bitpos[0][0];
-    for (ci = 0; ci < cinfo->num_components; ci++)
+    for (ci = 0; ci < cinfo->num_components; ci++) 
      for (coefi = 0; coefi < DCTSIZE2; coefi++)
-        *last_bitpos_ptr++ = -1;
+	*last_bitpos_ptr++ = -1;
 #else
    ERREXIT(cinfo, JERR_NOT_COMPILED);
 #endif
  } else {
    cinfo->progressive_mode = FALSE;
-    for (ci = 0; ci < cinfo->num_components; ci++)
+    for (ci = 0; ci < cinfo->num_components; ci++) 
      component_sent[ci] = FALSE;
  }

@@ -401,10 +395,10 @@ validate_script (j_compress_ptr cinfo)
    for (ci = 0; ci < ncomps; ci++) {
      thisi = scanptr->component_index[ci];
      if (thisi < 0 || thisi >= cinfo->num_components)
-        ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno);
+	ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno);
      /* Components must appear in SOF order within each scan */
      if (ci > 0 && thisi <= scanptr->component_index[ci-1])
-        ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno);
+	ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno);
    }
    /* Validate progression parameters */
    Ss = scanptr->Ss;
@@ -426,43 +420,43 @@ validate_script (j_compress_ptr cinfo)
 #define MAX_AH_AL 13
 #endif
      if (Ss < 0 || Ss >= DCTSIZE2 || Se < Ss || Se >= DCTSIZE2 ||
-          Ah < 0 || Ah > MAX_AH_AL || Al < 0 || Al > MAX_AH_AL)
-        ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
+	  Ah < 0 || Ah > MAX_AH_AL || Al < 0 || Al > MAX_AH_AL)
+	ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
      if (Ss == 0) {
-        if (Se != 0)		/* DC and AC together not OK */
-          ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
+	if (Se != 0)		/* DC and AC together not OK */
+	  ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
      } else {
-        if (ncomps != 1)	/* AC scans must be for only one component */
-          ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
+	if (ncomps != 1)	/* AC scans must be for only one component */
+	  ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
      }
      for (ci = 0; ci < ncomps; ci++) {
-        last_bitpos_ptr = & last_bitpos[scanptr->component_index[ci]][0];
-        if (Ss != 0 && last_bitpos_ptr[0] < 0) /* AC without prior DC scan */
-          ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
-        for (coefi = Ss; coefi <= Se; coefi++) {
-          if (last_bitpos_ptr[coefi] < 0) {
-            /* first scan of this coefficient */
-            if (Ah != 0)
-              ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
-          } else {
-            /* not first scan */
-            if (Ah != last_bitpos_ptr[coefi] || Al != Ah-1)
-              ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
-          }
-          last_bitpos_ptr[coefi] = Al;
-        }
+	last_bitpos_ptr = & last_bitpos[scanptr->component_index[ci]][0];
+	if (Ss != 0 && last_bitpos_ptr[0] < 0) /* AC without prior DC scan */
+	  ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
+	for (coefi = Ss; coefi <= Se; coefi++) {
+	  if (last_bitpos_ptr[coefi] < 0) {
+	    /* first scan of this coefficient */
+	    if (Ah != 0)
+	      ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
+	  } else {
+	    /* not first scan */
+	    if (Ah != last_bitpos_ptr[coefi] || Al != Ah-1)
+	      ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
+	  }
+	  last_bitpos_ptr[coefi] = Al;
+	}
      }
 #endif
    } else {
      /* For sequential JPEG, all progression parameters must be these: */
      if (Ss != 0 || Se != DCTSIZE2-1 || Ah != 0 || Al != 0)
-        ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
+	ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno);
      /* Make sure components are not sent twice */
      for (ci = 0; ci < ncomps; ci++) {
-        thisi = scanptr->component_index[ci];
-        if (component_sent[thisi])
-          ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno);
-        component_sent[thisi] = TRUE;
+	thisi = scanptr->component_index[ci];
+	if (component_sent[thisi])
+	  ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno);
+	component_sent[thisi] = TRUE;
      }
    }
  }
@@ -477,13 +471,13 @@ validate_script (j_compress_ptr cinfo)
     */
    for (ci = 0; ci < cinfo->num_components; ci++) {
      if (last_bitpos[ci][0] < 0)
-        ERREXIT(cinfo, JERR_MISSING_DATA);
+	ERREXIT(cinfo, JERR_MISSING_DATA);
    }
 #endif
  } else {
    for (ci = 0; ci < cinfo->num_components; ci++) {
      if (! component_sent[ci])
-        ERREXIT(cinfo, JERR_MISSING_DATA);
+	ERREXIT(cinfo, JERR_MISSING_DATA);
    }
  }
 }
@@ -539,7 +533,7 @@ select_scan_parameters (j_compress_ptr cinfo)
    cinfo->comps_in_scan = scanptr->comps_in_scan;
    for (ci = 0; ci < scanptr->comps_in_scan; ci++) {
      cinfo->cur_comp_info[ci] =
-        &cinfo->comp_info[scanptr->component_index[ci]];
+	&cinfo->comp_info[scanptr->component_index[ci]];
    }
    if (cinfo->progressive_mode) {
      cinfo->Ss = scanptr->Ss;
@@ -555,7 +549,7 @@ select_scan_parameters (j_compress_ptr cinfo)
    /* Prepare for single sequential-JPEG scan containing all components */
    if (cinfo->num_components > MAX_COMPS_IN_SCAN)
      ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components,
-               MAX_COMPS_IN_SCAN);
+	       MAX_COMPS_IN_SCAN);
    cinfo->comps_in_scan = cinfo->num_components;
    for (ci = 0; ci < cinfo->num_components; ci++) {
      cinfo->cur_comp_info[ci] = &cinfo->comp_info[ci];
@@ -575,16 +569,16 @@ per_scan_setup (j_compress_ptr cinfo)
 {
  int ci, mcublks, tmp;
  jpeg_component_info *compptr;
-
+  
  if (cinfo->comps_in_scan == 1) {
-
+    
    /* Noninterleaved (single-component) scan */
    compptr = cinfo->cur_comp_info[0];
-
+    
    /* Overall image size in MCUs */
    cinfo->MCUs_per_row = compptr->width_in_blocks;
    cinfo->MCU_rows_in_scan = compptr->height_in_blocks;
-
+    
    /* For noninterleaved scan, always one block per MCU */
    compptr->MCU_width = 1;
    compptr->MCU_height = 1;
@@ -597,28 +591,28 @@ per_scan_setup (j_compress_ptr cinfo)
    tmp = (int) (compptr->height_in_blocks % compptr->v_samp_factor);
    if (tmp == 0) tmp = compptr->v_samp_factor;
    compptr->last_row_height = tmp;
-
+    
    /* Prepare array describing MCU composition */
    cinfo->blocks_in_MCU = 1;
    cinfo->MCU_membership[0] = 0;
-
+    
  } else {
-
+    
    /* Interleaved (multi-component) scan */
    if (cinfo->comps_in_scan <= 0 || cinfo->comps_in_scan > MAX_COMPS_IN_SCAN)
      ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->comps_in_scan,
-               MAX_COMPS_IN_SCAN);
-
+	       MAX_COMPS_IN_SCAN);
+    
    /* Overall image size in MCUs */
    cinfo->MCUs_per_row = (JDIMENSION)
      jdiv_round_up((long) cinfo->jpeg_width,
-                    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
+		    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
    cinfo->MCU_rows_in_scan = (JDIMENSION)
      jdiv_round_up((long) cinfo->jpeg_height,
-                    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
-
+		    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+    
    cinfo->blocks_in_MCU = 0;
-
+    
    for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
      compptr = cinfo->cur_comp_info[ci];
      /* Sampling factors give # of blocks of component in each MCU */
@@ -636,12 +630,12 @@ per_scan_setup (j_compress_ptr cinfo)
      /* Prepare array describing MCU composition */
      mcublks = compptr->MCU_blocks;
      if (cinfo->blocks_in_MCU + mcublks > C_MAX_BLOCKS_IN_MCU)
-        ERREXIT(cinfo, JERR_BAD_MCU_SIZE);
+	ERREXIT(cinfo, JERR_BAD_MCU_SIZE);
      while (mcublks-- > 0) {
-        cinfo->MCU_membership[cinfo->blocks_in_MCU++] = ci;
+	cinfo->MCU_membership[cinfo->blocks_in_MCU++] = ci;
      }
    }
-
+    
  }

  /* Convert restart specified in rows to actual MCU count. */
@@ -681,8 +675,8 @@ prepare_for_pass (j_compress_ptr cinfo)
    (*cinfo->fdct->start_pass) (cinfo);
    (*cinfo->entropy->start_pass) (cinfo, cinfo->optimize_coding);
    (*cinfo->coef->start_pass) (cinfo,
-                                (master->total_passes > 1 ?
-                                 JBUF_SAVE_AND_PASS : JBUF_PASS_THRU));
+				(master->total_passes > 1 ?
+				 JBUF_SAVE_AND_PASS : JBUF_PASS_THRU));
    (*cinfo->main->start_pass) (cinfo, JBUF_PASS_THRU);
    if (cinfo->optimize_coding) {
      /* No immediate data output; postpone writing frame/scan headers */
@@ -810,8 +804,8 @@ jinit_c_master_control (j_compress_ptr cinfo, boolean transcode_only)

  master = (my_master_ptr)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  SIZEOF(my_comp_master));
-  cinfo->master = (struct jpeg_comp_master *) master;
+				  SIZEOF(my_comp_master));
+  cinfo->master = &master->pub;
  master->pub.prepare_for_pass = prepare_for_pass;
  master->pub.pass_startup = pass_startup;
  master->pub.finish_pass = finish_pass_master;
@@ -833,10 +827,14 @@ jinit_c_master_control (j_compress_ptr cinfo, boolean transcode_only)
    cinfo->num_scans = 1;
  }

-  if ((cinfo->progressive_mode || cinfo->block_size < DCTSIZE) &&
-      !cinfo->arith_code)			/*  TEMPORARY HACK ??? */
-    /* assume default tables no good for progressive or downscale mode */
-    cinfo->optimize_coding = TRUE;
+  if (cinfo->optimize_coding)
+    cinfo->arith_code = FALSE; /* disable arithmetic coding */
+  else if (! cinfo->arith_code &&
+	   (cinfo->progressive_mode ||
+	    (cinfo->block_size > 1 && cinfo->block_size < DCTSIZE)))
+    /* TEMPORARY HACK ??? */
+    /* assume default tables no good for progressive or reduced AC mode */
+    cinfo->optimize_coding = TRUE; /* force Huffman optimization */

  /* Initialize my private state */
  if (transcode_only) {
@@ -1,136 +1,52 @@
+/* jconfig.vc --- jconfig.h for Microsoft Visual C++ on Windows 9x or NT. */
+/* This file also works for Borland C++ 32-bit (bcc32) on Windows 9x or NT. */
+/* see jconfig.txt for explanations */
+
 #define HAVE_PROTOTYPES
 #define HAVE_UNSIGNED_CHAR
 #define HAVE_UNSIGNED_SHORT
-
-/* Define this if an ordinary "char" type is unsigned.
- * If you're not sure, leaving it undefined will work at some cost in speed.
- * If you defined HAVE_UNSIGNED_CHAR then the speed difference is minimal.
- */
+/* #define void char */
+/* #define const */
 #undef CHAR_IS_UNSIGNED
-
-#if defined __MINGW__ || defined __MINGW32__ || (!defined WIN32 && !defined _WIN32)
-/* Define this if your system has an ANSI-conforming <stddef.h> file.
- */
 #define HAVE_STDDEF_H
-
-/* Define this if your system has an ANSI-conforming <stdlib.h> file.
- */
 #define HAVE_STDLIB_H
-#endif
-
-/* Define this if your system does not have an ANSI/SysV <string.h>,
- * but does have a BSD-style <strings.h>.
- */
 #undef NEED_BSD_STRINGS
-
-/* Define this if your system does not provide typedef size_t in any of the
- * ANSI-standard places (stddef.h, stdlib.h, or stdio.h), but places it in
- * <sys/types.h> instead.
- */
 #undef NEED_SYS_TYPES_H
-
-/* For 80x86 machines, you need to define NEED_FAR_POINTERS,
- * unless you are using a large-data memory model or 80386 flat-memory mode.
- * On less brain-damaged CPUs this symbol must not be defined.
- * (Defining this symbol causes large data structures to be referenced through
- * "far" pointers and to be allocated with a special version of malloc.)
- */
-#undef NEED_FAR_POINTERS
-
-/* Define this if your linker needs global names to be unique in less
- * than the first 15 characters.
- */
+#undef NEED_FAR_POINTERS	/* we presume a 32-bit flat memory model */
 #undef NEED_SHORT_EXTERNAL_NAMES
-
-/* Although a real ANSI C compiler can deal perfectly well with pointers to
- * unspecified structures (see "incomplete types" in the spec), a few pre-ANSI
- * and pseudo-ANSI compilers get confused.  To keep one of these bozos happy,
- * define INCOMPLETE_TYPES_BROKEN.  This is not recommended unless you
- * actually get "missing structure definition" warnings or errors while
- * compiling the JPEG code.
- */
 #undef INCOMPLETE_TYPES_BROKEN

-/* Define "boolean" as unsigned char, not int, on Windows systems.
- */
-#ifdef _WIN32
+/* Define "boolean" as unsigned char, not enum, per Windows custom */
 #ifndef __RPCNDR_H__		/* don't conflict if rpcndr.h already read */
 typedef unsigned char boolean;
 #endif
-#define HAVE_BOOLEAN		/* prevent jmorecfg.h from redefining it */
+#ifndef FALSE			/* in case these macros already exist */
+#define FALSE	0		/* values of boolean */
 #endif
+#ifndef TRUE
+#define TRUE	1
+#endif
+#define HAVE_BOOLEAN		/* prevent jmorecfg.h from redefining it */


-/*
- * The following options affect code selection within the JPEG library,
- * but they don't need to be visible to applications using the library.
- * To minimize application namespace pollution, the symbols won't be
- * defined unless JPEG_INTERNALS has been defined.
- */
-
 #ifdef JPEG_INTERNALS

-/* Define this if your compiler implements ">>" on signed values as a logical
- * (unsigned) shift; leave it undefined if ">>" is a signed (arithmetic) shift,
- * which is the normal and rational definition.
- */
 #undef RIGHT_SHIFT_IS_UNSIGNED

-/* These are for configuring the JPEG memory manager. */
-#define DEFAULT_MAX_MEM 1073741824 /*1Gb*/
-
-#if !defined WIN32 && !defined _WIN32
-#define INLINE __inline__
-#undef NO_MKTEMP
-#endif
-
 #endif /* JPEG_INTERNALS */

-
-/*
- * The remaining options do not affect the JPEG library proper,
- * but only the sample applications cjpeg/djpeg (see cjpeg.c, djpeg.c).
- * Other applications can ignore these.
- */
-
 #ifdef JPEG_CJPEG_DJPEG

-/* These defines indicate which image (non-JPEG) file formats are allowed. */
-
 #define BMP_SUPPORTED		/* BMP image file format */
 #define GIF_SUPPORTED		/* GIF image file format */
 #define PPM_SUPPORTED		/* PBMPLUS PPM/PGM image file format */
 #undef RLE_SUPPORTED		/* Utah RLE image file format */
 #define TARGA_SUPPORTED		/* Targa image file format */

-/* Define this if you want to name both input and output files on the command
- * line, rather than using stdout and optionally stdin.  You MUST do this if
- * your system can't cope with binary I/O to stdin/stdout.  See comments at
- * head of cjpeg.c or djpeg.c.
- */
-#if defined WIN32 || defined _WIN32
 #define TWO_FILE_COMMANDLINE	/* optional */
 #define USE_SETMODE		/* Microsoft has setmode() */
-#else
-#undef TWO_FILE_COMMANDLINE
-#endif
-
-/* Define this if your system needs explicit cleanup of temporary files.
- * This is crucial under MS-DOS, where the temporary "files" may be areas
- * of extended memory; on most other systems it's not as important.
- */
 #undef NEED_SIGNAL_CATCHER
-
-/* By default, we open image files with fopen(...,"rb") or fopen(...,"wb").
- * This is necessary on systems that distinguish text files from binary files,
- * and is harmless on most systems that don't.  If you have one of the rare
- * systems that complains about the "b" spec, define this symbol.
- */
 #undef DONT_USE_B_MODE
-
-/* Define this if you want percent-done progress reports from cjpeg/djpeg.
- */
-#undef PROGRESS_REPORT
-
+#undef PROGRESS_REPORT		/* optional */

 #endif /* JPEG_CJPEG_DJPEG */
@@ -2,7 +2,7 @@
 * jcparam.c
 *
 * Copyright (C) 1991-1998, Thomas G. Lane.
- * Modified 2003-2012 by Guido Vollbeding.
+ * Modified 2003-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -22,8 +22,8 @@

 GLOBAL(void)
 jpeg_add_quant_table (j_compress_ptr cinfo, int which_tbl,
-                      const unsigned int *basic_table,
-                      int scale_factor, boolean force_baseline)
+		      const unsigned int *basic_table,
+		      int scale_factor, boolean force_baseline)
 /* Define a quantization table equal to the basic_table times
 * a scale factor (given as a percentage).
 * If force_baseline is TRUE, the computed quantization table entries
@@ -96,15 +96,15 @@ jpeg_default_qtables (j_compress_ptr cinfo, boolean force_baseline)
 {
  /* Set up two quantization tables using the specified scaling */
  jpeg_add_quant_table(cinfo, 0, std_luminance_quant_tbl,
-                       cinfo->q_scale_factor[0], force_baseline);
+		       cinfo->q_scale_factor[0], force_baseline);
  jpeg_add_quant_table(cinfo, 1, std_chrominance_quant_tbl,
-                       cinfo->q_scale_factor[1], force_baseline);
+		       cinfo->q_scale_factor[1], force_baseline);
 }


 GLOBAL(void)
 jpeg_set_linear_quality (j_compress_ptr cinfo, int scale_factor,
-                         boolean force_baseline)
+			 boolean force_baseline)
 /* Set or change the 'quality' (quantization) setting, using default tables
 * and a straight percentage-scaling quality scale.  In most cases it's better
 * to use jpeg_set_quality (below); this entry point is provided for
@@ -113,9 +113,9 @@ jpeg_set_linear_quality (j_compress_ptr cinfo, int scale_factor,
 {
  /* Set up two quantization tables using the specified scaling */
  jpeg_add_quant_table(cinfo, 0, std_luminance_quant_tbl,
-                       scale_factor, force_baseline);
+		       scale_factor, force_baseline);
  jpeg_add_quant_table(cinfo, 1, std_chrominance_quant_tbl,
-                       scale_factor, force_baseline);
+		       scale_factor, force_baseline);
 }


@@ -167,7 +167,7 @@ jpeg_set_quality (j_compress_ptr cinfo, int quality, boolean force_baseline)

 LOCAL(void)
 add_huff_table (j_compress_ptr cinfo,
-                JHUFF_TBL **htblptr, const UINT8 *bits, const UINT8 *val)
+		JHUFF_TBL **htblptr, const UINT8 *bits, const UINT8 *val)
 /* Define a Huffman table */
 {
  int nsymbols, len;
@@ -204,12 +204,12 @@ std_huff_tables (j_compress_ptr cinfo)
    { /* 0-base */ 0, 0, 1, 5, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0 };
  static const UINT8 val_dc_luminance[] =
    { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 };
-
+  
  static const UINT8 bits_dc_chrominance[17] =
    { /* 0-base */ 0, 0, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0 };
  static const UINT8 val_dc_chrominance[] =
    { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 };
-
+  
  static const UINT8 bits_ac_luminance[17] =
    { /* 0-base */ 0, 0, 2, 1, 3, 3, 2, 4, 3, 5, 5, 4, 4, 0, 0, 1, 0x7d };
  static const UINT8 val_ac_luminance[] =
@@ -234,7 +234,7 @@ std_huff_tables (j_compress_ptr cinfo)
      0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9, 0xea,
      0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8,
      0xf9, 0xfa };
-
+  
  static const UINT8 bits_ac_chrominance[17] =
    { /* 0-base */ 0, 0, 2, 1, 2, 4, 4, 3, 4, 7, 5, 4, 4, 0, 1, 2, 0x77 };
  static const UINT8 val_ac_chrominance[] =
@@ -259,15 +259,15 @@ std_huff_tables (j_compress_ptr cinfo)
      0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9,
      0xea, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8,
      0xf9, 0xfa };
-
+  
  add_huff_table(cinfo, &cinfo->dc_huff_tbl_ptrs[0],
-                 bits_dc_luminance, val_dc_luminance);
+		 bits_dc_luminance, val_dc_luminance);
  add_huff_table(cinfo, &cinfo->ac_huff_tbl_ptrs[0],
-                 bits_ac_luminance, val_ac_luminance);
+		 bits_ac_luminance, val_ac_luminance);
  add_huff_table(cinfo, &cinfo->dc_huff_tbl_ptrs[1],
-                 bits_dc_chrominance, val_dc_chrominance);
+		 bits_dc_chrominance, val_dc_chrominance);
  add_huff_table(cinfo, &cinfo->ac_huff_tbl_ptrs[1],
-                 bits_ac_chrominance, val_ac_chrominance);
+		 bits_ac_chrominance, val_ac_chrominance);
 }


@@ -297,7 +297,7 @@ jpeg_set_defaults (j_compress_ptr cinfo)
  if (cinfo->comp_info == NULL)
    cinfo->comp_info = (jpeg_component_info *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                  MAX_COMPONENTS * SIZEOF(jpeg_component_info));
+				  MAX_COMPONENTS * SIZEOF(jpeg_component_info));

  /* Initialize everything not dependent on the color space */

@@ -323,18 +323,17 @@ jpeg_set_defaults (j_compress_ptr cinfo)
  /* Expect normal source image, not raw downsampled data */
  cinfo->raw_data_in = FALSE;

-  /* Use Huffman coding, not arithmetic coding, by default */
-  cinfo->arith_code = FALSE;
+  /* The standard Huffman tables are only valid for 8-bit data precision.
+   * If the precision is higher, use arithmetic coding.
+   * (Alternatively, using Huffman coding would be possible with forcing
+   * optimization on so that usable tables will be computed, or by
+   * supplying default tables that are valid for the desired precision.)
+   * Otherwise, use Huffman coding by default.
+   */
+  cinfo->arith_code = cinfo->data_precision > 8 ? TRUE : FALSE;

  /* By default, don't do extra passes to optimize entropy coding */
  cinfo->optimize_coding = FALSE;
-  /* The standard Huffman tables are only valid for 8-bit data precision.
-   * If the precision is higher, force optimization on so that usable
-   * tables will be computed.  This test can be removed if default tables
-   * are supplied that are valid for the desired precision.
-   */
-  if (cinfo->data_precision > 8)
-    cinfo->optimize_coding = TRUE;

  /* By default, use the simpler non-cosited sampling alignment */
  cinfo->CCIR601_sampling = FALSE;
@@ -360,6 +359,9 @@ jpeg_set_defaults (j_compress_ptr cinfo)
   * JFIF_minor_version to 2.  We could probably get away with just defaulting
   * to 1.02, but there may still be some decoders in use that will complain
   * about that; saying 1.01 should minimize compatibility problems.
+   *
+   * For wide gamut colorspaces (BG_RGB and BG_YCC), the major version will be
+   * overridden by jpeg_set_colorspace and set to 2.
   */
  cinfo->JFIF_major_version = 1; /* Default JFIF version = 1.01 */
  cinfo->JFIF_minor_version = 1;
@@ -384,6 +386,9 @@ GLOBAL(void)
 jpeg_default_colorspace (j_compress_ptr cinfo)
 {
  switch (cinfo->in_color_space) {
+  case JCS_UNKNOWN:
+    jpeg_set_colorspace(cinfo, JCS_UNKNOWN);
+    break;
  case JCS_GRAYSCALE:
    jpeg_set_colorspace(cinfo, JCS_GRAYSCALE);
    break;
@@ -399,8 +404,12 @@ jpeg_default_colorspace (j_compress_ptr cinfo)
  case JCS_YCCK:
    jpeg_set_colorspace(cinfo, JCS_YCCK);
    break;
-  case JCS_UNKNOWN:
-    jpeg_set_colorspace(cinfo, JCS_UNKNOWN);
+  case JCS_BG_RGB:
+    /* No translation for now -- conversion to BG_YCC not yet supportet */
+    jpeg_set_colorspace(cinfo, JCS_BG_RGB);
+    break;
+  case JCS_BG_YCC:
+    jpeg_set_colorspace(cinfo, JCS_BG_YCC);
    break;
  default:
    ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE);
@@ -441,29 +450,40 @@ jpeg_set_colorspace (j_compress_ptr cinfo, J_COLOR_SPACE colorspace)
  cinfo->write_Adobe_marker = FALSE; /* write no Adobe marker by default */

  switch (colorspace) {
+  case JCS_UNKNOWN:
+    cinfo->num_components = cinfo->input_components;
+    if (cinfo->num_components < 1 || cinfo->num_components > MAX_COMPONENTS)
+      ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components,
+	       MAX_COMPONENTS);
+    for (ci = 0; ci < cinfo->num_components; ci++) {
+      SET_COMP(ci, ci, 1,1, 0, 0,0);
+    }
+    break;
  case JCS_GRAYSCALE:
    cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */
    cinfo->num_components = 1;
    /* JFIF specifies component ID 1 */
-    SET_COMP(0, 1, 1,1, 0, 0,0);
+    SET_COMP(0, 0x01, 1,1, 0, 0,0);
    break;
  case JCS_RGB:
    cinfo->write_Adobe_marker = TRUE; /* write Adobe marker to flag RGB */
    cinfo->num_components = 3;
-    SET_COMP(0, 0x52 /* 'R' */, 1,1, 0, 0,0);
-    SET_COMP(1, 0x47 /* 'G' */, 1,1, 0,
-                cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0,
-                cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0);
-    SET_COMP(2, 0x42 /* 'B' */, 1,1, 0, 0,0);
+    SET_COMP(0, 0x52 /* 'R' */, 1,1, 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0);
+    SET_COMP(1, 0x47 /* 'G' */, 1,1, 0, 0,0);
+    SET_COMP(2, 0x42 /* 'B' */, 1,1, 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0);
    break;
  case JCS_YCbCr:
    cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */
    cinfo->num_components = 3;
    /* JFIF specifies component IDs 1,2,3 */
    /* We default to 2x2 subsamples of chrominance */
-    SET_COMP(0, 1, 2,2, 0, 0,0);
-    SET_COMP(1, 2, 1,1, 1, 1,1);
-    SET_COMP(2, 3, 1,1, 1, 1,1);
+    SET_COMP(0, 0x01, 2,2, 0, 0,0);
+    SET_COMP(1, 0x02, 1,1, 1, 1,1);
+    SET_COMP(2, 0x03, 1,1, 1, 1,1);
    break;
  case JCS_CMYK:
    cinfo->write_Adobe_marker = TRUE; /* write Adobe marker to flag CMYK */
@@ -476,19 +496,33 @@ jpeg_set_colorspace (j_compress_ptr cinfo, J_COLOR_SPACE colorspace)
  case JCS_YCCK:
    cinfo->write_Adobe_marker = TRUE; /* write Adobe marker to flag YCCK */
    cinfo->num_components = 4;
-    SET_COMP(0, 1, 2,2, 0, 0,0);
-    SET_COMP(1, 2, 1,1, 1, 1,1);
-    SET_COMP(2, 3, 1,1, 1, 1,1);
-    SET_COMP(3, 4, 2,2, 0, 0,0);
+    SET_COMP(0, 0x01, 2,2, 0, 0,0);
+    SET_COMP(1, 0x02, 1,1, 1, 1,1);
+    SET_COMP(2, 0x03, 1,1, 1, 1,1);
+    SET_COMP(3, 0x04, 2,2, 0, 0,0);
    break;
-  case JCS_UNKNOWN:
-    cinfo->num_components = cinfo->input_components;
-    if (cinfo->num_components < 1 || cinfo->num_components > MAX_COMPONENTS)
-      ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components,
-               MAX_COMPONENTS);
-    for (ci = 0; ci < cinfo->num_components; ci++) {
-      SET_COMP(ci, ci, 1,1, 0, 0,0);
-    }
+  case JCS_BG_RGB:
+    cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */
+    cinfo->JFIF_major_version = 2;   /* Set JFIF major version = 2 */
+    cinfo->num_components = 3;
+    /* Add offset 0x20 to the normal R/G/B component IDs */
+    SET_COMP(0, 0x72 /* 'r' */, 1,1, 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0);
+    SET_COMP(1, 0x67 /* 'g' */, 1,1, 0, 0,0);
+    SET_COMP(2, 0x62 /* 'b' */, 1,1, 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0,
+		cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0);
+    break;
+  case JCS_BG_YCC:
+    cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */
+    cinfo->JFIF_major_version = 2;   /* Set JFIF major version = 2 */
+    cinfo->num_components = 3;
+    /* Add offset 0x20 to the normal Cb/Cr component IDs */
+    /* We default to 2x2 subsamples of chrominance */
+    SET_COMP(0, 0x01, 2,2, 0, 0,0);
+    SET_COMP(1, 0x22, 1,1, 1, 1,1);
+    SET_COMP(2, 0x23, 1,1, 1, 1,1);
    break;
  default:
    ERREXIT(cinfo, JERR_BAD_J_COLORSPACE);
@@ -500,7 +534,7 @@ jpeg_set_colorspace (j_compress_ptr cinfo, J_COLOR_SPACE colorspace)

 LOCAL(jpeg_scan_info *)
 fill_a_scan (jpeg_scan_info * scanptr, int ci,
-             int Ss, int Se, int Ah, int Al)
+	     int Ss, int Se, int Ah, int Al)
 /* Support routine: generate one scan for specified component */
 {
  scanptr->comps_in_scan = 1;
@@ -515,7 +549,7 @@ fill_a_scan (jpeg_scan_info * scanptr, int ci,

 LOCAL(jpeg_scan_info *)
 fill_scans (jpeg_scan_info * scanptr, int ncomps,
-            int Ss, int Se, int Ah, int Al)
+	    int Ss, int Se, int Ah, int Al)
 /* Support routine: generate one scan for each component */
 {
  int ci;
@@ -572,8 +606,10 @@ jpeg_simple_progression (j_compress_ptr cinfo)
    ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state);

  /* Figure space needed for script.  Calculation must match code below! */
-  if (ncomps == 3 && cinfo->jpeg_color_space == JCS_YCbCr) {
-    /* Custom script for YCbCr color images. */
+  if (ncomps == 3 &&
+      (cinfo->jpeg_color_space == JCS_YCbCr ||
+       cinfo->jpeg_color_space == JCS_BG_YCC)) {
+    /* Custom script for YCC color images. */
    nscans = 10;
  } else {
    /* All-purpose script for other color spaces. */
@@ -588,20 +624,22 @@ jpeg_simple_progression (j_compress_ptr cinfo)
   * multiple compressions without changing the settings.  To avoid a memory
   * leak if jpeg_simple_progression is called repeatedly for the same JPEG
   * object, we try to re-use previously allocated space, and we allocate
-   * enough space to handle YCbCr even if initially asked for grayscale.
+   * enough space to handle YCC even if initially asked for grayscale.
   */
  if (cinfo->script_space == NULL || cinfo->script_space_size < nscans) {
    cinfo->script_space_size = MAX(nscans, 10);
    cinfo->script_space = (jpeg_scan_info *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                        cinfo->script_space_size * SIZEOF(jpeg_scan_info));
+			cinfo->script_space_size * SIZEOF(jpeg_scan_info));
  }
  scanptr = cinfo->script_space;
  cinfo->scan_info = scanptr;
  cinfo->num_scans = nscans;

-  if (ncomps == 3 && cinfo->jpeg_color_space == JCS_YCbCr) {
-    /* Custom script for YCbCr color images. */
+  if (ncomps == 3 &&
+      (cinfo->jpeg_color_space == JCS_YCbCr ||
+       cinfo->jpeg_color_space == JCS_BG_YCC)) {
+    /* Custom script for YCC color images. */
    /* Initial DC scan */
    scanptr = fill_dc_scans(scanptr, ncomps, 0, 1);
    /* Initial AC scan: get some luma data out in a hurry */
@@ -104,13 +104,13 @@ start_pass_prep (j_compress_ptr cinfo, J_BUF_MODE pass_mode)

 LOCAL(void)
 expand_bottom_edge (JSAMPARRAY image_data, JDIMENSION num_cols,
-                    int input_rows, int output_rows)
+		    int input_rows, int output_rows)
 {
  register int row;

  for (row = input_rows; row < output_rows; row++) {
    jcopy_sample_rows(image_data, input_rows-1, image_data, row,
-                      1, num_cols);
+		      1, num_cols);
  }
 }

@@ -126,10 +126,10 @@ expand_bottom_edge (JSAMPARRAY image_data, JDIMENSION num_cols,

 METHODDEF(void)
 pre_process_data (j_compress_ptr cinfo,
-                  JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
-                  JDIMENSION in_rows_avail,
-                  JSAMPIMAGE output_buf, JDIMENSION *out_row_group_ctr,
-                  JDIMENSION out_row_groups_avail)
+		  JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
+		  JDIMENSION in_rows_avail,
+		  JSAMPIMAGE output_buf, JDIMENSION *out_row_group_ctr,
+		  JDIMENSION out_row_groups_avail)
 {
  my_prep_ptr prep = (my_prep_ptr) cinfo->prep;
  int numrows, ci;
@@ -137,32 +137,32 @@ pre_process_data (j_compress_ptr cinfo,
  jpeg_component_info * compptr;

  while (*in_row_ctr < in_rows_avail &&
-         *out_row_group_ctr < out_row_groups_avail) {
+	 *out_row_group_ctr < out_row_groups_avail) {
    /* Do color conversion to fill the conversion buffer. */
    inrows = in_rows_avail - *in_row_ctr;
    numrows = cinfo->max_v_samp_factor - prep->next_buf_row;
    numrows = (int) MIN((JDIMENSION) numrows, inrows);
    (*cinfo->cconvert->color_convert) (cinfo, input_buf + *in_row_ctr,
-                                       prep->color_buf,
-                                       (JDIMENSION) prep->next_buf_row,
-                                       numrows);
+				       prep->color_buf,
+				       (JDIMENSION) prep->next_buf_row,
+				       numrows);
    *in_row_ctr += numrows;
    prep->next_buf_row += numrows;
    prep->rows_to_go -= numrows;
    /* If at bottom of image, pad to fill the conversion buffer. */
    if (prep->rows_to_go == 0 &&
-        prep->next_buf_row < cinfo->max_v_samp_factor) {
+	prep->next_buf_row < cinfo->max_v_samp_factor) {
      for (ci = 0; ci < cinfo->num_components; ci++) {
-        expand_bottom_edge(prep->color_buf[ci], cinfo->image_width,
-                           prep->next_buf_row, cinfo->max_v_samp_factor);
+	expand_bottom_edge(prep->color_buf[ci], cinfo->image_width,
+			   prep->next_buf_row, cinfo->max_v_samp_factor);
      }
      prep->next_buf_row = cinfo->max_v_samp_factor;
    }
    /* If we've filled the conversion buffer, empty it. */
    if (prep->next_buf_row == cinfo->max_v_samp_factor) {
      (*cinfo->downsample->downsample) (cinfo,
-                                        prep->color_buf, (JDIMENSION) 0,
-                                        output_buf, *out_row_group_ctr);
+					prep->color_buf, (JDIMENSION) 0,
+					output_buf, *out_row_group_ctr);
      prep->next_buf_row = 0;
      (*out_row_group_ctr)++;
    }
@@ -170,15 +170,15 @@ pre_process_data (j_compress_ptr cinfo,
     * Note we assume the caller is providing a one-iMCU-height output buffer!
     */
    if (prep->rows_to_go == 0 &&
-        *out_row_group_ctr < out_row_groups_avail) {
+	*out_row_group_ctr < out_row_groups_avail) {
      for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-           ci++, compptr++) {
-        numrows = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) /
-                  cinfo->min_DCT_v_scaled_size;
-        expand_bottom_edge(output_buf[ci],
-                           compptr->width_in_blocks * compptr->DCT_h_scaled_size,
-                           (int) (*out_row_group_ctr * numrows),
-                           (int) (out_row_groups_avail * numrows));
+	   ci++, compptr++) {
+	numrows = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) /
+		  cinfo->min_DCT_v_scaled_size;
+	expand_bottom_edge(output_buf[ci],
+			   compptr->width_in_blocks * compptr->DCT_h_scaled_size,
+			   (int) (*out_row_group_ctr * numrows),
+			   (int) (out_row_groups_avail * numrows));
      }
      *out_row_group_ctr = out_row_groups_avail;
      break;			/* can exit outer loop without test */
@@ -195,10 +195,10 @@ pre_process_data (j_compress_ptr cinfo,

 METHODDEF(void)
 pre_process_context (j_compress_ptr cinfo,
-                     JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
-                     JDIMENSION in_rows_avail,
-                     JSAMPIMAGE output_buf, JDIMENSION *out_row_group_ctr,
-                     JDIMENSION out_row_groups_avail)
+		     JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
+		     JDIMENSION in_rows_avail,
+		     JSAMPIMAGE output_buf, JDIMENSION *out_row_group_ctr,
+		     JDIMENSION out_row_groups_avail)
 {
  my_prep_ptr prep = (my_prep_ptr) cinfo->prep;
  int numrows, ci;
@@ -212,19 +212,19 @@ pre_process_context (j_compress_ptr cinfo,
      numrows = prep->next_buf_stop - prep->next_buf_row;
      numrows = (int) MIN((JDIMENSION) numrows, inrows);
      (*cinfo->cconvert->color_convert) (cinfo, input_buf + *in_row_ctr,
-                                         prep->color_buf,
-                                         (JDIMENSION) prep->next_buf_row,
-                                         numrows);
+					 prep->color_buf,
+					 (JDIMENSION) prep->next_buf_row,
+					 numrows);
      /* Pad at top of image, if first time through */
      if (prep->rows_to_go == cinfo->image_height) {
-        for (ci = 0; ci < cinfo->num_components; ci++) {
-          int row;
-          for (row = 1; row <= cinfo->max_v_samp_factor; row++) {
-            jcopy_sample_rows(prep->color_buf[ci], 0,
-                              prep->color_buf[ci], -row,
-                              1, cinfo->image_width);
-          }
-        }
+	for (ci = 0; ci < cinfo->num_components; ci++) {
+	  int row;
+	  for (row = 1; row <= cinfo->max_v_samp_factor; row++) {
+	    jcopy_sample_rows(prep->color_buf[ci], 0,
+			      prep->color_buf[ci], -row,
+			      1, cinfo->image_width);
+	  }
+	}
      }
      *in_row_ctr += numrows;
      prep->next_buf_row += numrows;
@@ -232,29 +232,29 @@ pre_process_context (j_compress_ptr cinfo,
    } else {
      /* Return for more data, unless we are at the bottom of the image. */
      if (prep->rows_to_go != 0)
-        break;
+	break;
      /* When at bottom of image, pad to fill the conversion buffer. */
      if (prep->next_buf_row < prep->next_buf_stop) {
-        for (ci = 0; ci < cinfo->num_components; ci++) {
-          expand_bottom_edge(prep->color_buf[ci], cinfo->image_width,
-                             prep->next_buf_row, prep->next_buf_stop);
-        }
-        prep->next_buf_row = prep->next_buf_stop;
+	for (ci = 0; ci < cinfo->num_components; ci++) {
+	  expand_bottom_edge(prep->color_buf[ci], cinfo->image_width,
+			     prep->next_buf_row, prep->next_buf_stop);
+	}
+	prep->next_buf_row = prep->next_buf_stop;
      }
    }
    /* If we've gotten enough data, downsample a row group. */
    if (prep->next_buf_row == prep->next_buf_stop) {
      (*cinfo->downsample->downsample) (cinfo,
-                                        prep->color_buf,
-                                        (JDIMENSION) prep->this_row_group,
-                                        output_buf, *out_row_group_ctr);
+					prep->color_buf,
+					(JDIMENSION) prep->this_row_group,
+					output_buf, *out_row_group_ctr);
      (*out_row_group_ctr)++;
      /* Advance pointers with wraparound as necessary. */
      prep->this_row_group += cinfo->max_v_samp_factor;
      if (prep->this_row_group >= buf_height)
-        prep->this_row_group = 0;
+	prep->this_row_group = 0;
      if (prep->next_buf_row >= buf_height)
-        prep->next_buf_row = 0;
+	prep->next_buf_row = 0;
      prep->next_buf_stop = prep->next_buf_row + cinfo->max_v_samp_factor;
    }
  }
@@ -279,8 +279,8 @@ create_context_buffer (j_compress_ptr cinfo)
   */
  fake_buffer = (JSAMPARRAY)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (cinfo->num_components * 5 * rgroup_height) *
-                                SIZEOF(JSAMPROW));
+				(cinfo->num_components * 5 * rgroup_height) *
+				SIZEOF(JSAMPROW));

  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
       ci++, compptr++) {
@@ -291,12 +291,12 @@ create_context_buffer (j_compress_ptr cinfo)
    true_buffer = (*cinfo->mem->alloc_sarray)
      ((j_common_ptr) cinfo, JPOOL_IMAGE,
       (JDIMENSION) (((long) compptr->width_in_blocks *
-                      cinfo->min_DCT_h_scaled_size *
-                      cinfo->max_h_samp_factor) / compptr->h_samp_factor),
+		      cinfo->min_DCT_h_scaled_size *
+		      cinfo->max_h_samp_factor) / compptr->h_samp_factor),
       (JDIMENSION) (3 * rgroup_height));
    /* Copy true buffer row pointers into the middle of the fake row array */
    MEMCOPY(fake_buffer + rgroup_height, true_buffer,
-            3 * rgroup_height * SIZEOF(JSAMPROW));
+	    3 * rgroup_height * SIZEOF(JSAMPROW));
    /* Fill in the above and below wraparound pointers */
    for (i = 0; i < rgroup_height; i++) {
      fake_buffer[i] = true_buffer[2 * rgroup_height + i];
@@ -326,7 +326,7 @@ jinit_c_prep_controller (j_compress_ptr cinfo, boolean need_full_buffer)

  prep = (my_prep_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_prep_controller));
+				SIZEOF(my_prep_controller));
  cinfo->prep = (struct jpeg_c_prep_controller *) prep;
  prep->pub.start_pass = start_pass_prep;

@@ -346,13 +346,13 @@ jinit_c_prep_controller (j_compress_ptr cinfo, boolean need_full_buffer)
    /* No context, just make it tall enough for one row group */
    prep->pub.pre_process_data = pre_process_data;
    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      prep->color_buf[ci] = (*cinfo->mem->alloc_sarray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE,
-         (JDIMENSION) (((long) compptr->width_in_blocks *
-                        cinfo->min_DCT_h_scaled_size *
-                        cinfo->max_h_samp_factor) / compptr->h_samp_factor),
-         (JDIMENSION) cinfo->max_v_samp_factor);
+	((j_common_ptr) cinfo, JPOOL_IMAGE,
+	 (JDIMENSION) (((long) compptr->width_in_blocks *
+			cinfo->min_DCT_h_scaled_size *
+			cinfo->max_h_samp_factor) / compptr->h_samp_factor),
+	 (JDIMENSION) cinfo->max_v_samp_factor);
    }
  }
 }
@@ -52,8 +52,8 @@

 /* Pointer to routine to downsample a single component */
 typedef JMETHOD(void, downsample1_ptr,
-                (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                 JSAMPARRAY input_data, JSAMPARRAY output_data));
+		(j_compress_ptr cinfo, jpeg_component_info * compptr,
+		 JSAMPARRAY input_data, JSAMPARRAY output_data));

 /* Private subobject */

@@ -94,7 +94,7 @@ start_pass_downsample (j_compress_ptr cinfo)

 LOCAL(void)
 expand_right_edge (JSAMPARRAY image_data, int num_rows,
-                   JDIMENSION input_cols, JDIMENSION output_cols)
+		   JDIMENSION input_cols, JDIMENSION output_cols)
 {
  register JSAMPROW ptr;
  register JSAMPLE pixval;
@@ -107,7 +107,7 @@ expand_right_edge (JSAMPARRAY image_data, int num_rows,
      ptr = image_data[row] + input_cols;
      pixval = ptr[-1];		/* don't need GETJSAMPLE() here */
      for (count = numcols; count > 0; count--)
-        *ptr++ = pixval;
+	*ptr++ = pixval;
    }
  }
 }
@@ -121,8 +121,8 @@ expand_right_edge (JSAMPARRAY image_data, int num_rows,

 METHODDEF(void)
 sep_downsample (j_compress_ptr cinfo,
-                JSAMPIMAGE input_buf, JDIMENSION in_row_index,
-                JSAMPIMAGE output_buf, JDIMENSION out_row_group_index)
+		JSAMPIMAGE input_buf, JDIMENSION in_row_index,
+		JSAMPIMAGE output_buf, JDIMENSION out_row_group_index)
 {
  my_downsample_ptr downsample = (my_downsample_ptr) cinfo->downsample;
  int ci;
@@ -133,7 +133,7 @@ sep_downsample (j_compress_ptr cinfo,
       ci++, compptr++) {
    in_ptr = input_buf[ci] + in_row_index;
    out_ptr = output_buf[ci] +
-              (out_row_group_index * downsample->rowgroup_height[ci]);
+	      (out_row_group_index * downsample->rowgroup_height[ci]);
    (*downsample->methods[ci]) (cinfo, compptr, in_ptr, out_ptr);
  }
 }
@@ -148,7 +148,7 @@ sep_downsample (j_compress_ptr cinfo,

 METHODDEF(void)
 int_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                JSAMPARRAY input_data, JSAMPARRAY output_data)
+		JSAMPARRAY input_data, JSAMPARRAY output_data)
 {
  my_downsample_ptr downsample = (my_downsample_ptr) cinfo->downsample;
  int inrow, outrow, h_expand, v_expand, numpix, numpix2, h, v;
@@ -167,19 +167,19 @@ int_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
   * efficient.
   */
  expand_right_edge(input_data, cinfo->max_v_samp_factor,
-                    cinfo->image_width, output_cols * h_expand);
+		    cinfo->image_width, output_cols * h_expand);

  inrow = outrow = 0;
  while (inrow < cinfo->max_v_samp_factor) {
    outptr = output_data[outrow];
    for (outcol = 0, outcol_h = 0; outcol < output_cols;
-         outcol++, outcol_h += h_expand) {
+	 outcol++, outcol_h += h_expand) {
      outvalue = 0;
      for (v = 0; v < v_expand; v++) {
-        inptr = input_data[inrow+v] + outcol_h;
-        for (h = 0; h < h_expand; h++) {
-          outvalue += (INT32) GETJSAMPLE(*inptr++);
-        }
+	inptr = input_data[inrow+v] + outcol_h;
+	for (h = 0; h < h_expand; h++) {
+	  outvalue += (INT32) GETJSAMPLE(*inptr++);
+	}
      }
      *outptr++ = (JSAMPLE) ((outvalue + numpix2) / numpix);
    }
@@ -197,14 +197,14 @@ int_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 fullsize_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                     JSAMPARRAY input_data, JSAMPARRAY output_data)
+		     JSAMPARRAY input_data, JSAMPARRAY output_data)
 {
  /* Copy the data */
  jcopy_sample_rows(input_data, 0, output_data, 0,
-                    cinfo->max_v_samp_factor, cinfo->image_width);
+		    cinfo->max_v_samp_factor, cinfo->image_width);
  /* Edge-expand */
  expand_right_edge(output_data, cinfo->max_v_samp_factor, cinfo->image_width,
-                    compptr->width_in_blocks * compptr->DCT_h_scaled_size);
+		    compptr->width_in_blocks * compptr->DCT_h_scaled_size);
 }


@@ -222,7 +222,7 @@ fullsize_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 h2v1_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                 JSAMPARRAY input_data, JSAMPARRAY output_data)
+		 JSAMPARRAY input_data, JSAMPARRAY output_data)
 {
  int inrow;
  JDIMENSION outcol;
@@ -235,7 +235,7 @@ h2v1_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
   * efficient.
   */
  expand_right_edge(input_data, cinfo->max_v_samp_factor,
-                    cinfo->image_width, output_cols * 2);
+		    cinfo->image_width, output_cols * 2);

  for (inrow = 0; inrow < cinfo->max_v_samp_factor; inrow++) {
    outptr = output_data[inrow];
@@ -243,7 +243,7 @@ h2v1_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
    bias = 0;			/* bias = 0,1,0,1,... for successive samples */
    for (outcol = 0; outcol < output_cols; outcol++) {
      *outptr++ = (JSAMPLE) ((GETJSAMPLE(*inptr) + GETJSAMPLE(inptr[1])
-                              + bias) >> 1);
+			      + bias) >> 1);
      bias ^= 1;		/* 0=>1, 1=>0 */
      inptr += 2;
    }
@@ -259,7 +259,7 @@ h2v1_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 h2v2_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                 JSAMPARRAY input_data, JSAMPARRAY output_data)
+		 JSAMPARRAY input_data, JSAMPARRAY output_data)
 {
  int inrow, outrow;
  JDIMENSION outcol;
@@ -272,7 +272,7 @@ h2v2_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
   * efficient.
   */
  expand_right_edge(input_data, cinfo->max_v_samp_factor,
-                    cinfo->image_width, output_cols * 2);
+		    cinfo->image_width, output_cols * 2);

  inrow = outrow = 0;
  while (inrow < cinfo->max_v_samp_factor) {
@@ -282,8 +282,8 @@ h2v2_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
    bias = 1;			/* bias = 1,2,1,2,... for successive samples */
    for (outcol = 0; outcol < output_cols; outcol++) {
      *outptr++ = (JSAMPLE) ((GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) +
-                              GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1])
-                              + bias) >> 2);
+			      GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1])
+			      + bias) >> 2);
      bias ^= 3;		/* 1=>2, 2=>1 */
      inptr0 += 2; inptr1 += 2;
    }
@@ -303,7 +303,7 @@ h2v2_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                        JSAMPARRAY input_data, JSAMPARRAY output_data)
+			JSAMPARRAY input_data, JSAMPARRAY output_data)
 {
  int inrow, outrow;
  JDIMENSION colctr;
@@ -316,7 +316,7 @@ h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
   * efficient.
   */
  expand_right_edge(input_data - 1, cinfo->max_v_samp_factor + 2,
-                    cinfo->image_width, output_cols * 2);
+		    cinfo->image_width, output_cols * 2);

  /* We don't bother to form the individual "smoothed" input pixel values;
   * we can directly compute the output which is the average of the four
@@ -344,14 +344,14 @@ h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,

    /* Special case for first column: pretend column -1 is same as column 0 */
    membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) +
-                GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]);
+		GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]);
    neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) +
-               GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) +
-               GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[2]) +
-               GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[2]);
+	       GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) +
+	       GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[2]) +
+	       GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[2]);
    neighsum += neighsum;
    neighsum += GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[2]) +
-                GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[2]);
+		GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[2]);
    membersum = membersum * memberscale + neighsum * neighscale;
    *outptr++ = (JSAMPLE) ((membersum + 32768) >> 16);
    inptr0 += 2; inptr1 += 2; above_ptr += 2; below_ptr += 2;
@@ -359,17 +359,17 @@ h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,
    for (colctr = output_cols - 2; colctr > 0; colctr--) {
      /* sum of pixels directly mapped to this output element */
      membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) +
-                  GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]);
+		  GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]);
      /* sum of edge-neighbor pixels */
      neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) +
-                 GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) +
-                 GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[2]) +
-                 GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[2]);
+		 GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) +
+		 GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[2]) +
+		 GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[2]);
      /* The edge-neighbors count twice as much as corner-neighbors */
      neighsum += neighsum;
      /* Add in the corner-neighbors */
      neighsum += GETJSAMPLE(above_ptr[-1]) + GETJSAMPLE(above_ptr[2]) +
-                  GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[2]);
+		  GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[2]);
      /* form final output scaled up by 2^16 */
      membersum = membersum * memberscale + neighsum * neighscale;
      /* round, descale and output it */
@@ -379,14 +379,14 @@ h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,

    /* Special case for last column */
    membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) +
-                GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]);
+		GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]);
    neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) +
-               GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) +
-               GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[1]) +
-               GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[1]);
+	       GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) +
+	       GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[1]) +
+	       GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[1]);
    neighsum += neighsum;
    neighsum += GETJSAMPLE(above_ptr[-1]) + GETJSAMPLE(above_ptr[1]) +
-                GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[1]);
+		GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[1]);
    membersum = membersum * memberscale + neighsum * neighscale;
    *outptr = (JSAMPLE) ((membersum + 32768) >> 16);

@@ -404,7 +404,7 @@ h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 fullsize_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info *compptr,
-                            JSAMPARRAY input_data, JSAMPARRAY output_data)
+			    JSAMPARRAY input_data, JSAMPARRAY output_data)
 {
  int inrow;
  JDIMENSION colctr;
@@ -418,7 +418,7 @@ fullsize_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info *compptr,
   * efficient.
   */
  expand_right_edge(input_data - 1, cinfo->max_v_samp_factor + 2,
-                    cinfo->image_width, output_cols);
+		    cinfo->image_width, output_cols);

  /* Each of the eight neighbor pixels contributes a fraction SF to the
   * smoothed pixel, while the main pixel contributes (1-8*SF).  In order
@@ -437,10 +437,10 @@ fullsize_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info *compptr,

    /* Special case for first column */
    colsum = GETJSAMPLE(*above_ptr++) + GETJSAMPLE(*below_ptr++) +
-             GETJSAMPLE(*inptr);
+	     GETJSAMPLE(*inptr);
    membersum = GETJSAMPLE(*inptr++);
    nextcolsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(*below_ptr) +
-                 GETJSAMPLE(*inptr);
+		 GETJSAMPLE(*inptr);
    neighsum = colsum + (colsum - membersum) + nextcolsum;
    membersum = membersum * memberscale + neighsum * neighscale;
    *outptr++ = (JSAMPLE) ((membersum + 32768) >> 16);
@@ -450,7 +450,7 @@ fullsize_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info *compptr,
      membersum = GETJSAMPLE(*inptr++);
      above_ptr++; below_ptr++;
      nextcolsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(*below_ptr) +
-                   GETJSAMPLE(*inptr);
+		   GETJSAMPLE(*inptr);
      neighsum = lastcolsum + (colsum - membersum) + nextcolsum;
      membersum = membersum * memberscale + neighsum * neighscale;
      *outptr++ = (JSAMPLE) ((membersum + 32768) >> 16);
@@ -485,7 +485,7 @@ jinit_downsampler (j_compress_ptr cinfo)

  downsample = (my_downsample_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_downsampler));
+				SIZEOF(my_downsampler));
  cinfo->downsample = (struct jpeg_downsampler *) downsample;
  downsample->pub.start_pass = start_pass_downsample;
  downsample->pub.downsample = sep_downsample;
@@ -501,35 +501,35 @@ jinit_downsampler (j_compress_ptr cinfo)
     * are to be converted from max_h_samp_factor * max_v_samp_factor pixels.
     */
    h_out_group = (compptr->h_samp_factor * compptr->DCT_h_scaled_size) /
-                  cinfo->min_DCT_h_scaled_size;
+		  cinfo->min_DCT_h_scaled_size;
    v_out_group = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) /
-                  cinfo->min_DCT_v_scaled_size;
+		  cinfo->min_DCT_v_scaled_size;
    h_in_group = cinfo->max_h_samp_factor;
    v_in_group = cinfo->max_v_samp_factor;
    downsample->rowgroup_height[ci] = v_out_group; /* save for use later */
    if (h_in_group == h_out_group && v_in_group == v_out_group) {
 #ifdef INPUT_SMOOTHING_SUPPORTED
      if (cinfo->smoothing_factor) {
-        downsample->methods[ci] = fullsize_smooth_downsample;
-        downsample->pub.need_context_rows = TRUE;
+	downsample->methods[ci] = fullsize_smooth_downsample;
+	downsample->pub.need_context_rows = TRUE;
      } else
 #endif
-        downsample->methods[ci] = fullsize_downsample;
+	downsample->methods[ci] = fullsize_downsample;
    } else if (h_in_group == h_out_group * 2 &&
-               v_in_group == v_out_group) {
+	       v_in_group == v_out_group) {
      smoothok = FALSE;
      downsample->methods[ci] = h2v1_downsample;
    } else if (h_in_group == h_out_group * 2 &&
-               v_in_group == v_out_group * 2) {
+	       v_in_group == v_out_group * 2) {
 #ifdef INPUT_SMOOTHING_SUPPORTED
      if (cinfo->smoothing_factor) {
-        downsample->methods[ci] = h2v2_smooth_downsample;
-        downsample->pub.need_context_rows = TRUE;
+	downsample->methods[ci] = h2v2_smooth_downsample;
+	downsample->pub.need_context_rows = TRUE;
      } else
 #endif
-        downsample->methods[ci] = h2v2_downsample;
+	downsample->methods[ci] = h2v2_downsample;
    } else if ((h_in_group % h_out_group) == 0 &&
-               (v_in_group % v_out_group) == 0) {
+	       (v_in_group % v_out_group) == 0) {
      smoothok = FALSE;
      downsample->methods[ci] = int_downsample;
      downsample->h_expand[ci] = (UINT8) (h_in_group / h_out_group);
@@ -2,7 +2,7 @@
 * jctrans.c
 *
 * Copyright (C) 1995-1998, Thomas G. Lane.
- * Modified 2000-2012 by Guido Vollbeding.
+ * Modified 2000-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -18,9 +18,9 @@

 /* Forward declarations */
 LOCAL(void) transencode_master_selection
-        JPP((j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays));
+	JPP((j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays));
 LOCAL(void) transencode_coef_controller
-        JPP((j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays));
+	JPP((j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays));


 /*
@@ -62,7 +62,7 @@ jpeg_write_coefficients (j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays)

 GLOBAL(void)
 jpeg_copy_critical_parameters (j_decompress_ptr srcinfo,
-                               j_compress_ptr dstinfo)
+			       j_compress_ptr dstinfo)
 {
  JQUANT_TBL ** qtblptr;
  jpeg_component_info *incomp, *outcomp;
@@ -97,10 +97,10 @@ jpeg_copy_critical_parameters (j_decompress_ptr srcinfo,
    if (srcinfo->quant_tbl_ptrs[tblno] != NULL) {
      qtblptr = & dstinfo->quant_tbl_ptrs[tblno];
      if (*qtblptr == NULL)
-        *qtblptr = jpeg_alloc_quant_table((j_common_ptr) dstinfo);
+	*qtblptr = jpeg_alloc_quant_table((j_common_ptr) dstinfo);
      MEMCOPY((*qtblptr)->quantval,
-              srcinfo->quant_tbl_ptrs[tblno]->quantval,
-              SIZEOF((*qtblptr)->quantval));
+	      srcinfo->quant_tbl_ptrs[tblno]->quantval,
+	      SIZEOF((*qtblptr)->quantval));
      (*qtblptr)->sent_table = FALSE;
    }
  }
@@ -110,7 +110,7 @@ jpeg_copy_critical_parameters (j_decompress_ptr srcinfo,
  dstinfo->num_components = srcinfo->num_components;
  if (dstinfo->num_components < 1 || dstinfo->num_components > MAX_COMPONENTS)
    ERREXIT2(dstinfo, JERR_COMPONENT_COUNT, dstinfo->num_components,
-             MAX_COMPONENTS);
+	     MAX_COMPONENTS);
  for (ci = 0, incomp = srcinfo->comp_info, outcomp = dstinfo->comp_info;
       ci < dstinfo->num_components; ci++, incomp++, outcomp++) {
    outcomp->component_id = incomp->component_id;
@@ -123,14 +123,14 @@ jpeg_copy_critical_parameters (j_decompress_ptr srcinfo,
     */
    tblno = outcomp->quant_tbl_no;
    if (tblno < 0 || tblno >= NUM_QUANT_TBLS ||
-        srcinfo->quant_tbl_ptrs[tblno] == NULL)
+	srcinfo->quant_tbl_ptrs[tblno] == NULL)
      ERREXIT1(dstinfo, JERR_NO_QUANT_TABLE, tblno);
    slot_quant = srcinfo->quant_tbl_ptrs[tblno];
    c_quant = incomp->quant_table;
    if (c_quant != NULL) {
      for (coefi = 0; coefi < DCTSIZE2; coefi++) {
-        if (c_quant->quantval[coefi] != slot_quant->quantval[coefi])
-          ERREXIT1(dstinfo, JERR_MISMATCHED_QUANT_TABLE, tblno);
+	if (c_quant->quantval[coefi] != slot_quant->quantval[coefi])
+	  ERREXIT1(dstinfo, JERR_MISMATCHED_QUANT_TABLE, tblno);
      }
    }
    /* Note: we do not copy the source's entropy table assignments;
@@ -143,10 +143,10 @@ jpeg_copy_critical_parameters (j_decompress_ptr srcinfo,
   * if the application chooses to copy JFIF 1.02 extension markers from
   * the source file, we need to copy the version to make sure we don't
   * emit a file that has 1.02 extensions but a claimed version of 1.01.
-   * We will *not*, however, copy version info from mislabeled "2.01" files.
   */
  if (srcinfo->saw_JFIF_marker) {
-    if (srcinfo->JFIF_major_version == 1) {
+    if (srcinfo->JFIF_major_version == 1 ||
+	srcinfo->JFIF_major_version == 2) {
      dstinfo->JFIF_major_version = srcinfo->JFIF_major_version;
      dstinfo->JFIF_minor_version = srcinfo->JFIF_minor_version;
    }
@@ -164,7 +164,7 @@ jpeg_copy_critical_parameters (j_decompress_ptr srcinfo,

 LOCAL(void)
 transencode_master_selection (j_compress_ptr cinfo,
-                              jvirt_barray_ptr * coef_arrays)
+			      jvirt_barray_ptr * coef_arrays)
 {
  /* Initialize master control (includes parameter checking/processing) */
  jinit_c_master_control(cinfo, TRUE /* transcode only */);
@@ -298,44 +298,44 @@ compress_output (j_compress_ptr cinfo, JSAMPIMAGE input_buf)
  for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row;
       yoffset++) {
    for (MCU_col_num = coef->mcu_ctr; MCU_col_num < cinfo->MCUs_per_row;
-         MCU_col_num++) {
+	 MCU_col_num++) {
      /* Construct list of pointers to DCT blocks belonging to this MCU */
      blkn = 0;			/* index of current DCT block within MCU */
      for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
-        compptr = cinfo->cur_comp_info[ci];
-        start_col = MCU_col_num * compptr->MCU_width;
-        blockcnt = (MCU_col_num < last_MCU_col) ? compptr->MCU_width
-                                                : compptr->last_col_width;
-        for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
-          if (coef->iMCU_row_num < last_iMCU_row ||
-              yindex+yoffset < compptr->last_row_height) {
-            /* Fill in pointers to real blocks in this row */
-            buffer_ptr = buffer[ci][yindex+yoffset] + start_col;
-            for (xindex = 0; xindex < blockcnt; xindex++)
-              MCU_buffer[blkn++] = buffer_ptr++;
-          } else {
-            /* At bottom of image, need a whole row of dummy blocks */
-            xindex = 0;
-          }
-          /* Fill in any dummy blocks needed in this row.
-           * Dummy blocks are filled in the same way as in jccoefct.c:
-           * all zeroes in the AC entries, DC entries equal to previous
-           * block's DC value.  The init routine has already zeroed the
-           * AC entries, so we need only set the DC entries correctly.
-           */
-          for (; xindex < compptr->MCU_width; xindex++) {
-            MCU_buffer[blkn] = coef->dummy_buffer[blkn];
-            MCU_buffer[blkn][0][0] = MCU_buffer[blkn-1][0][0];
-            blkn++;
-          }
-        }
+	compptr = cinfo->cur_comp_info[ci];
+	start_col = MCU_col_num * compptr->MCU_width;
+	blockcnt = (MCU_col_num < last_MCU_col) ? compptr->MCU_width
+						: compptr->last_col_width;
+	for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
+	  if (coef->iMCU_row_num < last_iMCU_row ||
+	      yindex+yoffset < compptr->last_row_height) {
+	    /* Fill in pointers to real blocks in this row */
+	    buffer_ptr = buffer[ci][yindex+yoffset] + start_col;
+	    for (xindex = 0; xindex < blockcnt; xindex++)
+	      MCU_buffer[blkn++] = buffer_ptr++;
+	  } else {
+	    /* At bottom of image, need a whole row of dummy blocks */
+	    xindex = 0;
+	  }
+	  /* Fill in any dummy blocks needed in this row.
+	   * Dummy blocks are filled in the same way as in jccoefct.c:
+	   * all zeroes in the AC entries, DC entries equal to previous
+	   * block's DC value.  The init routine has already zeroed the
+	   * AC entries, so we need only set the DC entries correctly.
+	   */
+	  for (; xindex < compptr->MCU_width; xindex++) {
+	    MCU_buffer[blkn] = coef->dummy_buffer[blkn];
+	    MCU_buffer[blkn][0][0] = MCU_buffer[blkn-1][0][0];
+	    blkn++;
+	  }
+	}
      }
      /* Try to write the MCU. */
      if (! (*cinfo->entropy->encode_mcu) (cinfo, MCU_buffer)) {
-        /* Suspension forced; update state counters and exit */
-        coef->MCU_vert_offset = yoffset;
-        coef->mcu_ctr = MCU_col_num;
-        return FALSE;
+	/* Suspension forced; update state counters and exit */
+	coef->MCU_vert_offset = yoffset;
+	coef->mcu_ctr = MCU_col_num;
+	return FALSE;
      }
    }
    /* Completed an MCU row, but perhaps not an iMCU row */
@@ -358,7 +358,7 @@ compress_output (j_compress_ptr cinfo, JSAMPIMAGE input_buf)

 LOCAL(void)
 transencode_coef_controller (j_compress_ptr cinfo,
-                             jvirt_barray_ptr * coef_arrays)
+			     jvirt_barray_ptr * coef_arrays)
 {
  my_coef_ptr coef;
  JBLOCKROW buffer;
@@ -366,7 +366,7 @@ transencode_coef_controller (j_compress_ptr cinfo,

  coef = (my_coef_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_coef_controller));
+				SIZEOF(my_coef_controller));
  cinfo->coef = &coef->pub;
  coef->pub.start_pass = start_pass_coef;
  coef->pub.compress_data = compress_output;
@@ -377,7 +377,7 @@ transencode_coef_controller (j_compress_ptr cinfo,
  /* Allocate and pre-zero space for dummy DCT blocks. */
  buffer = (JBLOCKROW)
    (*cinfo->mem->alloc_large) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                C_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK));
+				C_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK));
  FMEMZERO((void FAR *) buffer, C_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK));
  for (i = 0; i < C_MAX_BLOCKS_IN_MCU; i++) {
    coef->dummy_buffer[i] = buffer + i;
@@ -2,7 +2,7 @@
 * jdapimin.c
 *
 * Copyright (C) 1994-1998, Thomas G. Lane.
- * Modified 2009 by Guido Vollbeding.
+ * Modified 2009-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -37,8 +37,8 @@ jpeg_CreateDecompress (j_decompress_ptr cinfo, int version, size_t structsize)
  if (version != JPEG_LIB_VERSION)
    ERREXIT2(cinfo, JERR_BAD_LIB_VERSION, JPEG_LIB_VERSION, version);
  if (structsize != SIZEOF(struct jpeg_decompress_struct))
-    ERREXIT2(cinfo, JERR_BAD_STRUCT_SIZE,
-             (int) SIZEOF(struct jpeg_decompress_struct), (int) structsize);
+    ERREXIT2(cinfo, JERR_BAD_STRUCT_SIZE, 
+	     (int) SIZEOF(struct jpeg_decompress_struct), (int) structsize);

  /* For debugging purposes, we zero the whole master structure.
   * But the application has already set the err pointer, and may have set
@@ -114,63 +114,66 @@ jpeg_abort_decompress (j_decompress_ptr cinfo)
 LOCAL(void)
 default_decompress_parms (j_decompress_ptr cinfo)
 {
+  int cid0, cid1, cid2;
+
  /* Guess the input colorspace, and set output colorspace accordingly. */
-  /* (Wish JPEG committee had provided a real way to specify this...) */
  /* Note application may override our guesses. */
  switch (cinfo->num_components) {
  case 1:
    cinfo->jpeg_color_space = JCS_GRAYSCALE;
    cinfo->out_color_space = JCS_GRAYSCALE;
    break;
-
+    
  case 3:
-    if (cinfo->saw_JFIF_marker) {
-      cinfo->jpeg_color_space = JCS_YCbCr; /* JFIF implies YCbCr */
-    } else if (cinfo->saw_Adobe_marker) {
+    cid0 = cinfo->comp_info[0].component_id;
+    cid1 = cinfo->comp_info[1].component_id;
+    cid2 = cinfo->comp_info[2].component_id;
+
+    /* First try to guess from the component IDs */
+    if      (cid0 == 0x01 && cid1 == 0x02 && cid2 == 0x03)
+      cinfo->jpeg_color_space = JCS_YCbCr;
+    else if (cid0 == 0x01 && cid1 == 0x22 && cid2 == 0x23)
+      cinfo->jpeg_color_space = JCS_BG_YCC;
+    else if (cid0 == 0x52 && cid1 == 0x47 && cid2 == 0x42)
+      cinfo->jpeg_color_space = JCS_RGB;	/* ASCII 'R', 'G', 'B' */
+    else if (cid0 == 0x72 && cid1 == 0x67 && cid2 == 0x62)
+      cinfo->jpeg_color_space = JCS_BG_RGB;	/* ASCII 'r', 'g', 'b' */
+    else if (cinfo->saw_JFIF_marker)
+      cinfo->jpeg_color_space = JCS_YCbCr;	/* assume it's YCbCr */
+    else if (cinfo->saw_Adobe_marker) {
      switch (cinfo->Adobe_transform) {
      case 0:
-        cinfo->jpeg_color_space = JCS_RGB;
-        break;
+	cinfo->jpeg_color_space = JCS_RGB;
+	break;
      case 1:
-        cinfo->jpeg_color_space = JCS_YCbCr;
-        break;
+	cinfo->jpeg_color_space = JCS_YCbCr;
+	break;
      default:
-        WARNMS1(cinfo, JWRN_ADOBE_XFORM, cinfo->Adobe_transform);
-        cinfo->jpeg_color_space = JCS_YCbCr; /* assume it's YCbCr */
-        break;
+	WARNMS1(cinfo, JWRN_ADOBE_XFORM, cinfo->Adobe_transform);
+	cinfo->jpeg_color_space = JCS_YCbCr;	/* assume it's YCbCr */
+	break;
      }
    } else {
-      /* Saw no special markers, try to guess from the component IDs */
-      int cid0 = cinfo->comp_info[0].component_id;
-      int cid1 = cinfo->comp_info[1].component_id;
-      int cid2 = cinfo->comp_info[2].component_id;
-
-      if (cid0 == 1 && cid1 == 2 && cid2 == 3)
-        cinfo->jpeg_color_space = JCS_YCbCr; /* assume JFIF w/out marker */
-      else if (cid0 == 82 && cid1 == 71 && cid2 == 66)
-        cinfo->jpeg_color_space = JCS_RGB; /* ASCII 'R', 'G', 'B' */
-      else {
-        TRACEMS3(cinfo, 1, JTRC_UNKNOWN_IDS, cid0, cid1, cid2);
-        cinfo->jpeg_color_space = JCS_YCbCr; /* assume it's YCbCr */
-      }
+      TRACEMS3(cinfo, 1, JTRC_UNKNOWN_IDS, cid0, cid1, cid2);
+      cinfo->jpeg_color_space = JCS_YCbCr;	/* assume it's YCbCr */
    }
    /* Always guess RGB is proper output colorspace. */
    cinfo->out_color_space = JCS_RGB;
    break;
-
+    
  case 4:
    if (cinfo->saw_Adobe_marker) {
      switch (cinfo->Adobe_transform) {
      case 0:
-        cinfo->jpeg_color_space = JCS_CMYK;
-        break;
+	cinfo->jpeg_color_space = JCS_CMYK;
+	break;
      case 2:
-        cinfo->jpeg_color_space = JCS_YCCK;
-        break;
+	cinfo->jpeg_color_space = JCS_YCCK;
+	break;
      default:
-        WARNMS1(cinfo, JWRN_ADOBE_XFORM, cinfo->Adobe_transform);
-        cinfo->jpeg_color_space = JCS_YCCK; /* assume it's YCCK */
-        break;
+	WARNMS1(cinfo, JWRN_ADOBE_XFORM, cinfo->Adobe_transform);
+	cinfo->jpeg_color_space = JCS_YCCK;	/* assume it's YCCK */
+	break;
      }
    } else {
      /* No special markers, assume straight CMYK. */
@@ -178,7 +181,7 @@ default_decompress_parms (j_decompress_ptr cinfo)
    }
    cinfo->out_color_space = JCS_CMYK;
    break;
-
+    
  default:
    cinfo->jpeg_color_space = JCS_UNKNOWN;
    cinfo->out_color_space = JCS_UNKNOWN;
@@ -2,6 +2,7 @@
 * jdapistd.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
+ * Modified 2002-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -52,24 +53,24 @@ jpeg_start_decompress (j_decompress_ptr cinfo)
    if (cinfo->inputctl->has_multiple_scans) {
 #ifdef D_MULTISCAN_FILES_SUPPORTED
      for (;;) {
-        int retcode;
-        /* Call progress monitor hook if present */
-        if (cinfo->progress != NULL)
-          (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
-        /* Absorb some more input */
-        retcode = (*cinfo->inputctl->consume_input) (cinfo);
-        if (retcode == JPEG_SUSPENDED)
-          return FALSE;
-        if (retcode == JPEG_REACHED_EOI)
-          break;
-        /* Advance progress counter if appropriate */
-        if (cinfo->progress != NULL &&
-            (retcode == JPEG_ROW_COMPLETED || retcode == JPEG_REACHED_SOS)) {
-          if (++cinfo->progress->pass_counter >= cinfo->progress->pass_limit) {
-            /* jdmaster underestimated number of scans; ratchet up one scan */
-            cinfo->progress->pass_limit += (long) cinfo->total_iMCU_rows;
-          }
-        }
+	int retcode;
+	/* Call progress monitor hook if present */
+	if (cinfo->progress != NULL)
+	  (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
+	/* Absorb some more input */
+	retcode = (*cinfo->inputctl->consume_input) (cinfo);
+	if (retcode == JPEG_SUSPENDED)
+	  return FALSE;
+	if (retcode == JPEG_REACHED_EOI)
+	  break;
+	/* Advance progress counter if appropriate */
+	if (cinfo->progress != NULL &&
+	    (retcode == JPEG_ROW_COMPLETED || retcode == JPEG_REACHED_SOS)) {
+	  if (++cinfo->progress->pass_counter >= cinfo->progress->pass_limit) {
+	    /* jdmaster underestimated number of scans; ratchet up one scan */
+	    cinfo->progress->pass_limit += (long) cinfo->total_iMCU_rows;
+	  }
+	}
      }
 #else
      ERREXIT(cinfo, JERR_NOT_COMPILED);
@@ -108,16 +109,16 @@ output_pass_setup (j_decompress_ptr cinfo)
      JDIMENSION last_scanline;
      /* Call progress monitor hook if present */
      if (cinfo->progress != NULL) {
-        cinfo->progress->pass_counter = (long) cinfo->output_scanline;
-        cinfo->progress->pass_limit = (long) cinfo->output_height;
-        (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
+	cinfo->progress->pass_counter = (long) cinfo->output_scanline;
+	cinfo->progress->pass_limit = (long) cinfo->output_height;
+	(*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
      }
      /* Process some data */
      last_scanline = cinfo->output_scanline;
      (*cinfo->main->process_data) (cinfo, (JSAMPARRAY) NULL,
-                                    &cinfo->output_scanline, (JDIMENSION) 0);
+				    &cinfo->output_scanline, (JDIMENSION) 0);
      if (cinfo->output_scanline == last_scanline)
-        return FALSE;		/* No progress made, must suspend */
+	return FALSE;		/* No progress made, must suspend */
    }
    /* Finish up dummy pass, and set up for another one */
    (*cinfo->master->finish_output_pass) (cinfo);
@@ -150,7 +151,7 @@ output_pass_setup (j_decompress_ptr cinfo)

 GLOBAL(JDIMENSION)
 jpeg_read_scanlines (j_decompress_ptr cinfo, JSAMPARRAY scanlines,
-                     JDIMENSION max_lines)
+		     JDIMENSION max_lines)
 {
  JDIMENSION row_ctr;

@@ -183,7 +184,7 @@ jpeg_read_scanlines (j_decompress_ptr cinfo, JSAMPARRAY scanlines,

 GLOBAL(JDIMENSION)
 jpeg_read_raw_data (j_decompress_ptr cinfo, JSAMPIMAGE data,
-                    JDIMENSION max_lines)
+		    JDIMENSION max_lines)
 {
  JDIMENSION lines_per_iMCU_row;

@@ -264,7 +265,7 @@ jpeg_finish_output (j_decompress_ptr cinfo)
  }
  /* Read markers looking for SOS or EOI */
  while (cinfo->input_scan_number <= cinfo->output_scan_number &&
-         ! cinfo->inputctl->eoi_reached) {
+	 ! cinfo->inputctl->eoi_reached) {
    if ((*cinfo->inputctl->consume_input) (cinfo) == JPEG_SUSPENDED)
      return FALSE;		/* Suspend, come back later */
  }
@@ -1,7 +1,7 @@
 /*
 * jdarith.c
 *
- * Developed 1997-2012 by Guido Vollbeding.
+ * Developed 1997-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -94,7 +94,7 @@ get_byte (j_decompress_ptr cinfo)
 * (instead of fixed) with the bit shift counter CT.
 * Thus, we also need only one (variable instead of
 * fixed size) shift for the LPS/MPS decision, and
- * we can get away with any renormalization update
+ * we can do away with any renormalization update
 * of C (except for new data insertion, of course).
 *
 * I've also introduced a new scheme for accessing
@@ -115,32 +115,32 @@ arith_decode (j_decompress_ptr cinfo, unsigned char *st)
    if (--e->ct < 0) {
      /* Need to fetch next data byte */
      if (cinfo->unread_marker)
-        data = 0;		/* stuff zero data */
+	data = 0;		/* stuff zero data */
      else {
-        data = get_byte(cinfo);	/* read next input byte */
-        if (data == 0xFF) {	/* zero stuff or marker code */
-          do data = get_byte(cinfo);
-          while (data == 0xFF);	/* swallow extra 0xFF bytes */
-          if (data == 0)
-            data = 0xFF;	/* discard stuffed zero byte */
-          else {
-            /* Note: Different from the Huffman decoder, hitting
-             * a marker while processing the compressed data
-             * segment is legal in arithmetic coding.
-             * The convention is to supply zero data
-             * then until decoding is complete.
-             */
-            cinfo->unread_marker = data;
-            data = 0;
-          }
-        }
+	data = get_byte(cinfo);	/* read next input byte */
+	if (data == 0xFF) {	/* zero stuff or marker code */
+	  do data = get_byte(cinfo);
+	  while (data == 0xFF);	/* swallow extra 0xFF bytes */
+	  if (data == 0)
+	    data = 0xFF;	/* discard stuffed zero byte */
+	  else {
+	    /* Note: Different from the Huffman decoder, hitting
+	     * a marker while processing the compressed data
+	     * segment is legal in arithmetic coding.
+	     * The convention is to supply zero data
+	     * then until decoding is complete.
+	     */
+	    cinfo->unread_marker = data;
+	    data = 0;
+	  }
+	}
      }
      e->c = (e->c << 8) | data; /* insert data into C register */
      if ((e->ct += 8) < 0)	 /* update bit shift counter */
-        /* Need more initial bytes */
-        if (++e->ct == 0)
-          /* Got 2 initial bytes -> re-init A and exit loop */
-          e->a = 0x8000L; /* => e->a = 0x10000L after loop exit */
+	/* Need more initial bytes */
+	if (++e->ct == 0)
+	  /* Got 2 initial bytes -> re-init A and exit loop */
+	  e->a = 0x8000L; /* => e->a = 0x10000L after loop exit */
    }
    e->a <<= 1;
  }
@@ -207,7 +207,7 @@ process_restart (j_decompress_ptr cinfo)
      entropy->dc_context[ci] = 0;
    }
    if ((! cinfo->progressive_mode && cinfo->lim_Se) ||
-        (cinfo->progressive_mode && cinfo->Ss)) {
+	(cinfo->progressive_mode && cinfo->Ss)) {
      MEMZERO(entropy->ac_stats[compptr->ac_tbl_no], AC_STAT_BINS);
    }
  }
@@ -278,28 +278,28 @@ decode_mcu_DC_first (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)
      st += 2; st += sign;
      /* Figure F.23: Decoding the magnitude category of v */
      if ((m = arith_decode(cinfo, st)) != 0) {
-        st = entropy->dc_stats[tbl] + 20;	/* Table F.4: X1 = 20 */
-        while (arith_decode(cinfo, st)) {
-          if ((m <<= 1) == 0x8000) {
-            WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
-            entropy->ct = -1;			/* magnitude overflow */
-            return TRUE;
-          }
-          st += 1;
-        }
+	st = entropy->dc_stats[tbl] + 20;	/* Table F.4: X1 = 20 */
+	while (arith_decode(cinfo, st)) {
+	  if ((m <<= 1) == 0x8000) {
+	    WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
+	    entropy->ct = -1;			/* magnitude overflow */
+	    return TRUE;
+	  }
+	  st += 1;
+	}
      }
      /* Section F.1.4.4.1.2: Establish dc_context conditioning category */
      if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1))
-        entropy->dc_context[ci] = 0;		   /* zero diff category */
+	entropy->dc_context[ci] = 0;		   /* zero diff category */
      else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1))
-        entropy->dc_context[ci] = 12 + (sign * 4); /* large diff category */
+	entropy->dc_context[ci] = 12 + (sign * 4); /* large diff category */
      else
-        entropy->dc_context[ci] = 4 + (sign * 4);  /* small diff category */
+	entropy->dc_context[ci] = 4 + (sign * 4);  /* small diff category */
      v = m;
      /* Figure F.24: Decoding the magnitude bit pattern of v */
      st += 14;
      while (m >>= 1)
-        if (arith_decode(cinfo, st)) v |= m;
+	if (arith_decode(cinfo, st)) v |= m;
      v += 1; if (sign) v = -v;
      entropy->last_dc_val[ci] += v;
    }
@@ -354,9 +354,9 @@ decode_mcu_AC_first (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)
      if (arith_decode(cinfo, st + 1)) break;
      st += 3;
      if (k >= cinfo->Se) {
-        WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
-        entropy->ct = -1;			/* spectral overflow */
-        return TRUE;
+	WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
+	entropy->ct = -1;			/* spectral overflow */
+	return TRUE;
      }
    }
    /* Figure F.21: Decoding nonzero value v */
@@ -366,17 +366,17 @@ decode_mcu_AC_first (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)
    /* Figure F.23: Decoding the magnitude category of v */
    if ((m = arith_decode(cinfo, st)) != 0) {
      if (arith_decode(cinfo, st)) {
-        m <<= 1;
-        st = entropy->ac_stats[tbl] +
-             (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
-        while (arith_decode(cinfo, st)) {
-          if ((m <<= 1) == 0x8000) {
-            WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
-            entropy->ct = -1;			/* magnitude overflow */
-            return TRUE;
-          }
-          st += 1;
-        }
+	m <<= 1;
+	st = entropy->ac_stats[tbl] +
+	     (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
+	while (arith_decode(cinfo, st)) {
+	  if ((m <<= 1) == 0x8000) {
+	    WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
+	    entropy->ct = -1;			/* magnitude overflow */
+	    return TRUE;
+	  }
+	  st += 1;
+	}
      }
    }
    v = m;
@@ -395,6 +395,8 @@ decode_mcu_AC_first (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)

 /*
 * MCU decoding for DC successive approximation refinement scan.
+ * Note: we assume such scans can be multi-component,
+ * although the spec is not very clear on the point.
 */

 METHODDEF(boolean)
@@ -473,26 +475,26 @@ decode_mcu_AC_refine (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)
    for (;;) {
      thiscoef = *block + natural_order[++k];
      if (*thiscoef) {				/* previously nonzero coef */
-        if (arith_decode(cinfo, st + 2)) {
-          if (*thiscoef < 0)
-            *thiscoef += m1;
-          else
-            *thiscoef += p1;
-        }
-        break;
+	if (arith_decode(cinfo, st + 2)) {
+	  if (*thiscoef < 0)
+	    *thiscoef += m1;
+	  else
+	    *thiscoef += p1;
+	}
+	break;
      }
      if (arith_decode(cinfo, st + 1)) {	/* newly nonzero coef */
-        if (arith_decode(cinfo, entropy->fixed_bin))
-          *thiscoef = m1;
-        else
-          *thiscoef = p1;
-        break;
+	if (arith_decode(cinfo, entropy->fixed_bin))
+	  *thiscoef = m1;
+	else
+	  *thiscoef = p1;
+	break;
      }
      st += 3;
      if (k >= cinfo->Se) {
-        WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
-        entropy->ct = -1;			/* spectral overflow */
-        return TRUE;
+	WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
+	entropy->ct = -1;			/* spectral overflow */
+	return TRUE;
      }
    }
  } while (k < cinfo->Se);
@@ -551,28 +553,28 @@ decode_mcu (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)
      st += 2; st += sign;
      /* Figure F.23: Decoding the magnitude category of v */
      if ((m = arith_decode(cinfo, st)) != 0) {
-        st = entropy->dc_stats[tbl] + 20;	/* Table F.4: X1 = 20 */
-        while (arith_decode(cinfo, st)) {
-          if ((m <<= 1) == 0x8000) {
-            WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
-            entropy->ct = -1;			/* magnitude overflow */
-            return TRUE;
-          }
-          st += 1;
-        }
+	st = entropy->dc_stats[tbl] + 20;	/* Table F.4: X1 = 20 */
+	while (arith_decode(cinfo, st)) {
+	  if ((m <<= 1) == 0x8000) {
+	    WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
+	    entropy->ct = -1;			/* magnitude overflow */
+	    return TRUE;
+	  }
+	  st += 1;
+	}
      }
      /* Section F.1.4.4.1.2: Establish dc_context conditioning category */
      if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1))
-        entropy->dc_context[ci] = 0;		   /* zero diff category */
+	entropy->dc_context[ci] = 0;		   /* zero diff category */
      else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1))
-        entropy->dc_context[ci] = 12 + (sign * 4); /* large diff category */
+	entropy->dc_context[ci] = 12 + (sign * 4); /* large diff category */
      else
-        entropy->dc_context[ci] = 4 + (sign * 4);  /* small diff category */
+	entropy->dc_context[ci] = 4 + (sign * 4);  /* small diff category */
      v = m;
      /* Figure F.24: Decoding the magnitude bit pattern of v */
      st += 14;
      while (m >>= 1)
-        if (arith_decode(cinfo, st)) v |= m;
+	if (arith_decode(cinfo, st)) v |= m;
      v += 1; if (sign) v = -v;
      entropy->last_dc_val[ci] += v;
    }
@@ -590,14 +592,14 @@ decode_mcu (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)
      st = entropy->ac_stats[tbl] + 3 * k;
      if (arith_decode(cinfo, st)) break;	/* EOB flag */
      for (;;) {
-        k++;
-        if (arith_decode(cinfo, st + 1)) break;
-        st += 3;
-        if (k >= cinfo->lim_Se) {
-          WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
-          entropy->ct = -1;			/* spectral overflow */
-          return TRUE;
-        }
+	k++;
+	if (arith_decode(cinfo, st + 1)) break;
+	st += 3;
+	if (k >= cinfo->lim_Se) {
+	  WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
+	  entropy->ct = -1;			/* spectral overflow */
+	  return TRUE;
+	}
      }
      /* Figure F.21: Decoding nonzero value v */
      /* Figure F.22: Decoding the sign of v */
@@ -605,25 +607,25 @@ decode_mcu (j_decompress_ptr cinfo, JBLOCKROW *MCU_data)
      st += 2;
      /* Figure F.23: Decoding the magnitude category of v */
      if ((m = arith_decode(cinfo, st)) != 0) {
-        if (arith_decode(cinfo, st)) {
-          m <<= 1;
-          st = entropy->ac_stats[tbl] +
-               (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
-          while (arith_decode(cinfo, st)) {
-            if ((m <<= 1) == 0x8000) {
-              WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
-              entropy->ct = -1;			/* magnitude overflow */
-              return TRUE;
-            }
-            st += 1;
-          }
-        }
+	if (arith_decode(cinfo, st)) {
+	  m <<= 1;
+	  st = entropy->ac_stats[tbl] +
+	       (k <= cinfo->arith_ac_K[tbl] ? 189 : 217);
+	  while (arith_decode(cinfo, st)) {
+	    if ((m <<= 1) == 0x8000) {
+	      WARNMS(cinfo, JWRN_ARITH_BAD_CODE);
+	      entropy->ct = -1;			/* magnitude overflow */
+	      return TRUE;
+	    }
+	    st += 1;
+	  }
+	}
      }
      v = m;
      /* Figure F.24: Decoding the magnitude bit pattern of v */
      st += 14;
      while (m >>= 1)
-        if (arith_decode(cinfo, st)) v |= m;
+	if (arith_decode(cinfo, st)) v |= m;
      v += 1; if (sign) v = -v;
      (*block)[natural_order[k]] = (JCOEF) v;
    } while (k < cinfo->lim_Se);
@@ -648,24 +650,24 @@ start_pass (j_decompress_ptr cinfo)
    /* Validate progressive scan parameters */
    if (cinfo->Ss == 0) {
      if (cinfo->Se != 0)
-        goto bad;
+	goto bad;
    } else {
      /* need not check Ss/Se < 0 since they came from unsigned bytes */
      if (cinfo->Se < cinfo->Ss || cinfo->Se > cinfo->lim_Se)
-        goto bad;
+	goto bad;
      /* AC scans may have only one component */
      if (cinfo->comps_in_scan != 1)
-        goto bad;
+	goto bad;
    }
    if (cinfo->Ah != 0) {
      /* Successive approximation refinement scan: must have Al = Ah-1. */
      if (cinfo->Ah-1 != cinfo->Al)
-        goto bad;
+	goto bad;
    }
    if (cinfo->Al > 13) {	/* need not check for < 0 */
      bad:
      ERREXIT4(cinfo, JERR_BAD_PROGRESSION,
-               cinfo->Ss, cinfo->Se, cinfo->Ah, cinfo->Al);
+	       cinfo->Ss, cinfo->Se, cinfo->Ah, cinfo->Al);
    }
    /* Update progression status, and verify that scan order is legal.
     * Note that inter-scan inconsistencies are treated as warnings
@@ -675,32 +677,32 @@ start_pass (j_decompress_ptr cinfo)
      int coefi, cindex = cinfo->cur_comp_info[ci]->component_index;
      int *coef_bit_ptr = & cinfo->coef_bits[cindex][0];
      if (cinfo->Ss && coef_bit_ptr[0] < 0) /* AC without prior DC scan */
-        WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, 0);
+	WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, 0);
      for (coefi = cinfo->Ss; coefi <= cinfo->Se; coefi++) {
-        int expected = (coef_bit_ptr[coefi] < 0) ? 0 : coef_bit_ptr[coefi];
-        if (cinfo->Ah != expected)
-          WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, coefi);
-        coef_bit_ptr[coefi] = cinfo->Al;
+	int expected = (coef_bit_ptr[coefi] < 0) ? 0 : coef_bit_ptr[coefi];
+	if (cinfo->Ah != expected)
+	  WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, coefi);
+	coef_bit_ptr[coefi] = cinfo->Al;
      }
    }
    /* Select MCU decoding routine */
    if (cinfo->Ah == 0) {
      if (cinfo->Ss == 0)
-        entropy->pub.decode_mcu = decode_mcu_DC_first;
+	entropy->pub.decode_mcu = decode_mcu_DC_first;
      else
-        entropy->pub.decode_mcu = decode_mcu_AC_first;
+	entropy->pub.decode_mcu = decode_mcu_AC_first;
    } else {
      if (cinfo->Ss == 0)
-        entropy->pub.decode_mcu = decode_mcu_DC_refine;
+	entropy->pub.decode_mcu = decode_mcu_DC_refine;
      else
-        entropy->pub.decode_mcu = decode_mcu_AC_refine;
+	entropy->pub.decode_mcu = decode_mcu_AC_refine;
    }
  } else {
    /* Check that the scan parameters Ss, Se, Ah/Al are OK for sequential JPEG.
     * This ought to be an error condition, but we make it a warning.
     */
    if (cinfo->Ss != 0 || cinfo->Ah != 0 || cinfo->Al != 0 ||
-        (cinfo->Se < DCTSIZE2 && cinfo->Se != cinfo->lim_Se))
+	(cinfo->Se < DCTSIZE2 && cinfo->Se != cinfo->lim_Se))
      WARNMS(cinfo, JWRN_NOT_SEQUENTIAL);
    /* Select MCU decoding routine */
    entropy->pub.decode_mcu = decode_mcu;
@@ -712,23 +714,23 @@ start_pass (j_decompress_ptr cinfo)
    if (! cinfo->progressive_mode || (cinfo->Ss == 0 && cinfo->Ah == 0)) {
      tbl = compptr->dc_tbl_no;
      if (tbl < 0 || tbl >= NUM_ARITH_TBLS)
-        ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
+	ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
      if (entropy->dc_stats[tbl] == NULL)
-        entropy->dc_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
-          ((j_common_ptr) cinfo, JPOOL_IMAGE, DC_STAT_BINS);
+	entropy->dc_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
+	  ((j_common_ptr) cinfo, JPOOL_IMAGE, DC_STAT_BINS);
      MEMZERO(entropy->dc_stats[tbl], DC_STAT_BINS);
      /* Initialize DC predictions to 0 */
      entropy->last_dc_val[ci] = 0;
      entropy->dc_context[ci] = 0;
    }
    if ((! cinfo->progressive_mode && cinfo->lim_Se) ||
-        (cinfo->progressive_mode && cinfo->Ss)) {
+	(cinfo->progressive_mode && cinfo->Ss)) {
      tbl = compptr->ac_tbl_no;
      if (tbl < 0 || tbl >= NUM_ARITH_TBLS)
-        ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
+	ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl);
      if (entropy->ac_stats[tbl] == NULL)
-        entropy->ac_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
-          ((j_common_ptr) cinfo, JPOOL_IMAGE, AC_STAT_BINS);
+	entropy->ac_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small)
+	  ((j_common_ptr) cinfo, JPOOL_IMAGE, AC_STAT_BINS);
      MEMZERO(entropy->ac_stats[tbl], AC_STAT_BINS);
    }
  }
@@ -743,6 +745,17 @@ start_pass (j_decompress_ptr cinfo)
 }


+/*
+ * Finish up at the end of an arithmetic-compressed scan.
+ */
+
+METHODDEF(void)
+finish_pass (j_decompress_ptr cinfo)
+{
+  /* no work necessary here */
+}
+
+
 /*
 * Module initialization routine for arithmetic entropy decoding.
 */
@@ -755,9 +768,10 @@ jinit_arith_decoder (j_decompress_ptr cinfo)

  entropy = (arith_entropy_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(arith_entropy_decoder));
+				SIZEOF(arith_entropy_decoder));
  cinfo->entropy = &entropy->pub;
  entropy->pub.start_pass = start_pass;
+  entropy->pub.finish_pass = finish_pass;

  /* Mark tables unallocated */
  for (i = 0; i < NUM_ARITH_TBLS; i++) {
@@ -773,10 +787,10 @@ jinit_arith_decoder (j_decompress_ptr cinfo)
    int *coef_bit_ptr, ci;
    cinfo->coef_bits = (int (*)[DCTSIZE2])
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  cinfo->num_components*DCTSIZE2*SIZEOF(int));
+				  cinfo->num_components*DCTSIZE2*SIZEOF(int));
    coef_bit_ptr = & cinfo->coef_bits[0][0];
-    for (ci = 0; ci < cinfo->num_components; ci++)
+    for (ci = 0; ci < cinfo->num_components; ci++) 
      for (i = 0; i < DCTSIZE2; i++)
-        *coef_bit_ptr++ = -1;
+	*coef_bit_ptr++ = -1;
  }
 }
@@ -68,7 +68,7 @@ init_destination (j_compress_ptr cinfo)
  /* Allocate the output buffer --- it will be released when done with image */
  dest->buffer = (JOCTET *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  OUTPUT_BUF_SIZE * SIZEOF(JOCTET));
+				  OUTPUT_BUF_SIZE * SIZEOF(JOCTET));

  dest->pub.next_output_byte = dest->buffer;
  dest->pub.free_in_buffer = OUTPUT_BUF_SIZE;
@@ -206,7 +206,7 @@ jpeg_stdio_dest (j_compress_ptr cinfo, FILE * outfile)
  if (cinfo->dest == NULL) {	/* first time for this JPEG object? */
    cinfo->dest = (struct jpeg_destination_mgr *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                  SIZEOF(my_destination_mgr));
+				  SIZEOF(my_destination_mgr));
  }

  dest = (my_dest_ptr) cinfo->dest;
@@ -233,7 +233,7 @@ jpeg_stdio_dest (j_compress_ptr cinfo, FILE * outfile)

 GLOBAL(void)
 jpeg_mem_dest (j_compress_ptr cinfo,
-               unsigned char ** outbuffer, unsigned long * outsize)
+	       unsigned char ** outbuffer, unsigned long * outsize)
 {
  my_mem_dest_ptr dest;

@@ -246,7 +246,7 @@ jpeg_mem_dest (j_compress_ptr cinfo,
  if (cinfo->dest == NULL) {	/* first time for this JPEG object? */
    cinfo->dest = (struct jpeg_destination_mgr *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                  SIZEOF(my_mem_destination_mgr));
+				  SIZEOF(my_mem_destination_mgr));
  }

  dest = (my_mem_dest_ptr) cinfo->dest;
@@ -2,7 +2,7 @@
 * jdatasrc.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
- * Modified 2009-2011 by Guido Vollbeding.
+ * Modified 2009-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -221,11 +221,11 @@ jpeg_stdio_src (j_decompress_ptr cinfo, FILE * infile)
  if (cinfo->src == NULL) {	/* first time for this JPEG object? */
    cinfo->src = (struct jpeg_source_mgr *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                  SIZEOF(my_source_mgr));
+				  SIZEOF(my_source_mgr));
    src = (my_src_ptr) cinfo->src;
    src->buffer = (JOCTET *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                  INPUT_BUF_SIZE * SIZEOF(JOCTET));
+				  INPUT_BUF_SIZE * SIZEOF(JOCTET));
  }

  src = (my_src_ptr) cinfo->src;
@@ -247,7 +247,7 @@ jpeg_stdio_src (j_decompress_ptr cinfo, FILE * infile)

 GLOBAL(void)
 jpeg_mem_src (j_decompress_ptr cinfo,
-              unsigned char * inbuffer, unsigned long insize)
+	      const unsigned char * inbuffer, unsigned long insize)
 {
  struct jpeg_source_mgr * src;

@@ -261,7 +261,7 @@ jpeg_mem_src (j_decompress_ptr cinfo,
  if (cinfo->src == NULL) {	/* first time for this JPEG object? */
    cinfo->src = (struct jpeg_source_mgr *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                  SIZEOF(struct jpeg_source_mgr));
+				  SIZEOF(struct jpeg_source_mgr));
  }

  src = cinfo->src;
@@ -271,5 +271,5 @@ jpeg_mem_src (j_decompress_ptr cinfo,
  src->resync_to_restart = jpeg_resync_to_restart; /* use default method */
  src->term_source = term_source;
  src->bytes_in_buffer = (size_t) insize;
-  src->next_input_byte = (JOCTET *) inbuffer;
+  src->next_input_byte = (const JOCTET *) inbuffer;
 }
@@ -64,15 +64,15 @@ typedef my_coef_controller * my_coef_ptr;

 /* Forward declarations */
 METHODDEF(int) decompress_onepass
-        JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf));
+	JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf));
 #ifdef D_MULTISCAN_FILES_SUPPORTED
 METHODDEF(int) decompress_data
-        JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf));
+	JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf));
 #endif
 #ifdef BLOCK_SMOOTHING_SUPPORTED
 LOCAL(boolean) smoothing_ok JPP((j_decompress_ptr cinfo));
 METHODDEF(int) decompress_smooth_data
-        JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf));
+	JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf));
 #endif


@@ -161,16 +161,16 @@ decompress_onepass (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)
  for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row;
       yoffset++) {
    for (MCU_col_num = coef->MCU_ctr; MCU_col_num <= last_MCU_col;
-         MCU_col_num++) {
+	 MCU_col_num++) {
      /* Try to fetch an MCU.  Entropy decoder expects buffer to be zeroed. */
      if (cinfo->lim_Se)	/* can bypass in DC only case */
-        FMEMZERO((void FAR *) coef->MCU_buffer[0],
-                 (size_t) (cinfo->blocks_in_MCU * SIZEOF(JBLOCK)));
+	FMEMZERO((void FAR *) coef->MCU_buffer[0],
+		 (size_t) (cinfo->blocks_in_MCU * SIZEOF(JBLOCK)));
      if (! (*cinfo->entropy->decode_mcu) (cinfo, coef->MCU_buffer)) {
-        /* Suspension forced; update state counters and exit */
-        coef->MCU_vert_offset = yoffset;
-        coef->MCU_ctr = MCU_col_num;
-        return JPEG_SUSPENDED;
+	/* Suspension forced; update state counters and exit */
+	coef->MCU_vert_offset = yoffset;
+	coef->MCU_ctr = MCU_col_num;
+	return JPEG_SUSPENDED;
      }
      /* Determine where data should go in output_buf and do the IDCT thing.
       * We skip dummy blocks at the right and bottom edges (but blkn gets
@@ -179,32 +179,32 @@ decompress_onepass (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)
       */
      blkn = 0;			/* index of current DCT block within MCU */
      for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
-        compptr = cinfo->cur_comp_info[ci];
-        /* Don't bother to IDCT an uninteresting component. */
-        if (! compptr->component_needed) {
-          blkn += compptr->MCU_blocks;
-          continue;
-        }
-        inverse_DCT = cinfo->idct->inverse_DCT[compptr->component_index];
-        useful_width = (MCU_col_num < last_MCU_col) ? compptr->MCU_width
-                                                    : compptr->last_col_width;
-        output_ptr = output_buf[compptr->component_index] +
-          yoffset * compptr->DCT_v_scaled_size;
-        start_col = MCU_col_num * compptr->MCU_sample_width;
-        for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
-          if (cinfo->input_iMCU_row < last_iMCU_row ||
-              yoffset+yindex < compptr->last_row_height) {
-            output_col = start_col;
-            for (xindex = 0; xindex < useful_width; xindex++) {
-              (*inverse_DCT) (cinfo, compptr,
-                              (JCOEFPTR) coef->MCU_buffer[blkn+xindex],
-                              output_ptr, output_col);
-              output_col += compptr->DCT_h_scaled_size;
-            }
-          }
-          blkn += compptr->MCU_width;
-          output_ptr += compptr->DCT_v_scaled_size;
-        }
+	compptr = cinfo->cur_comp_info[ci];
+	/* Don't bother to IDCT an uninteresting component. */
+	if (! compptr->component_needed) {
+	  blkn += compptr->MCU_blocks;
+	  continue;
+	}
+	inverse_DCT = cinfo->idct->inverse_DCT[compptr->component_index];
+	useful_width = (MCU_col_num < last_MCU_col) ? compptr->MCU_width
+						    : compptr->last_col_width;
+	output_ptr = output_buf[compptr->component_index] +
+	  yoffset * compptr->DCT_v_scaled_size;
+	start_col = MCU_col_num * compptr->MCU_sample_width;
+	for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
+	  if (cinfo->input_iMCU_row < last_iMCU_row ||
+	      yoffset+yindex < compptr->last_row_height) {
+	    output_col = start_col;
+	    for (xindex = 0; xindex < useful_width; xindex++) {
+	      (*inverse_DCT) (cinfo, compptr,
+			      (JCOEFPTR) coef->MCU_buffer[blkn+xindex],
+			      output_ptr, output_col);
+	      output_col += compptr->DCT_h_scaled_size;
+	    }
+	  }
+	  blkn += compptr->MCU_width;
+	  output_ptr += compptr->DCT_v_scaled_size;
+	}
      }
    }
    /* Completed an MCU row, but perhaps not an iMCU row */
@@ -270,25 +270,25 @@ consume_data (j_decompress_ptr cinfo)
  for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row;
       yoffset++) {
    for (MCU_col_num = coef->MCU_ctr; MCU_col_num < cinfo->MCUs_per_row;
-         MCU_col_num++) {
+	 MCU_col_num++) {
      /* Construct list of pointers to DCT blocks belonging to this MCU */
      blkn = 0;			/* index of current DCT block within MCU */
      for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
-        compptr = cinfo->cur_comp_info[ci];
-        start_col = MCU_col_num * compptr->MCU_width;
-        for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
-          buffer_ptr = buffer[ci][yindex+yoffset] + start_col;
-          for (xindex = 0; xindex < compptr->MCU_width; xindex++) {
-            coef->MCU_buffer[blkn++] = buffer_ptr++;
-          }
-        }
+	compptr = cinfo->cur_comp_info[ci];
+	start_col = MCU_col_num * compptr->MCU_width;
+	for (yindex = 0; yindex < compptr->MCU_height; yindex++) {
+	  buffer_ptr = buffer[ci][yindex+yoffset] + start_col;
+	  for (xindex = 0; xindex < compptr->MCU_width; xindex++) {
+	    coef->MCU_buffer[blkn++] = buffer_ptr++;
+	  }
+	}
      }
      /* Try to fetch the MCU. */
      if (! (*cinfo->entropy->decode_mcu) (cinfo, coef->MCU_buffer)) {
-        /* Suspension forced; update state counters and exit */
-        coef->MCU_vert_offset = yoffset;
-        coef->MCU_ctr = MCU_col_num;
-        return JPEG_SUSPENDED;
+	/* Suspension forced; update state counters and exit */
+	coef->MCU_vert_offset = yoffset;
+	coef->MCU_ctr = MCU_col_num;
+	return JPEG_SUSPENDED;
      }
    }
    /* Completed an MCU row, but perhaps not an iMCU row */
@@ -329,8 +329,8 @@ decompress_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)

  /* Force some input to be done if we are getting ahead of the input. */
  while (cinfo->input_scan_number < cinfo->output_scan_number ||
-         (cinfo->input_scan_number == cinfo->output_scan_number &&
-          cinfo->input_iMCU_row <= cinfo->output_iMCU_row)) {
+	 (cinfo->input_scan_number == cinfo->output_scan_number &&
+	  cinfo->input_iMCU_row <= cinfo->output_iMCU_row)) {
    if ((*cinfo->inputctl->consume_input)(cinfo) == JPEG_SUSPENDED)
      return JPEG_SUSPENDED;
  }
@@ -361,10 +361,10 @@ decompress_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)
      buffer_ptr = buffer[block_row];
      output_col = 0;
      for (block_num = 0; block_num < compptr->width_in_blocks; block_num++) {
-        (*inverse_DCT) (cinfo, compptr, (JCOEFPTR) buffer_ptr,
-                        output_ptr, output_col);
-        buffer_ptr++;
-        output_col += compptr->DCT_h_scaled_size;
+	(*inverse_DCT) (cinfo, compptr, (JCOEFPTR) buffer_ptr,
+			output_ptr, output_col);
+	buffer_ptr++;
+	output_col += compptr->DCT_h_scaled_size;
      }
      output_ptr += compptr->DCT_v_scaled_size;
    }
@@ -421,8 +421,8 @@ smoothing_ok (j_decompress_ptr cinfo)
  if (coef->coef_bits_latch == NULL)
    coef->coef_bits_latch = (int *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  cinfo->num_components *
-                                  (SAVED_COEFS * SIZEOF(int)));
+				  cinfo->num_components *
+				  (SAVED_COEFS * SIZEOF(int)));
  coef_bits_latch = coef->coef_bits_latch;

  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
@@ -432,11 +432,11 @@ smoothing_ok (j_decompress_ptr cinfo)
      return FALSE;
    /* Verify DC & first 5 AC quantizers are nonzero to avoid zero-divide. */
    if (qtable->quantval[0] == 0 ||
-        qtable->quantval[Q01_POS] == 0 ||
-        qtable->quantval[Q10_POS] == 0 ||
-        qtable->quantval[Q20_POS] == 0 ||
-        qtable->quantval[Q11_POS] == 0 ||
-        qtable->quantval[Q02_POS] == 0)
+	qtable->quantval[Q01_POS] == 0 ||
+	qtable->quantval[Q10_POS] == 0 ||
+	qtable->quantval[Q20_POS] == 0 ||
+	qtable->quantval[Q11_POS] == 0 ||
+	qtable->quantval[Q02_POS] == 0)
      return FALSE;
    /* DC values must be at least partly known for all components. */
    coef_bits = cinfo->coef_bits[ci];
@@ -446,7 +446,7 @@ smoothing_ok (j_decompress_ptr cinfo)
    for (coefi = 1; coefi <= 5; coefi++) {
      coef_bits_latch[coefi] = coef_bits[coefi];
      if (coef_bits[coefi] != 0)
-        smoothing_useful = TRUE;
+	smoothing_useful = TRUE;
    }
    coef_bits_latch += SAVED_COEFS;
  }
@@ -482,7 +482,7 @@ decompress_smooth_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)

  /* Force some input to be done if we are getting ahead of the input. */
  while (cinfo->input_scan_number <= cinfo->output_scan_number &&
-         ! cinfo->inputctl->eoi_reached) {
+	 ! cinfo->inputctl->eoi_reached) {
    if (cinfo->input_scan_number == cinfo->output_scan_number) {
      /* If input is working on current scan, we ordinarily want it to
       * have completed the current row.  But if input scan is DC,
@@ -491,7 +491,7 @@ decompress_smooth_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)
       */
      JDIMENSION delta = (cinfo->Ss == 0) ? 1 : 0;
      if (cinfo->input_iMCU_row > cinfo->output_iMCU_row+delta)
-        break;
+	break;
    }
    if ((*cinfo->inputctl->consume_input)(cinfo) == JPEG_SUSPENDED)
      return JPEG_SUSPENDED;
@@ -519,15 +519,15 @@ decompress_smooth_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)
    if (cinfo->output_iMCU_row > 0) {
      access_rows += compptr->v_samp_factor; /* prior iMCU row too */
      buffer = (*cinfo->mem->access_virt_barray)
-        ((j_common_ptr) cinfo, coef->whole_image[ci],
-         (cinfo->output_iMCU_row - 1) * compptr->v_samp_factor,
-         (JDIMENSION) access_rows, FALSE);
+	((j_common_ptr) cinfo, coef->whole_image[ci],
+	 (cinfo->output_iMCU_row - 1) * compptr->v_samp_factor,
+	 (JDIMENSION) access_rows, FALSE);
      buffer += compptr->v_samp_factor;	/* point to current iMCU row */
      first_row = FALSE;
    } else {
      buffer = (*cinfo->mem->access_virt_barray)
-        ((j_common_ptr) cinfo, coef->whole_image[ci],
-         (JDIMENSION) 0, (JDIMENSION) access_rows, FALSE);
+	((j_common_ptr) cinfo, coef->whole_image[ci],
+	 (JDIMENSION) 0, (JDIMENSION) access_rows, FALSE);
      first_row = TRUE;
    }
    /* Fetch component-dependent info */
@@ -545,13 +545,13 @@ decompress_smooth_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)
    for (block_row = 0; block_row < block_rows; block_row++) {
      buffer_ptr = buffer[block_row];
      if (first_row && block_row == 0)
-        prev_block_row = buffer_ptr;
+	prev_block_row = buffer_ptr;
      else
-        prev_block_row = buffer[block_row-1];
+	prev_block_row = buffer[block_row-1];
      if (last_row && block_row == block_rows-1)
-        next_block_row = buffer_ptr;
+	next_block_row = buffer_ptr;
      else
-        next_block_row = buffer[block_row+1];
+	next_block_row = buffer[block_row+1];
      /* We fetch the surrounding DC values using a sliding-register approach.
       * Initialize all nine here so as to do the right thing on narrow pics.
       */
@@ -561,102 +561,102 @@ decompress_smooth_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf)
      output_col = 0;
      last_block_column = compptr->width_in_blocks - 1;
      for (block_num = 0; block_num <= last_block_column; block_num++) {
-        /* Fetch current DCT block into workspace so we can modify it. */
-        jcopy_block_row(buffer_ptr, (JBLOCKROW) workspace, (JDIMENSION) 1);
-        /* Update DC values */
-        if (block_num < last_block_column) {
-          DC3 = (int) prev_block_row[1][0];
-          DC6 = (int) buffer_ptr[1][0];
-          DC9 = (int) next_block_row[1][0];
-        }
-        /* Compute coefficient estimates per K.8.
-         * An estimate is applied only if coefficient is still zero,
-         * and is not known to be fully accurate.
-         */
-        /* AC01 */
-        if ((Al=coef_bits[1]) != 0 && workspace[1] == 0) {
-          num = 36 * Q00 * (DC4 - DC6);
-          if (num >= 0) {
-            pred = (int) (((Q01<<7) + num) / (Q01<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-          } else {
-            pred = (int) (((Q01<<7) - num) / (Q01<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-            pred = -pred;
-          }
-          workspace[1] = (JCOEF) pred;
-        }
-        /* AC10 */
-        if ((Al=coef_bits[2]) != 0 && workspace[8] == 0) {
-          num = 36 * Q00 * (DC2 - DC8);
-          if (num >= 0) {
-            pred = (int) (((Q10<<7) + num) / (Q10<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-          } else {
-            pred = (int) (((Q10<<7) - num) / (Q10<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-            pred = -pred;
-          }
-          workspace[8] = (JCOEF) pred;
-        }
-        /* AC20 */
-        if ((Al=coef_bits[3]) != 0 && workspace[16] == 0) {
-          num = 9 * Q00 * (DC2 + DC8 - 2*DC5);
-          if (num >= 0) {
-            pred = (int) (((Q20<<7) + num) / (Q20<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-          } else {
-            pred = (int) (((Q20<<7) - num) / (Q20<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-            pred = -pred;
-          }
-          workspace[16] = (JCOEF) pred;
-        }
-        /* AC11 */
-        if ((Al=coef_bits[4]) != 0 && workspace[9] == 0) {
-          num = 5 * Q00 * (DC1 - DC3 - DC7 + DC9);
-          if (num >= 0) {
-            pred = (int) (((Q11<<7) + num) / (Q11<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-          } else {
-            pred = (int) (((Q11<<7) - num) / (Q11<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-            pred = -pred;
-          }
-          workspace[9] = (JCOEF) pred;
-        }
-        /* AC02 */
-        if ((Al=coef_bits[5]) != 0 && workspace[2] == 0) {
-          num = 9 * Q00 * (DC4 + DC6 - 2*DC5);
-          if (num >= 0) {
-            pred = (int) (((Q02<<7) + num) / (Q02<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-          } else {
-            pred = (int) (((Q02<<7) - num) / (Q02<<8));
-            if (Al > 0 && pred >= (1<<Al))
-              pred = (1<<Al)-1;
-            pred = -pred;
-          }
-          workspace[2] = (JCOEF) pred;
-        }
-        /* OK, do the IDCT */
-        (*inverse_DCT) (cinfo, compptr, (JCOEFPTR) workspace,
-                        output_ptr, output_col);
-        /* Advance for next column */
-        DC1 = DC2; DC2 = DC3;
-        DC4 = DC5; DC5 = DC6;
-        DC7 = DC8; DC8 = DC9;
-        buffer_ptr++, prev_block_row++, next_block_row++;
-        output_col += compptr->DCT_h_scaled_size;
+	/* Fetch current DCT block into workspace so we can modify it. */
+	jcopy_block_row(buffer_ptr, (JBLOCKROW) workspace, (JDIMENSION) 1);
+	/* Update DC values */
+	if (block_num < last_block_column) {
+	  DC3 = (int) prev_block_row[1][0];
+	  DC6 = (int) buffer_ptr[1][0];
+	  DC9 = (int) next_block_row[1][0];
+	}
+	/* Compute coefficient estimates per K.8.
+	 * An estimate is applied only if coefficient is still zero,
+	 * and is not known to be fully accurate.
+	 */
+	/* AC01 */
+	if ((Al=coef_bits[1]) != 0 && workspace[1] == 0) {
+	  num = 36 * Q00 * (DC4 - DC6);
+	  if (num >= 0) {
+	    pred = (int) (((Q01<<7) + num) / (Q01<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	  } else {
+	    pred = (int) (((Q01<<7) - num) / (Q01<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	    pred = -pred;
+	  }
+	  workspace[1] = (JCOEF) pred;
+	}
+	/* AC10 */
+	if ((Al=coef_bits[2]) != 0 && workspace[8] == 0) {
+	  num = 36 * Q00 * (DC2 - DC8);
+	  if (num >= 0) {
+	    pred = (int) (((Q10<<7) + num) / (Q10<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	  } else {
+	    pred = (int) (((Q10<<7) - num) / (Q10<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	    pred = -pred;
+	  }
+	  workspace[8] = (JCOEF) pred;
+	}
+	/* AC20 */
+	if ((Al=coef_bits[3]) != 0 && workspace[16] == 0) {
+	  num = 9 * Q00 * (DC2 + DC8 - 2*DC5);
+	  if (num >= 0) {
+	    pred = (int) (((Q20<<7) + num) / (Q20<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	  } else {
+	    pred = (int) (((Q20<<7) - num) / (Q20<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	    pred = -pred;
+	  }
+	  workspace[16] = (JCOEF) pred;
+	}
+	/* AC11 */
+	if ((Al=coef_bits[4]) != 0 && workspace[9] == 0) {
+	  num = 5 * Q00 * (DC1 - DC3 - DC7 + DC9);
+	  if (num >= 0) {
+	    pred = (int) (((Q11<<7) + num) / (Q11<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	  } else {
+	    pred = (int) (((Q11<<7) - num) / (Q11<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	    pred = -pred;
+	  }
+	  workspace[9] = (JCOEF) pred;
+	}
+	/* AC02 */
+	if ((Al=coef_bits[5]) != 0 && workspace[2] == 0) {
+	  num = 9 * Q00 * (DC4 + DC6 - 2*DC5);
+	  if (num >= 0) {
+	    pred = (int) (((Q02<<7) + num) / (Q02<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	  } else {
+	    pred = (int) (((Q02<<7) - num) / (Q02<<8));
+	    if (Al > 0 && pred >= (1<<Al))
+	      pred = (1<<Al)-1;
+	    pred = -pred;
+	  }
+	  workspace[2] = (JCOEF) pred;
+	}
+	/* OK, do the IDCT */
+	(*inverse_DCT) (cinfo, compptr, (JCOEFPTR) workspace,
+			output_ptr, output_col);
+	/* Advance for next column */
+	DC1 = DC2; DC2 = DC3;
+	DC4 = DC5; DC5 = DC6;
+	DC7 = DC8; DC8 = DC9;
+	buffer_ptr++, prev_block_row++, next_block_row++;
+	output_col += compptr->DCT_h_scaled_size;
      }
      output_ptr += compptr->DCT_v_scaled_size;
    }
@@ -681,7 +681,7 @@ jinit_d_coef_controller (j_decompress_ptr cinfo, boolean need_full_buffer)

  coef = (my_coef_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_coef_controller));
+				SIZEOF(my_coef_controller));
  cinfo->coef = (struct jpeg_d_coef_controller *) coef;
  coef->pub.start_input_pass = start_input_pass;
  coef->pub.start_output_pass = start_output_pass;
@@ -699,20 +699,20 @@ jinit_d_coef_controller (j_decompress_ptr cinfo, boolean need_full_buffer)
    jpeg_component_info *compptr;

    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      access_rows = compptr->v_samp_factor;
 #ifdef BLOCK_SMOOTHING_SUPPORTED
      /* If block smoothing could be used, need a bigger window */
      if (cinfo->progressive_mode)
-        access_rows *= 3;
+	access_rows *= 3;
 #endif
      coef->whole_image[ci] = (*cinfo->mem->request_virt_barray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE, TRUE,
-         (JDIMENSION) jround_up((long) compptr->width_in_blocks,
-                                (long) compptr->h_samp_factor),
-         (JDIMENSION) jround_up((long) compptr->height_in_blocks,
-                                (long) compptr->v_samp_factor),
-         (JDIMENSION) access_rows);
+	((j_common_ptr) cinfo, JPOOL_IMAGE, TRUE,
+	 (JDIMENSION) jround_up((long) compptr->width_in_blocks,
+				(long) compptr->h_samp_factor),
+	 (JDIMENSION) jround_up((long) compptr->height_in_blocks,
+				(long) compptr->v_samp_factor),
+	 (JDIMENSION) access_rows);
    }
    coef->pub.consume_data = consume_data;
    coef->pub.decompress_data = decompress_data;
@@ -727,13 +727,13 @@ jinit_d_coef_controller (j_decompress_ptr cinfo, boolean need_full_buffer)

    buffer = (JBLOCKROW)
      (*cinfo->mem->alloc_large) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  D_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK));
+				  D_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK));
    for (i = 0; i < D_MAX_BLOCKS_IN_MCU; i++) {
      coef->MCU_buffer[i] = buffer + i;
    }
    if (cinfo->lim_Se == 0)	/* DC only case: want to bypass later */
      FMEMZERO((void FAR *) buffer,
-               (size_t) (D_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK)));
+	       (size_t) (D_MAX_BLOCKS_IN_MCU * SIZEOF(JBLOCK)));
    coef->pub.consume_data = dummy_consume_data;
    coef->pub.decompress_data = decompress_onepass;
    coef->pub.coef_arrays = NULL; /* flag for no virtual arrays */
@@ -2,7 +2,7 @@
 * jdcolor.c
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
- * Modified 2011-2012 by Guido Vollbeding.
+ * Modified 2011-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -19,7 +19,7 @@
 typedef struct {
  struct jpeg_color_deconverter pub; /* public fields */

-  /* Private state for YCC->RGB conversion */
+  /* Private state for YCbCr->RGB and BG_YCC->RGB conversion */
  int * Cr_r_tab;		/* => table for Cr to R conversion */
  int * Cb_b_tab;		/* => table for Cb to B conversion */
  INT32 * Cr_g_tab;		/* => table for Cr to G conversion */
@@ -32,22 +32,44 @@ typedef struct {
 typedef my_color_deconverter * my_cconvert_ptr;


-/**************** YCbCr -> RGB conversion: most common case **************/
-/****************   RGB -> Y   conversion: less common case **************/
+/***************  YCbCr -> RGB conversion: most common case **************/
+/*************** BG_YCC -> RGB conversion: less common case **************/
+/***************    RGB -> Y   conversion: less common case **************/

 /*
- * YCbCr is defined per CCIR 601-1, except that Cb and Cr are
- * normalized to the range 0..MAXJSAMPLE rather than -0.5 .. 0.5.
- * The conversion equations to be implemented are therefore
+ * YCbCr is defined per Recommendation ITU-R BT.601-7 (03/2011),
+ * previously known as Recommendation CCIR 601-1, except that Cb and Cr
+ * are normalized to the range 0..MAXJSAMPLE rather than -0.5 .. 0.5.
+ * sRGB (standard RGB color space) is defined per IEC 61966-2-1:1999.
+ * sYCC (standard luma-chroma-chroma color space with extended gamut)
+ * is defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex F.
+ * bg-sRGB and bg-sYCC (big gamut standard color spaces)
+ * are defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex G.
+ * Note that the derived conversion coefficients given in some of these
+ * documents are imprecise.  The general conversion equations are
 *
- *	R = Y                + 1.40200 * Cr
- *	G = Y - 0.34414 * Cb - 0.71414 * Cr
- *	B = Y + 1.77200 * Cb
+ *	R = Y + K * (1 - Kr) * Cr
+ *	G = Y - K * (Kb * (1 - Kb) * Cb + Kr * (1 - Kr) * Cr) / (1 - Kr - Kb)
+ *	B = Y + K * (1 - Kb) * Cb
 *
- *	Y = 0.29900 * R + 0.58700 * G + 0.11400 * B
+ *	Y = Kr * R + (1 - Kr - Kb) * G + Kb * B
+ *
+ * With Kr = 0.299 and Kb = 0.114 (derived according to SMPTE RP 177-1993
+ * from the 1953 FCC NTSC primaries and CIE Illuminant C), K = 2 for sYCC,
+ * the conversion equations to be implemented are therefore
+ *
+ *	R = Y + 1.402 * Cr
+ *	G = Y - 0.344136286 * Cb - 0.714136286 * Cr
+ *	B = Y + 1.772 * Cb
+ *
+ *	Y = 0.299 * R + 0.587 * G + 0.114 * B
 *
 * where Cb and Cr represent the incoming values less CENTERJSAMPLE.
- * (These numbers are derived from TIFF 6.0 section 21, dated 3-June-92.)
+ * For bg-sYCC, with K = 4, the equations are
+ *
+ *	R = Y + 2.804 * Cr
+ *	G = Y - 0.688272572 * Cb - 1.428272572 * Cr
+ *	B = Y + 3.544 * Cb
 *
 * To avoid floating-point arithmetic, we represent the fractional constants
 * as integers scaled up by 2^16 (about 4 digits precision); we have to divide
@@ -58,9 +80,9 @@ typedef my_color_deconverter * my_cconvert_ptr;
 * For even more speed, we avoid doing any multiplications in the inner loop
 * by precalculating the constants times Cb and Cr for all possible values.
 * For 8-bit JSAMPLEs this is very reasonable (only 256 entries per table);
- * for 12-bit samples it is still acceptable.  It's not very reasonable for
- * 16-bit samples, but if you want lossless storage you shouldn't be changing
- * colorspace anyway.
+ * for 9-bit to 12-bit samples it is still acceptable.  It's not very
+ * reasonable for 16-bit samples, but if you want lossless storage you
+ * shouldn't be changing colorspace anyway.
 * The Cr=>R and Cb=>B values can be rounded to integers in advance; the
 * values for the G calculation are left scaled up, since we must add them
 * together before rounding.
@@ -84,11 +106,12 @@ typedef my_color_deconverter * my_cconvert_ptr;


 /*
- * Initialize tables for YCC->RGB colorspace conversion.
+ * Initialize tables for YCbCr->RGB and BG_YCC->RGB colorspace conversion.
 */

 LOCAL(void)
 build_ycc_rgb_table (j_decompress_ptr cinfo)
+/* Normal case, sYCC */
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  int i;
@@ -97,31 +120,71 @@ build_ycc_rgb_table (j_decompress_ptr cinfo)

  cconvert->Cr_r_tab = (int *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(int));
+				(MAXJSAMPLE+1) * SIZEOF(int));
  cconvert->Cb_b_tab = (int *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(int));
+				(MAXJSAMPLE+1) * SIZEOF(int));
  cconvert->Cr_g_tab = (INT32 *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(INT32));
+				(MAXJSAMPLE+1) * SIZEOF(INT32));
  cconvert->Cb_g_tab = (INT32 *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(INT32));
+				(MAXJSAMPLE+1) * SIZEOF(INT32));

  for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) {
    /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */
    /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */
-    /* Cr=>R value is nearest int to 1.40200 * x */
+    /* Cr=>R value is nearest int to 1.402 * x */
    cconvert->Cr_r_tab[i] = (int)
-                    RIGHT_SHIFT(FIX(1.40200) * x + ONE_HALF, SCALEBITS);
-    /* Cb=>B value is nearest int to 1.77200 * x */
+		    RIGHT_SHIFT(FIX(1.402) * x + ONE_HALF, SCALEBITS);
+    /* Cb=>B value is nearest int to 1.772 * x */
    cconvert->Cb_b_tab[i] = (int)
-                    RIGHT_SHIFT(FIX(1.77200) * x + ONE_HALF, SCALEBITS);
-    /* Cr=>G value is scaled-up -0.71414 * x */
-    cconvert->Cr_g_tab[i] = (- FIX(0.71414)) * x;
-    /* Cb=>G value is scaled-up -0.34414 * x */
+		    RIGHT_SHIFT(FIX(1.772) * x + ONE_HALF, SCALEBITS);
+    /* Cr=>G value is scaled-up -0.714136286 * x */
+    cconvert->Cr_g_tab[i] = (- FIX(0.714136286)) * x;
+    /* Cb=>G value is scaled-up -0.344136286 * x */
    /* We also add in ONE_HALF so that need not do it in inner loop */
-    cconvert->Cb_g_tab[i] = (- FIX(0.34414)) * x + ONE_HALF;
+    cconvert->Cb_g_tab[i] = (- FIX(0.344136286)) * x + ONE_HALF;
+  }
+}
+
+
+LOCAL(void)
+build_bg_ycc_rgb_table (j_decompress_ptr cinfo)
+/* Wide gamut case, bg-sYCC */
+{
+  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
+  int i;
+  INT32 x;
+  SHIFT_TEMPS
+
+  cconvert->Cr_r_tab = (int *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(int));
+  cconvert->Cb_b_tab = (int *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(int));
+  cconvert->Cr_g_tab = (INT32 *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(INT32));
+  cconvert->Cb_g_tab = (INT32 *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(INT32));
+
+  for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) {
+    /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */
+    /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */
+    /* Cr=>R value is nearest int to 2.804 * x */
+    cconvert->Cr_r_tab[i] = (int)
+		    RIGHT_SHIFT(FIX(2.804) * x + ONE_HALF, SCALEBITS);
+    /* Cb=>B value is nearest int to 3.544 * x */
+    cconvert->Cb_b_tab[i] = (int)
+		    RIGHT_SHIFT(FIX(3.544) * x + ONE_HALF, SCALEBITS);
+    /* Cr=>G value is scaled-up -1.428272572 * x */
+    cconvert->Cr_g_tab[i] = (- FIX(1.428272572)) * x;
+    /* Cb=>G value is scaled-up -0.688272572 * x */
+    /* We also add in ONE_HALF so that need not do it in inner loop */
+    cconvert->Cb_g_tab[i] = (- FIX(0.688272572)) * x + ONE_HALF;
  }
 }

@@ -139,8 +202,8 @@ build_ycc_rgb_table (j_decompress_ptr cinfo)

 METHODDEF(void)
 ycc_rgb_convert (j_decompress_ptr cinfo,
-                 JSAMPIMAGE input_buf, JDIMENSION input_row,
-                 JSAMPARRAY output_buf, int num_rows)
+		 JSAMPIMAGE input_buf, JDIMENSION input_row,
+		 JSAMPARRAY output_buf, int num_rows)
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  register int y, cb, cr;
@@ -166,19 +229,21 @@ ycc_rgb_convert (j_decompress_ptr cinfo,
      y  = GETJSAMPLE(inptr0[col]);
      cb = GETJSAMPLE(inptr1[col]);
      cr = GETJSAMPLE(inptr2[col]);
-      /* Range-limiting is essential due to noise introduced by DCT losses. */
-      outptr[RGB_RED] =   range_limit[y + Crrtab[cr]];
+      /* Range-limiting is essential due to noise introduced by DCT losses,
+       * for extended gamut (sYCC) and wide gamut (bg-sYCC) encodings.
+       */
+      outptr[RGB_RED]   = range_limit[y + Crrtab[cr]];
      outptr[RGB_GREEN] = range_limit[y +
-                              ((int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr],
-                                                 SCALEBITS))];
-      outptr[RGB_BLUE] =  range_limit[y + Cbbtab[cb]];
+			      ((int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr],
+						 SCALEBITS))];
+      outptr[RGB_BLUE]  = range_limit[y + Cbbtab[cb]];
      outptr += RGB_PIXELSIZE;
    }
  }
 }


-/**************** Cases other than YCbCr -> RGB **************/
+/**************** Cases other than YCC -> RGB ****************/


 /*
@@ -195,12 +260,12 @@ build_rgb_y_table (j_decompress_ptr cinfo)
  /* Allocate and fill in the conversion tables. */
  cconvert->rgb_y_tab = rgb_y_tab = (INT32 *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (TABLE_SIZE * SIZEOF(INT32)));
+				(TABLE_SIZE * SIZEOF(INT32)));

  for (i = 0; i <= MAXJSAMPLE; i++) {
-    rgb_y_tab[i+R_Y_OFF] = FIX(0.29900) * i;
-    rgb_y_tab[i+G_Y_OFF] = FIX(0.58700) * i;
-    rgb_y_tab[i+B_Y_OFF] = FIX(0.11400) * i + ONE_HALF;
+    rgb_y_tab[i+R_Y_OFF] = FIX(0.299) * i;
+    rgb_y_tab[i+G_Y_OFF] = FIX(0.587) * i;
+    rgb_y_tab[i+B_Y_OFF] = FIX(0.114) * i + ONE_HALF;
  }
 }

@@ -211,8 +276,8 @@ build_rgb_y_table (j_decompress_ptr cinfo)

 METHODDEF(void)
 rgb_gray_convert (j_decompress_ptr cinfo,
-                  JSAMPIMAGE input_buf, JDIMENSION input_row,
-                  JSAMPARRAY output_buf, int num_rows)
+		  JSAMPIMAGE input_buf, JDIMENSION input_row,
+		  JSAMPARRAY output_buf, int num_rows)
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  register INT32 * ctab = cconvert->rgb_y_tab;
@@ -234,8 +299,8 @@ rgb_gray_convert (j_decompress_ptr cinfo,
      b = GETJSAMPLE(inptr2[col]);
      /* Y */
      outptr[col] = (JSAMPLE)
-                ((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
+		 >> SCALEBITS);
    }
  }
 }
@@ -244,12 +309,15 @@ rgb_gray_convert (j_decompress_ptr cinfo,
 /*
 * [R-G,G,B-G] to [R,G,B] conversion with modulo calculation
 * (inverse color transform).
+ * This can be seen as an adaption of the general YCbCr->RGB
+ * conversion equation with Kr = Kb = 0, while replacing the
+ * normalization by modulo calculation.
 */

 METHODDEF(void)
 rgb1_rgb_convert (j_decompress_ptr cinfo,
-                  JSAMPIMAGE input_buf, JDIMENSION input_row,
-                  JSAMPARRAY output_buf, int num_rows)
+		  JSAMPIMAGE input_buf, JDIMENSION input_row,
+		  JSAMPARRAY output_buf, int num_rows)
 {
  register int r, g, b;
  register JSAMPROW outptr;
@@ -286,8 +354,8 @@ rgb1_rgb_convert (j_decompress_ptr cinfo,

 METHODDEF(void)
 rgb1_gray_convert (j_decompress_ptr cinfo,
-                   JSAMPIMAGE input_buf, JDIMENSION input_row,
-                   JSAMPARRAY output_buf, int num_rows)
+		   JSAMPIMAGE input_buf, JDIMENSION input_row,
+		   JSAMPARRAY output_buf, int num_rows)
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  register INT32 * ctab = cconvert->rgb_y_tab;
@@ -314,8 +382,8 @@ rgb1_gray_convert (j_decompress_ptr cinfo,
      b = (b + g - CENTERJSAMPLE) & MAXJSAMPLE;
      /* Y */
      outptr[col] = (JSAMPLE)
-                ((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
-                 >> SCALEBITS);
+		((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF])
+		 >> SCALEBITS);
    }
  }
 }
@@ -328,8 +396,8 @@ rgb1_gray_convert (j_decompress_ptr cinfo,

 METHODDEF(void)
 rgb_convert (j_decompress_ptr cinfo,
-             JSAMPIMAGE input_buf, JDIMENSION input_row,
-             JSAMPARRAY output_buf, int num_rows)
+	     JSAMPIMAGE input_buf, JDIMENSION input_row,
+	     JSAMPARRAY output_buf, int num_rows)
 {
  register JSAMPROW outptr;
  register JSAMPROW inptr0, inptr1, inptr2;
@@ -360,8 +428,8 @@ rgb_convert (j_decompress_ptr cinfo,

 METHODDEF(void)
 null_convert (j_decompress_ptr cinfo,
-              JSAMPIMAGE input_buf, JDIMENSION input_row,
-              JSAMPARRAY output_buf, int num_rows)
+	      JSAMPIMAGE input_buf, JDIMENSION input_row,
+	      JSAMPARRAY output_buf, int num_rows)
 {
  int ci;
  register int nc = cinfo->num_components;
@@ -375,8 +443,8 @@ null_convert (j_decompress_ptr cinfo,
      inptr = input_buf[ci][input_row];
      outptr = output_buf[0] + ci;
      for (col = 0; col < num_cols; col++) {
-        *outptr = *inptr++;	/* needn't bother with GETJSAMPLE() here */
-        outptr += nc;
+	*outptr = *inptr++;	/* needn't bother with GETJSAMPLE() here */
+	outptr += nc;
      }
    }
    input_row++;
@@ -387,17 +455,17 @@ null_convert (j_decompress_ptr cinfo,

 /*
 * Color conversion for grayscale: just copy the data.
- * This also works for YCbCr -> grayscale conversion, in which
+ * This also works for YCC -> grayscale conversion, in which
 * we just copy the Y (luminance) component and ignore chrominance.
 */

 METHODDEF(void)
 grayscale_convert (j_decompress_ptr cinfo,
-                   JSAMPIMAGE input_buf, JDIMENSION input_row,
-                   JSAMPARRAY output_buf, int num_rows)
+		   JSAMPIMAGE input_buf, JDIMENSION input_row,
+		   JSAMPARRAY output_buf, int num_rows)
 {
  jcopy_sample_rows(input_buf[0], (int) input_row, output_buf, 0,
-                    num_rows, cinfo->output_width);
+		    num_rows, cinfo->output_width);
 }


@@ -409,8 +477,8 @@ grayscale_convert (j_decompress_ptr cinfo,

 METHODDEF(void)
 gray_rgb_convert (j_decompress_ptr cinfo,
-                  JSAMPIMAGE input_buf, JDIMENSION input_row,
-                  JSAMPARRAY output_buf, int num_rows)
+		  JSAMPIMAGE input_buf, JDIMENSION input_row,
+		  JSAMPARRAY output_buf, int num_rows)
 {
  register JSAMPROW outptr;
  register JSAMPROW inptr;
@@ -438,8 +506,8 @@ gray_rgb_convert (j_decompress_ptr cinfo,

 METHODDEF(void)
 ycck_cmyk_convert (j_decompress_ptr cinfo,
-                   JSAMPIMAGE input_buf, JDIMENSION input_row,
-                   JSAMPARRAY output_buf, int num_rows)
+		   JSAMPIMAGE input_buf, JDIMENSION input_row,
+		   JSAMPARRAY output_buf, int num_rows)
 {
  my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert;
  register int y, cb, cr;
@@ -466,11 +534,13 @@ ycck_cmyk_convert (j_decompress_ptr cinfo,
      y  = GETJSAMPLE(inptr0[col]);
      cb = GETJSAMPLE(inptr1[col]);
      cr = GETJSAMPLE(inptr2[col]);
-      /* Range-limiting is essential due to noise introduced by DCT losses. */
+      /* Range-limiting is essential due to noise introduced by DCT losses,
+       * and for extended gamut encodings (sYCC).
+       */
      outptr[0] = range_limit[MAXJSAMPLE - (y + Crrtab[cr])];	/* red */
      outptr[1] = range_limit[MAXJSAMPLE - (y +			/* green */
-                              ((int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr],
-                                                 SCALEBITS)))];
+			      ((int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr],
+						 SCALEBITS)))];
      outptr[2] = range_limit[MAXJSAMPLE - (y + Cbbtab[cb])];	/* blue */
      /* K passes through unchanged */
      outptr[3] = inptr3[col];	/* don't need GETJSAMPLE here */
@@ -503,7 +573,7 @@ jinit_color_deconverter (j_decompress_ptr cinfo)

  cconvert = (my_cconvert_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_color_deconverter));
+				SIZEOF(my_color_deconverter));
  cinfo->cconvert = &cconvert->pub;
  cconvert->pub.start_pass = start_pass_dcolor;

@@ -516,6 +586,8 @@ jinit_color_deconverter (j_decompress_ptr cinfo)

  case JCS_RGB:
  case JCS_YCbCr:
+  case JCS_BG_RGB:
+  case JCS_BG_YCC:
    if (cinfo->num_components != 3)
      ERREXIT(cinfo, JERR_BAD_J_COLORSPACE);
    break;
@@ -532,8 +604,10 @@ jinit_color_deconverter (j_decompress_ptr cinfo)
    break;
  }

-  /* Support color transform only for RGB colorspace */
-  if (cinfo->color_transform && cinfo->jpeg_color_space != JCS_RGB)
+  /* Support color transform only for RGB colorspaces */
+  if (cinfo->color_transform &&
+      cinfo->jpeg_color_space != JCS_RGB &&
+      cinfo->jpeg_color_space != JCS_BG_RGB)
    ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);

  /* Set out_color_components and conversion method based on requested space.
@@ -544,47 +618,76 @@ jinit_color_deconverter (j_decompress_ptr cinfo)
  switch (cinfo->out_color_space) {
  case JCS_GRAYSCALE:
    cinfo->out_color_components = 1;
-    if (cinfo->jpeg_color_space == JCS_GRAYSCALE ||
-        cinfo->jpeg_color_space == JCS_YCbCr) {
+    switch (cinfo->jpeg_color_space) {
+    case JCS_GRAYSCALE:
+    case JCS_YCbCr:
+    case JCS_BG_YCC:
      cconvert->pub.color_convert = grayscale_convert;
      /* For color->grayscale conversion, only the Y (0) component is needed */
      for (ci = 1; ci < cinfo->num_components; ci++)
-        cinfo->comp_info[ci].component_needed = FALSE;
-    } else if (cinfo->jpeg_color_space == JCS_RGB) {
+	cinfo->comp_info[ci].component_needed = FALSE;
+      break;
+    case JCS_RGB:
      switch (cinfo->color_transform) {
      case JCT_NONE:
-        cconvert->pub.color_convert = rgb_gray_convert;
-        break;
+	cconvert->pub.color_convert = rgb_gray_convert;
+	break;
      case JCT_SUBTRACT_GREEN:
-        cconvert->pub.color_convert = rgb1_gray_convert;
-        break;
+	cconvert->pub.color_convert = rgb1_gray_convert;
+	break;
      default:
-        ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
-        break;
+	ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
      }
      build_rgb_y_table(cinfo);
-    } else
+      break;
+    default:
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+    }
    break;

  case JCS_RGB:
    cinfo->out_color_components = RGB_PIXELSIZE;
-    if (cinfo->jpeg_color_space == JCS_YCbCr) {
+    switch (cinfo->jpeg_color_space) {
+    case JCS_GRAYSCALE:
+      cconvert->pub.color_convert = gray_rgb_convert;
+      break;
+    case JCS_YCbCr:
      cconvert->pub.color_convert = ycc_rgb_convert;
      build_ycc_rgb_table(cinfo);
-    } else if (cinfo->jpeg_color_space == JCS_GRAYSCALE) {
-      cconvert->pub.color_convert = gray_rgb_convert;
-    } else if (cinfo->jpeg_color_space == JCS_RGB) {
+      break;
+    case JCS_BG_YCC:
+      cconvert->pub.color_convert = ycc_rgb_convert;
+      build_bg_ycc_rgb_table(cinfo);
+      break;
+    case JCS_RGB:
      switch (cinfo->color_transform) {
      case JCT_NONE:
-        cconvert->pub.color_convert = rgb_convert;
-        break;
+	cconvert->pub.color_convert = rgb_convert;
+	break;
      case JCT_SUBTRACT_GREEN:
-        cconvert->pub.color_convert = rgb1_rgb_convert;
-        break;
+	cconvert->pub.color_convert = rgb1_rgb_convert;
+	break;
      default:
-        ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
-        break;
+	ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+      }
+      break;
+    default:
+      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+    }
+    break;
+
+  case JCS_BG_RGB:
+    cinfo->out_color_components = RGB_PIXELSIZE;
+    if (cinfo->jpeg_color_space == JCS_BG_RGB) {
+      switch (cinfo->color_transform) {
+      case JCT_NONE:
+	cconvert->pub.color_convert = rgb_convert;
+	break;
+      case JCT_SUBTRACT_GREEN:
+	cconvert->pub.color_convert = rgb1_rgb_convert;
+	break;
+      default:
+	ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
      }
    } else
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
@@ -592,13 +695,17 @@ jinit_color_deconverter (j_decompress_ptr cinfo)

  case JCS_CMYK:
    cinfo->out_color_components = 4;
-    if (cinfo->jpeg_color_space == JCS_YCCK) {
+    switch (cinfo->jpeg_color_space) {
+    case JCS_YCCK:
      cconvert->pub.color_convert = ycck_cmyk_convert;
      build_ycc_rgb_table(cinfo);
-    } else if (cinfo->jpeg_color_space == JCS_CMYK) {
+      break;
+    case JCS_CMYK:
      cconvert->pub.color_convert = null_convert;
-    } else
+      break;
+    default:
      ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL);
+    }
    break;

  default:
@@ -2,13 +2,14 @@
 * jdct.h
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
+ * Modified 2002-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
 * This include file contains common declarations for the forward and
 * inverse DCT modules.  These declarations are private to the DCT managers
 * (jcdctmgr.c, jddctmgr.c) and the individual DCT algorithms.
- * The individual DCT algorithms are kept in separate files to ease
+ * The individual DCT algorithms are kept in separate files to ease 
 * machine-dependent tuning (e.g., assembly coding).
 */

@@ -38,11 +39,11 @@ typedef INT32 DCTELEM;		/* must have 32 bits */
 #endif

 typedef JMETHOD(void, forward_DCT_method_ptr, (DCTELEM * data,
-                                               JSAMPARRAY sample_data,
-                                               JDIMENSION start_col));
+					       JSAMPARRAY sample_data,
+					       JDIMENSION start_col));
 typedef JMETHOD(void, float_DCT_method_ptr, (FAST_FLOAT * data,
-                                             JSAMPARRAY sample_data,
-                                             JDIMENSION start_col));
+					     JSAMPARRAY sample_data,
+					     JDIMENSION start_col));


 /*
@@ -78,13 +79,16 @@ typedef FAST_FLOAT FLOAT_MULT_TYPE; /* preferred floating type */
 * converting them to unsigned form (0..MAXJSAMPLE).  The raw outputs could
 * be quite far out of range if the input data is corrupt, so a bulletproof
 * range-limiting step is required.  We use a mask-and-table-lookup method
- * to do the combined operations quickly.  See the comments with
- * prepare_range_limit_table (in jdmaster.c) for more info.
+ * to do the combined operations quickly, assuming that MAXJSAMPLE+1
+ * is a power of 2.  See the comments with prepare_range_limit_table
+ * (in jdmaster.c) for more info.
 */

-#define IDCT_range_limit(cinfo)  ((cinfo)->sample_range_limit + CENTERJSAMPLE)
-
 #define RANGE_MASK  (MAXJSAMPLE * 4 + 3) /* 2 bits wider than legal samples */
+#define RANGE_CENTER  (MAXJSAMPLE * 2 + 2)
+#define RANGE_SUBSET  (RANGE_CENTER - CENTERJSAMPLE)
+
+#define IDCT_range_limit(cinfo)  ((cinfo)->sample_range_limit - RANGE_SUBSET)


 /* Short forms of external names for systems with brain-damaged linkers. */
@@ -233,106 +237,106 @@ EXTERN(void) jpeg_fdct_1x2

 EXTERN(void) jpeg_idct_islow
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_ifast
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_float
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_7x7
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_6x6
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_5x5
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_4x4
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_3x3
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_2x2
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_1x1
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_9x9
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_10x10
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_11x11
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_12x12
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_13x13
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_14x14
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_15x15
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_16x16
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_16x8
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_14x7
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_12x6
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_10x5
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_8x4
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_6x3
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_4x2
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_2x1
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_8x16
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_7x14
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_6x12
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_5x10
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_4x8
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_3x6
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_2x4
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
 EXTERN(void) jpeg_idct_1x2
    JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
-         JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
+	 JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));


 /*
@@ -391,3 +395,23 @@ EXTERN(void) jpeg_idct_1x2
 #ifndef MULTIPLY16V16		/* default definition */
 #define MULTIPLY16V16(var1,var2)  ((var1) * (var2))
 #endif
+
+/* Like RIGHT_SHIFT, but applies to a DCTELEM.
+ * We assume that int right shift is unsigned if INT32 right shift is.
+ */
+
+#ifdef RIGHT_SHIFT_IS_UNSIGNED
+#define ISHIFT_TEMPS	DCTELEM ishift_temp;
+#if BITS_IN_JSAMPLE == 8
+#define DCTELEMBITS  16		/* DCTELEM may be 16 or 32 bits */
+#else
+#define DCTELEMBITS  32		/* DCTELEM must be 32 bits */
+#endif
+#define IRIGHT_SHIFT(x,shft)  \
+    ((ishift_temp = (x)) < 0 ? \
+     (ishift_temp >> (shft)) | ((~((DCTELEM) 0)) << (DCTELEMBITS-(shft))) : \
+     (ishift_temp >> (shft)))
+#else
+#define ISHIFT_TEMPS
+#define IRIGHT_SHIFT(x,shft)	((x) >> (shft))
+#endif
@@ -2,7 +2,7 @@
 * jddctmgr.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
- * Modified 2002-2010 by Guido Vollbeding.
+ * Modified 2002-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -230,30 +230,30 @@ start_pass (j_decompress_ptr cinfo)
      switch (cinfo->dct_method) {
 #ifdef DCT_ISLOW_SUPPORTED
      case JDCT_ISLOW:
-        method_ptr = jpeg_idct_islow;
-        method = JDCT_ISLOW;
-        break;
+	method_ptr = jpeg_idct_islow;
+	method = JDCT_ISLOW;
+	break;
 #endif
 #ifdef DCT_IFAST_SUPPORTED
      case JDCT_IFAST:
-        method_ptr = jpeg_idct_ifast;
-        method = JDCT_IFAST;
-        break;
+	method_ptr = jpeg_idct_ifast;
+	method = JDCT_IFAST;
+	break;
 #endif
 #ifdef DCT_FLOAT_SUPPORTED
      case JDCT_FLOAT:
-        method_ptr = jpeg_idct_float;
-        method = JDCT_FLOAT;
-        break;
+	method_ptr = jpeg_idct_float;
+	method = JDCT_FLOAT;
+	break;
 #endif
      default:
-        ERREXIT(cinfo, JERR_NOT_COMPILED);
-        break;
+	ERREXIT(cinfo, JERR_NOT_COMPILED);
+	break;
      }
      break;
    default:
      ERREXIT2(cinfo, JERR_BAD_DCTSIZE,
-               compptr->DCT_h_scaled_size, compptr->DCT_v_scaled_size);
+	       compptr->DCT_h_scaled_size, compptr->DCT_v_scaled_size);
      break;
    }
    idct->pub.inverse_DCT[ci] = method_ptr;
@@ -274,75 +274,75 @@ start_pass (j_decompress_ptr cinfo)
 #ifdef PROVIDE_ISLOW_TABLES
    case JDCT_ISLOW:
      {
-        /* For LL&M IDCT method, multipliers are equal to raw quantization
-         * coefficients, but are stored as ints to ensure access efficiency.
-         */
-        ISLOW_MULT_TYPE * ismtbl = (ISLOW_MULT_TYPE *) compptr->dct_table;
-        for (i = 0; i < DCTSIZE2; i++) {
-          ismtbl[i] = (ISLOW_MULT_TYPE) qtbl->quantval[i];
-        }
+	/* For LL&M IDCT method, multipliers are equal to raw quantization
+	 * coefficients, but are stored as ints to ensure access efficiency.
+	 */
+	ISLOW_MULT_TYPE * ismtbl = (ISLOW_MULT_TYPE *) compptr->dct_table;
+	for (i = 0; i < DCTSIZE2; i++) {
+	  ismtbl[i] = (ISLOW_MULT_TYPE) qtbl->quantval[i];
+	}
      }
      break;
 #endif
 #ifdef DCT_IFAST_SUPPORTED
    case JDCT_IFAST:
      {
-        /* For AA&N IDCT method, multipliers are equal to quantization
-         * coefficients scaled by scalefactor[row]*scalefactor[col], where
-         *   scalefactor[0] = 1
-         *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
-         * For integer operation, the multiplier table is to be scaled by
-         * IFAST_SCALE_BITS.
-         */
-        IFAST_MULT_TYPE * ifmtbl = (IFAST_MULT_TYPE *) compptr->dct_table;
+	/* For AA&N IDCT method, multipliers are equal to quantization
+	 * coefficients scaled by scalefactor[row]*scalefactor[col], where
+	 *   scalefactor[0] = 1
+	 *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
+	 * For integer operation, the multiplier table is to be scaled by
+	 * IFAST_SCALE_BITS.
+	 */
+	IFAST_MULT_TYPE * ifmtbl = (IFAST_MULT_TYPE *) compptr->dct_table;
 #define CONST_BITS 14
-        static const INT16 aanscales[DCTSIZE2] = {
-          /* precomputed values scaled up by 14 bits */
-          16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
-          22725, 31521, 29692, 26722, 22725, 17855, 12299,  6270,
-          21407, 29692, 27969, 25172, 21407, 16819, 11585,  5906,
-          19266, 26722, 25172, 22654, 19266, 15137, 10426,  5315,
-          16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
-          12873, 17855, 16819, 15137, 12873, 10114,  6967,  3552,
-           8867, 12299, 11585, 10426,  8867,  6967,  4799,  2446,
-           4520,  6270,  5906,  5315,  4520,  3552,  2446,  1247
-        };
-        SHIFT_TEMPS
+	static const INT16 aanscales[DCTSIZE2] = {
+	  /* precomputed values scaled up by 14 bits */
+	  16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
+	  22725, 31521, 29692, 26722, 22725, 17855, 12299,  6270,
+	  21407, 29692, 27969, 25172, 21407, 16819, 11585,  5906,
+	  19266, 26722, 25172, 22654, 19266, 15137, 10426,  5315,
+	  16384, 22725, 21407, 19266, 16384, 12873,  8867,  4520,
+	  12873, 17855, 16819, 15137, 12873, 10114,  6967,  3552,
+	   8867, 12299, 11585, 10426,  8867,  6967,  4799,  2446,
+	   4520,  6270,  5906,  5315,  4520,  3552,  2446,  1247
+	};
+	SHIFT_TEMPS

-        for (i = 0; i < DCTSIZE2; i++) {
-          ifmtbl[i] = (IFAST_MULT_TYPE)
-            DESCALE(MULTIPLY16V16((INT32) qtbl->quantval[i],
-                                  (INT32) aanscales[i]),
-                    CONST_BITS-IFAST_SCALE_BITS);
-        }
+	for (i = 0; i < DCTSIZE2; i++) {
+	  ifmtbl[i] = (IFAST_MULT_TYPE)
+	    DESCALE(MULTIPLY16V16((INT32) qtbl->quantval[i],
+				  (INT32) aanscales[i]),
+		    CONST_BITS-IFAST_SCALE_BITS);
+	}
      }
      break;
 #endif
 #ifdef DCT_FLOAT_SUPPORTED
    case JDCT_FLOAT:
      {
-        /* For float AA&N IDCT method, multipliers are equal to quantization
-         * coefficients scaled by scalefactor[row]*scalefactor[col], where
-         *   scalefactor[0] = 1
-         *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
-         * We apply a further scale factor of 1/8.
-         */
-        FLOAT_MULT_TYPE * fmtbl = (FLOAT_MULT_TYPE *) compptr->dct_table;
-        int row, col;
-        static const double aanscalefactor[DCTSIZE] = {
-          1.0, 1.387039845, 1.306562965, 1.175875602,
-          1.0, 0.785694958, 0.541196100, 0.275899379
-        };
+	/* For float AA&N IDCT method, multipliers are equal to quantization
+	 * coefficients scaled by scalefactor[row]*scalefactor[col], where
+	 *   scalefactor[0] = 1
+	 *   scalefactor[k] = cos(k*PI/16) * sqrt(2)    for k=1..7
+	 * We apply a further scale factor of 1/8.
+	 */
+	FLOAT_MULT_TYPE * fmtbl = (FLOAT_MULT_TYPE *) compptr->dct_table;
+	int row, col;
+	static const double aanscalefactor[DCTSIZE] = {
+	  1.0, 1.387039845, 1.306562965, 1.175875602,
+	  1.0, 0.785694958, 0.541196100, 0.275899379
+	};

-        i = 0;
-        for (row = 0; row < DCTSIZE; row++) {
-          for (col = 0; col < DCTSIZE; col++) {
-            fmtbl[i] = (FLOAT_MULT_TYPE)
-              ((double) qtbl->quantval[i] *
-               aanscalefactor[row] * aanscalefactor[col] * 0.125);
-            i++;
-          }
-        }
+	i = 0;
+	for (row = 0; row < DCTSIZE; row++) {
+	  for (col = 0; col < DCTSIZE; col++) {
+	    fmtbl[i] = (FLOAT_MULT_TYPE)
+	      ((double) qtbl->quantval[i] *
+	       aanscalefactor[row] * aanscalefactor[col] * 0.125);
+	    i++;
+	  }
+	}
      }
      break;
 #endif
@@ -367,8 +367,8 @@ jinit_inverse_dct (j_decompress_ptr cinfo)

  idct = (my_idct_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_idct_controller));
-  cinfo->idct = (struct jpeg_inverse_dct *) idct;
+				SIZEOF(my_idct_controller));
+  cinfo->idct = &idct->pub;
  idct->pub.start_pass = start_pass;

  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
@@ -376,7 +376,7 @@ jinit_inverse_dct (j_decompress_ptr cinfo)
    /* Allocate and pre-zero a multiplier table for each component */
    compptr->dct_table =
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  SIZEOF(multiplier_table));
+				  SIZEOF(multiplier_table));
    MEMZERO(compptr->dct_table, SIZEOF(multiplier_table));
    /* Mark multiplier table not yet set up for any method */
    idct->cur_method[ci] = -1;
@@ -2,7 +2,7 @@
 * jdinput.c
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
- * Modified 2002-2009 by Guido Vollbeding.
+ * Modified 2002-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -196,7 +196,7 @@ jpeg_core_output_dimensions (j_decompress_ptr cinfo)
  /* Hardwire it to "no scaling" */
  cinfo->output_width = cinfo->image_width;
  cinfo->output_height = cinfo->image_height;
-  /* jdinput.c has already initialized DCT_scaled_size,
+  /* initial_setup has already initialized DCT_scaled_size,
   * and has computed unscaled downsampled_width and downsampled_height.
   */

@@ -216,14 +216,14 @@ initial_setup (j_decompress_ptr cinfo)
      (long) cinfo->image_width > (long) JPEG_MAX_DIMENSION)
    ERREXIT1(cinfo, JERR_IMAGE_TOO_BIG, (unsigned int) JPEG_MAX_DIMENSION);

-  /* For now, precision must match compiled-in value... */
-  if (cinfo->data_precision != BITS_IN_JSAMPLE)
+  /* Only 8 to 12 bits data precision are supported for DCT based JPEG */
+  if (cinfo->data_precision < 8 || cinfo->data_precision > 12)
    ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision);

  /* Check that number of components won't exceed internal array sizes */
  if (cinfo->num_components > MAX_COMPONENTS)
    ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components,
-             MAX_COMPONENTS);
+	     MAX_COMPONENTS);

  /* Compute maximum sampling factors; check factor validity */
  cinfo->max_h_samp_factor = 1;
@@ -231,12 +231,12 @@ initial_setup (j_decompress_ptr cinfo)
  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
       ci++, compptr++) {
    if (compptr->h_samp_factor<=0 || compptr->h_samp_factor>MAX_SAMP_FACTOR ||
-        compptr->v_samp_factor<=0 || compptr->v_samp_factor>MAX_SAMP_FACTOR)
+	compptr->v_samp_factor<=0 || compptr->v_samp_factor>MAX_SAMP_FACTOR)
      ERREXIT(cinfo, JERR_BAD_SAMPLING);
    cinfo->max_h_samp_factor = MAX(cinfo->max_h_samp_factor,
-                                   compptr->h_samp_factor);
+				   compptr->h_samp_factor);
    cinfo->max_v_samp_factor = MAX(cinfo->max_v_samp_factor,
-                                   compptr->v_samp_factor);
+				   compptr->v_samp_factor);
  }

  /* Derive block_size, natural_order, and lim_Se */
@@ -329,7 +329,7 @@ initial_setup (j_decompress_ptr cinfo)
      break;
    default:
      ERREXIT4(cinfo, JERR_BAD_PROGRESSION,
-               cinfo->Ss, cinfo->Se, cinfo->Ah, cinfo->Al);
+	       cinfo->Ss, cinfo->Se, cinfo->Ah, cinfo->Al);
      break;
    }

@@ -350,10 +350,10 @@ initial_setup (j_decompress_ptr cinfo)
    /* Size in DCT blocks */
    compptr->width_in_blocks = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_width * (long) compptr->h_samp_factor,
-                    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
+		    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
    compptr->height_in_blocks = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_height * (long) compptr->v_samp_factor,
-                    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+		    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
    /* downsampled_width and downsampled_height will also be overridden by
     * jdmaster.c if we are doing full decompression.  The transcoder library
     * doesn't use these values, but the calling application might.
@@ -361,10 +361,10 @@ initial_setup (j_decompress_ptr cinfo)
    /* Size in samples */
    compptr->downsampled_width = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_width * (long) compptr->h_samp_factor,
-                    (long) cinfo->max_h_samp_factor);
+		    (long) cinfo->max_h_samp_factor);
    compptr->downsampled_height = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_height * (long) compptr->v_samp_factor,
-                    (long) cinfo->max_v_samp_factor);
+		    (long) cinfo->max_v_samp_factor);
    /* Mark component needed, until color conversion says otherwise */
    compptr->component_needed = TRUE;
    /* Mark no quantization table yet saved for component */
@@ -374,7 +374,7 @@ initial_setup (j_decompress_ptr cinfo)
  /* Compute number of fully interleaved MCU rows. */
  cinfo->total_iMCU_rows = (JDIMENSION)
    jdiv_round_up((long) cinfo->image_height,
-                  (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+	          (long) (cinfo->max_v_samp_factor * cinfo->block_size));

  /* Decide whether file contains multiple scans */
  if (cinfo->comps_in_scan < cinfo->num_components || cinfo->progressive_mode)
@@ -391,16 +391,16 @@ per_scan_setup (j_decompress_ptr cinfo)
 {
  int ci, mcublks, tmp;
  jpeg_component_info *compptr;
-
+  
  if (cinfo->comps_in_scan == 1) {
-
+    
    /* Noninterleaved (single-component) scan */
    compptr = cinfo->cur_comp_info[0];
-
+    
    /* Overall image size in MCUs */
    cinfo->MCUs_per_row = compptr->width_in_blocks;
    cinfo->MCU_rows_in_scan = compptr->height_in_blocks;
-
+    
    /* For noninterleaved scan, always one block per MCU */
    compptr->MCU_width = 1;
    compptr->MCU_height = 1;
@@ -413,28 +413,28 @@ per_scan_setup (j_decompress_ptr cinfo)
    tmp = (int) (compptr->height_in_blocks % compptr->v_samp_factor);
    if (tmp == 0) tmp = compptr->v_samp_factor;
    compptr->last_row_height = tmp;
-
+    
    /* Prepare array describing MCU composition */
    cinfo->blocks_in_MCU = 1;
    cinfo->MCU_membership[0] = 0;
-
+    
  } else {
-
+    
    /* Interleaved (multi-component) scan */
    if (cinfo->comps_in_scan <= 0 || cinfo->comps_in_scan > MAX_COMPS_IN_SCAN)
      ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->comps_in_scan,
-               MAX_COMPS_IN_SCAN);
-
+	       MAX_COMPS_IN_SCAN);
+    
    /* Overall image size in MCUs */
    cinfo->MCUs_per_row = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_width,
-                    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
+		    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
    cinfo->MCU_rows_in_scan = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_height,
-                    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
-
+		    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+    
    cinfo->blocks_in_MCU = 0;
-
+    
    for (ci = 0; ci < cinfo->comps_in_scan; ci++) {
      compptr = cinfo->cur_comp_info[ci];
      /* Sampling factors give # of blocks of component in each MCU */
@@ -452,12 +452,12 @@ per_scan_setup (j_decompress_ptr cinfo)
      /* Prepare array describing MCU composition */
      mcublks = compptr->MCU_blocks;
      if (cinfo->blocks_in_MCU + mcublks > D_MAX_BLOCKS_IN_MCU)
-        ERREXIT(cinfo, JERR_BAD_MCU_SIZE);
+	ERREXIT(cinfo, JERR_BAD_MCU_SIZE);
      while (mcublks-- > 0) {
-        cinfo->MCU_membership[cinfo->blocks_in_MCU++] = ci;
+	cinfo->MCU_membership[cinfo->blocks_in_MCU++] = ci;
      }
    }
-
+    
  }
 }

@@ -498,12 +498,12 @@ latch_quant_tables (j_decompress_ptr cinfo)
    /* Make sure specified quantization table is present */
    qtblno = compptr->quant_tbl_no;
    if (qtblno < 0 || qtblno >= NUM_QUANT_TBLS ||
-        cinfo->quant_tbl_ptrs[qtblno] == NULL)
+	cinfo->quant_tbl_ptrs[qtblno] == NULL)
      ERREXIT1(cinfo, JERR_NO_QUANT_TABLE, qtblno);
    /* OK, save away the quantization table */
    qtbl = (JQUANT_TBL *)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  SIZEOF(JQUANT_TBL));
+				  SIZEOF(JQUANT_TBL));
    MEMCOPY(qtbl, cinfo->quant_tbl_ptrs[qtblno], SIZEOF(JQUANT_TBL));
    compptr->quant_table = qtbl;
  }
@@ -537,6 +537,7 @@ start_input_pass (j_decompress_ptr cinfo)
 METHODDEF(void)
 finish_input_pass (j_decompress_ptr cinfo)
 {
+  (*cinfo->entropy->finish_pass) (cinfo);
  cinfo->inputctl->consume_input = consume_markers;
 }

@@ -570,36 +571,36 @@ consume_markers (j_decompress_ptr cinfo)
    switch (val) {
    case JPEG_REACHED_SOS:	/* Found SOS */
      if (inputctl->inheaders) { /* 1st SOS */
-        if (inputctl->inheaders == 1)
-          initial_setup(cinfo);
-        if (cinfo->comps_in_scan == 0) { /* pseudo SOS marker */
-          inputctl->inheaders = 2;
-          break;
-        }
-        inputctl->inheaders = 0;
-        /* Note: start_input_pass must be called by jdmaster.c
-         * before any more input can be consumed.  jdapimin.c is
-         * responsible for enforcing this sequencing.
-         */
+	if (inputctl->inheaders == 1)
+	  initial_setup(cinfo);
+	if (cinfo->comps_in_scan == 0) { /* pseudo SOS marker */
+	  inputctl->inheaders = 2;
+	  break;
+	}
+	inputctl->inheaders = 0;
+	/* Note: start_input_pass must be called by jdmaster.c
+	 * before any more input can be consumed.  jdapimin.c is
+	 * responsible for enforcing this sequencing.
+	 */
      } else {			/* 2nd or later SOS marker */
-        if (! inputctl->pub.has_multiple_scans)
-          ERREXIT(cinfo, JERR_EOI_EXPECTED); /* Oops, I wasn't expecting this! */
-        if (cinfo->comps_in_scan == 0) /* unexpected pseudo SOS marker */
-          break;
-        start_input_pass(cinfo);
+	if (! inputctl->pub.has_multiple_scans)
+	  ERREXIT(cinfo, JERR_EOI_EXPECTED); /* Oops, I wasn't expecting this! */
+	if (cinfo->comps_in_scan == 0) /* unexpected pseudo SOS marker */
+	  break;
+	start_input_pass(cinfo);
      }
      return val;
    case JPEG_REACHED_EOI:	/* Found EOI */
      inputctl->pub.eoi_reached = TRUE;
      if (inputctl->inheaders) { /* Tables-only datastream, apparently */
-        if (cinfo->marker->saw_SOF)
-          ERREXIT(cinfo, JERR_SOF_NO_SOS);
+	if (cinfo->marker->saw_SOF)
+	  ERREXIT(cinfo, JERR_SOF_NO_SOS);
      } else {
-        /* Prevent infinite loop in coef ctlr's decompress_data routine
-         * if user set output_scan_number larger than number of scans.
-         */
-        if (cinfo->output_scan_number > cinfo->input_scan_number)
-          cinfo->output_scan_number = cinfo->input_scan_number;
+	/* Prevent infinite loop in coef ctlr's decompress_data routine
+	 * if user set output_scan_number larger than number of scans.
+	 */
+	if (cinfo->output_scan_number > cinfo->input_scan_number)
+	  cinfo->output_scan_number = cinfo->input_scan_number;
      }
      return val;
    case JPEG_SUSPENDED:
@@ -645,8 +646,8 @@ jinit_input_controller (j_decompress_ptr cinfo)
  /* Create subobject in permanent pool */
  inputctl = (my_inputctl_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                SIZEOF(my_input_controller));
-  cinfo->inputctl = (struct jpeg_input_controller *) inputctl;
+				SIZEOF(my_input_controller));
+  cinfo->inputctl = &inputctl->pub;
  /* Initialize method pointers */
  inputctl->pub.consume_input = consume_markers;
  inputctl->pub.reset_input_controller = reset_input_controller;
@@ -142,15 +142,15 @@ typedef my_main_controller * my_main_ptr;

 /* Forward declarations */
 METHODDEF(void) process_data_simple_main
-        JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf,
-             JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail));
+	JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf,
+	     JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail));
 METHODDEF(void) process_data_context_main
-        JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf,
-             JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail));
+	JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf,
+	     JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail));
 #ifdef QUANT_2PASS_SUPPORTED
 METHODDEF(void) process_data_crank_post
-        JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf,
-             JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail));
+	JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf,
+	     JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail));
 #endif


@@ -171,7 +171,7 @@ alloc_funny_pointers (j_decompress_ptr cinfo)
   */
  mainp->xbuffer[0] = (JSAMPIMAGE)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                cinfo->num_components * 2 * SIZEOF(JSAMPARRAY));
+				cinfo->num_components * 2 * SIZEOF(JSAMPARRAY));
  mainp->xbuffer[1] = mainp->xbuffer[0] + cinfo->num_components;

  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
@@ -183,7 +183,7 @@ alloc_funny_pointers (j_decompress_ptr cinfo)
     */
    xbuf = (JSAMPARRAY)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  2 * (rgroup * (M + 4)) * SIZEOF(JSAMPROW));
+				  2 * (rgroup * (M + 4)) * SIZEOF(JSAMPROW));
    xbuf += rgroup;		/* want one row group at negative offsets */
    mainp->xbuffer[0][ci] = xbuf;
    xbuf += rgroup * (M + 4);
@@ -344,8 +344,8 @@ start_pass_main (j_decompress_ptr cinfo, J_BUF_MODE pass_mode)

 METHODDEF(void)
 process_data_simple_main (j_decompress_ptr cinfo,
-                          JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                          JDIMENSION out_rows_avail)
+			  JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+			  JDIMENSION out_rows_avail)
 {
  my_main_ptr mainp = (my_main_ptr) cinfo->main;
  JDIMENSION rowgroups_avail;
@@ -366,8 +366,8 @@ process_data_simple_main (j_decompress_ptr cinfo,

  /* Feed the postprocessor */
  (*cinfo->post->post_process_data) (cinfo, mainp->buffer,
-                                     &mainp->rowgroup_ctr, rowgroups_avail,
-                                     output_buf, out_row_ctr, out_rows_avail);
+				     &mainp->rowgroup_ctr, rowgroups_avail,
+				     output_buf, out_row_ctr, out_rows_avail);

  /* Has postprocessor consumed all the data yet? If so, mark buffer empty */
  if (mainp->rowgroup_ctr >= rowgroups_avail) {
@@ -384,15 +384,15 @@ process_data_simple_main (j_decompress_ptr cinfo,

 METHODDEF(void)
 process_data_context_main (j_decompress_ptr cinfo,
-                           JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                           JDIMENSION out_rows_avail)
+			   JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+			   JDIMENSION out_rows_avail)
 {
  my_main_ptr mainp = (my_main_ptr) cinfo->main;

  /* Read input data if we haven't filled the main buffer yet */
  if (! mainp->buffer_full) {
    if (! (*cinfo->coef->decompress_data) (cinfo,
-                                           mainp->xbuffer[mainp->whichptr]))
+					   mainp->xbuffer[mainp->whichptr]))
      return;			/* suspension forced, can do nothing more */
    mainp->buffer_full = TRUE;	/* OK, we have an iMCU row to work with */
    mainp->iMCU_row_ctr++;	/* count rows received */
@@ -407,8 +407,8 @@ process_data_context_main (j_decompress_ptr cinfo,
  case CTX_POSTPONED_ROW:
    /* Call postprocessor using previously set pointers for postponed row */
    (*cinfo->post->post_process_data) (cinfo, mainp->xbuffer[mainp->whichptr],
-                        &mainp->rowgroup_ctr, mainp->rowgroups_avail,
-                        output_buf, out_row_ctr, out_rows_avail);
+			&mainp->rowgroup_ctr, mainp->rowgroups_avail,
+			output_buf, out_row_ctr, out_rows_avail);
    if (mainp->rowgroup_ctr < mainp->rowgroups_avail)
      return;			/* Need to suspend */
    mainp->context_state = CTX_PREPARE_FOR_IMCU;
@@ -429,8 +429,8 @@ process_data_context_main (j_decompress_ptr cinfo,
  case CTX_PROCESS_IMCU:
    /* Call postprocessor using previously set pointers */
    (*cinfo->post->post_process_data) (cinfo, mainp->xbuffer[mainp->whichptr],
-                        &mainp->rowgroup_ctr, mainp->rowgroups_avail,
-                        output_buf, out_row_ctr, out_rows_avail);
+			&mainp->rowgroup_ctr, mainp->rowgroups_avail,
+			output_buf, out_row_ctr, out_rows_avail);
    if (mainp->rowgroup_ctr < mainp->rowgroups_avail)
      return;			/* Need to suspend */
    /* After the first iMCU, change wraparound pointers to normal state */
@@ -458,12 +458,12 @@ process_data_context_main (j_decompress_ptr cinfo,

 METHODDEF(void)
 process_data_crank_post (j_decompress_ptr cinfo,
-                         JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                         JDIMENSION out_rows_avail)
+			 JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+			 JDIMENSION out_rows_avail)
 {
  (*cinfo->post->post_process_data) (cinfo, (JSAMPIMAGE) NULL,
-                                     (JDIMENSION *) NULL, (JDIMENSION) 0,
-                                     output_buf, out_row_ctr, out_rows_avail);
+				     (JDIMENSION *) NULL, (JDIMENSION) 0,
+				     output_buf, out_row_ctr, out_rows_avail);
 }

 #endif /* QUANT_2PASS_SUPPORTED */
@@ -482,7 +482,7 @@ jinit_d_main_controller (j_decompress_ptr cinfo, boolean need_full_buffer)

  mainp = (my_main_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_main_controller));
+				SIZEOF(my_main_controller));
  cinfo->main = &mainp->pub;
  mainp->pub.start_pass = start_pass_main;

@@ -2,7 +2,7 @@
 * jdmarker.c
 *
 * Copyright (C) 1991-1998, Thomas G. Lane.
- * Modified 2009-2012 by Guido Vollbeding.
+ * Modified 2009-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -119,49 +119,49 @@ typedef my_marker_reader * my_marker_ptr;

 /* Declare and initialize local copies of input pointer/count */
 #define INPUT_VARS(cinfo)  \
-        struct jpeg_source_mgr * datasrc = (cinfo)->src;  \
-        const JOCTET * next_input_byte = datasrc->next_input_byte;  \
-        size_t bytes_in_buffer = datasrc->bytes_in_buffer
+	struct jpeg_source_mgr * datasrc = (cinfo)->src;  \
+	const JOCTET * next_input_byte = datasrc->next_input_byte;  \
+	size_t bytes_in_buffer = datasrc->bytes_in_buffer

 /* Unload the local copies --- do this only at a restart boundary */
 #define INPUT_SYNC(cinfo)  \
-        ( datasrc->next_input_byte = next_input_byte,  \
-          datasrc->bytes_in_buffer = bytes_in_buffer )
+	( datasrc->next_input_byte = next_input_byte,  \
+	  datasrc->bytes_in_buffer = bytes_in_buffer )

 /* Reload the local copies --- used only in MAKE_BYTE_AVAIL */
 #define INPUT_RELOAD(cinfo)  \
-        ( next_input_byte = datasrc->next_input_byte,  \
-          bytes_in_buffer = datasrc->bytes_in_buffer )
+	( next_input_byte = datasrc->next_input_byte,  \
+	  bytes_in_buffer = datasrc->bytes_in_buffer )

 /* Internal macro for INPUT_BYTE and INPUT_2BYTES: make a byte available.
 * Note we do *not* do INPUT_SYNC before calling fill_input_buffer,
 * but we must reload the local copies after a successful fill.
 */
 #define MAKE_BYTE_AVAIL(cinfo,action)  \
-        if (bytes_in_buffer == 0) {  \
-          if (! (*datasrc->fill_input_buffer) (cinfo))  \
-            { action; }  \
-          INPUT_RELOAD(cinfo);  \
-        }
+	if (bytes_in_buffer == 0) {  \
+	  if (! (*datasrc->fill_input_buffer) (cinfo))  \
+	    { action; }  \
+	  INPUT_RELOAD(cinfo);  \
+	}

 /* Read a byte into variable V.
 * If must suspend, take the specified action (typically "return FALSE").
 */
 #define INPUT_BYTE(cinfo,V,action)  \
-        MAKESTMT( MAKE_BYTE_AVAIL(cinfo,action); \
-                  bytes_in_buffer--; \
-                  V = GETJOCTET(*next_input_byte++); )
+	MAKESTMT( MAKE_BYTE_AVAIL(cinfo,action); \
+		  bytes_in_buffer--; \
+		  V = GETJOCTET(*next_input_byte++); )

 /* As above, but read two bytes interpreted as an unsigned 16-bit integer.
 * V should be declared unsigned int or perhaps INT32.
 */
 #define INPUT_2BYTES(cinfo,V,action)  \
-        MAKESTMT( MAKE_BYTE_AVAIL(cinfo,action); \
-                  bytes_in_buffer--; \
-                  V = ((unsigned int) GETJOCTET(*next_input_byte++)) << 8; \
-                  MAKE_BYTE_AVAIL(cinfo,action); \
-                  bytes_in_buffer--; \
-                  V += GETJOCTET(*next_input_byte++); )
+	MAKESTMT( MAKE_BYTE_AVAIL(cinfo,action); \
+		  bytes_in_buffer--; \
+		  V = ((unsigned int) GETJOCTET(*next_input_byte++)) << 8; \
+		  MAKE_BYTE_AVAIL(cinfo,action); \
+		  bytes_in_buffer--; \
+		  V += GETJOCTET(*next_input_byte++); )


 /*
@@ -200,7 +200,7 @@ get_soi (j_decompress_ptr cinfo)
 /* Process an SOI marker */
 {
  int i;
-
+  
  TRACEMS(cinfo, 1, JTRC_SOI);

  if (cinfo->marker->saw_SOI)
@@ -238,7 +238,7 @@ get_soi (j_decompress_ptr cinfo)

 LOCAL(boolean)
 get_sof (j_decompress_ptr cinfo, boolean is_baseline, boolean is_prog,
-         boolean is_arith)
+	 boolean is_arith)
 /* Process a SOFn marker */
 {
  INT32 length;
@@ -260,8 +260,8 @@ get_sof (j_decompress_ptr cinfo, boolean is_baseline, boolean is_prog,
  length -= 8;

  TRACEMS4(cinfo, 1, JTRC_SOF, cinfo->unread_marker,
-           (int) cinfo->image_width, (int) cinfo->image_height,
-           cinfo->num_components);
+	   (int) cinfo->image_width, (int) cinfo->image_height,
+	   cinfo->num_components);

  if (cinfo->marker->saw_SOF)
    ERREXIT(cinfo, JERR_SOF_DUPLICATE);
@@ -269,8 +269,8 @@ get_sof (j_decompress_ptr cinfo, boolean is_baseline, boolean is_prog,
  /* We don't support files in which the image height is initially specified */
  /* as 0 and is later redefined by DNL.  As long as we have to check that,  */
  /* might as well have a general sanity check. */
-  if (cinfo->image_height <= 0 || cinfo->image_width <= 0
-      || cinfo->num_components <= 0)
+  if (cinfo->image_height <= 0 || cinfo->image_width <= 0 ||
+      cinfo->num_components <= 0)
    ERREXIT(cinfo, JERR_EMPTY_IMAGE);

  if (length != (cinfo->num_components * 3))
@@ -278,8 +278,8 @@ get_sof (j_decompress_ptr cinfo, boolean is_baseline, boolean is_prog,

  if (cinfo->comp_info == NULL)	/* do only once, even if suspend */
    cinfo->comp_info = (jpeg_component_info *) (*cinfo->mem->alloc_small)
-                        ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                         cinfo->num_components * SIZEOF(jpeg_component_info));
+			((j_common_ptr) cinfo, JPOOL_IMAGE,
+			 cinfo->num_components * SIZEOF(jpeg_component_info));

  for (ci = 0; ci < cinfo->num_components; ci++) {
    INPUT_BYTE(cinfo, c, return FALSE);
@@ -289,14 +289,14 @@ get_sof (j_decompress_ptr cinfo, boolean is_baseline, boolean is_prog,
    /* max id seen so far + 1. */
    for (i = 0, compptr = cinfo->comp_info; i < ci; i++, compptr++) {
      if (c == compptr->component_id) {
-        compptr = cinfo->comp_info;
-        c = compptr->component_id;
-        compptr++;
-        for (i = 1; i < ci; i++, compptr++) {
-          if (compptr->component_id > c) c = compptr->component_id;
-        }
-        c++;
-        break;
+	compptr = cinfo->comp_info;
+	c = compptr->component_id;
+	compptr++;
+	for (i = 1; i < ci; i++, compptr++) {
+	  if (compptr->component_id > c) c = compptr->component_id;
+	}
+	c++;
+	break;
      }
    }
    compptr->component_id = c;
@@ -307,8 +307,8 @@ get_sof (j_decompress_ptr cinfo, boolean is_baseline, boolean is_prog,
    INPUT_BYTE(cinfo, compptr->quant_tbl_no, return FALSE);

    TRACEMS4(cinfo, 1, JTRC_SOF_COMPONENT,
-             compptr->component_id, compptr->h_samp_factor,
-             compptr->v_samp_factor, compptr->quant_tbl_no);
+	     compptr->component_id, compptr->h_samp_factor,
+	     compptr->v_samp_factor, compptr->quant_tbl_no);
  }

  cinfo->marker->saw_SOF = TRUE;
@@ -350,22 +350,25 @@ get_sos (j_decompress_ptr cinfo)

    /* Detect the case where component id's are not unique, and, if so, */
    /* create a fake component id using the same logic as in get_sof.   */
+    /* Note:  This also ensures that all of the SOF components are      */
+    /* referenced in the single scan case, which prevents access to     */
+    /* uninitialized memory in later decoding stages. */
    for (ci = 0; ci < i; ci++) {
      if (c == cinfo->cur_comp_info[ci]->component_id) {
-        c = cinfo->cur_comp_info[0]->component_id;
-        for (ci = 1; ci < i; ci++) {
-          compptr = cinfo->cur_comp_info[ci];
-          if (compptr->component_id > c) c = compptr->component_id;
-        }
-        c++;
-        break;
+	c = cinfo->cur_comp_info[0]->component_id;
+	for (ci = 1; ci < i; ci++) {
+	  compptr = cinfo->cur_comp_info[ci];
+	  if (compptr->component_id > c) c = compptr->component_id;
+	}
+	c++;
+	break;
      }
    }

    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      if (c == compptr->component_id)
-        goto id_found;
+	goto id_found;
    }

    ERREXIT1(cinfo, JERR_BAD_COMPONENT_ID, c);
@@ -378,7 +381,7 @@ get_sos (j_decompress_ptr cinfo)
    compptr->ac_tbl_no = (c     ) & 15;

    TRACEMS3(cinfo, 1, JTRC_SOS_COMPONENT, compptr->component_id,
-             compptr->dc_tbl_no, compptr->ac_tbl_no);
+	     compptr->dc_tbl_no, compptr->ac_tbl_no);
  }

  /* Collect the additional scan parameters Ss, Se, Ah/Al. */
@@ -391,7 +394,7 @@ get_sos (j_decompress_ptr cinfo)
  cinfo->Al = (c     ) & 15;

  TRACEMS4(cinfo, 1, JTRC_SOS_PARAMS, cinfo->Ss, cinfo->Se,
-           cinfo->Ah, cinfo->Al);
+	   cinfo->Ah, cinfo->Al);

  /* Prepare to scan data & restart markers */
  cinfo->marker->next_restart_num = 0;
@@ -416,7 +419,7 @@ get_dac (j_decompress_ptr cinfo)

  INPUT_2BYTES(cinfo, length, return FALSE);
  length -= 2;
-
+  
  while (length > 0) {
    INPUT_BYTE(cinfo, index, return FALSE);
    INPUT_BYTE(cinfo, val, return FALSE);
@@ -434,7 +437,7 @@ get_dac (j_decompress_ptr cinfo)
      cinfo->arith_dc_L[index] = (UINT8) (val & 0x0F);
      cinfo->arith_dc_U[index] = (UINT8) (val >> 4);
      if (cinfo->arith_dc_L[index] > cinfo->arith_dc_U[index])
-        ERREXIT1(cinfo, JERR_DAC_VALUE, val);
+	ERREXIT1(cinfo, JERR_DAC_VALUE, val);
    }
  }

@@ -465,12 +468,12 @@ get_dht (j_decompress_ptr cinfo)

  INPUT_2BYTES(cinfo, length, return FALSE);
  length -= 2;
-
+  
  while (length > 16) {
    INPUT_BYTE(cinfo, index, return FALSE);

    TRACEMS1(cinfo, 1, JTRC_DHT, index);
-
+      
    bits[0] = 0;
    count = 0;
    for (i = 1; i <= 16; i++) {
@@ -481,11 +484,11 @@ get_dht (j_decompress_ptr cinfo)
    length -= 1 + 16;

    TRACEMS8(cinfo, 2, JTRC_HUFFBITS,
-             bits[1], bits[2], bits[3], bits[4],
-             bits[5], bits[6], bits[7], bits[8]);
+	     bits[1], bits[2], bits[3], bits[4],
+	     bits[5], bits[6], bits[7], bits[8]);
    TRACEMS8(cinfo, 2, JTRC_HUFFBITS,
-             bits[9], bits[10], bits[11], bits[12],
-             bits[13], bits[14], bits[15], bits[16]);
+	     bits[9], bits[10], bits[11], bits[12],
+	     bits[13], bits[14], bits[15], bits[16]);

    /* Here we just do minimal validation of the counts to avoid walking
     * off the end of our table space.  jdhuff.c will check more carefully.
@@ -493,6 +496,8 @@ get_dht (j_decompress_ptr cinfo)
    if (count > 256 || ((INT32) count) > length)
      ERREXIT(cinfo, JERR_BAD_HUFF_TABLE);

+    MEMZERO(huffval, SIZEOF(huffval)); /* pre-zero array for later copy */
+
    for (i = 0; i < count; i++)
      INPUT_BYTE(cinfo, huffval[i], return FALSE);

@@ -510,7 +515,7 @@ get_dht (j_decompress_ptr cinfo)

    if (*htblptr == NULL)
      *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo);
-
+  
    MEMCOPY((*htblptr)->bits, bits, SIZEOF((*htblptr)->bits));
    MEMCOPY((*htblptr)->huffval, huffval, SIZEOF((*htblptr)->huffval));
  }
@@ -547,29 +552,29 @@ get_dqt (j_decompress_ptr cinfo)

    if (n >= NUM_QUANT_TBLS)
      ERREXIT1(cinfo, JERR_DQT_INDEX, n);
-
+      
    if (cinfo->quant_tbl_ptrs[n] == NULL)
      cinfo->quant_tbl_ptrs[n] = jpeg_alloc_quant_table((j_common_ptr) cinfo);
    quant_ptr = cinfo->quant_tbl_ptrs[n];

    if (prec) {
      if (length < DCTSIZE2 * 2) {
-        /* Initialize full table for safety. */
-        for (i = 0; i < DCTSIZE2; i++) {
-          quant_ptr->quantval[i] = 1;
-        }
-        count = length >> 1;
+	/* Initialize full table for safety. */
+	for (i = 0; i < DCTSIZE2; i++) {
+	  quant_ptr->quantval[i] = 1;
+	}
+	count = length >> 1;
      } else
-        count = DCTSIZE2;
+	count = DCTSIZE2;
    } else {
      if (length < DCTSIZE2) {
-        /* Initialize full table for safety. */
-        for (i = 0; i < DCTSIZE2; i++) {
-          quant_ptr->quantval[i] = 1;
-        }
-        count = length;
+	/* Initialize full table for safety. */
+	for (i = 0; i < DCTSIZE2; i++) {
+	  quant_ptr->quantval[i] = 1;
+	}
+	count = length;
      } else
-        count = DCTSIZE2;
+	count = DCTSIZE2;
    }

    switch (count) {
@@ -584,20 +589,20 @@ get_dqt (j_decompress_ptr cinfo)

    for (i = 0; i < count; i++) {
      if (prec)
-        INPUT_2BYTES(cinfo, tmp, return FALSE);
+	INPUT_2BYTES(cinfo, tmp, return FALSE);
      else
-        INPUT_BYTE(cinfo, tmp, return FALSE);
+	INPUT_BYTE(cinfo, tmp, return FALSE);
      /* We convert the zigzag-order table to natural array order. */
      quant_ptr->quantval[natural_order[i]] = (UINT16) tmp;
    }

    if (cinfo->err->trace_level >= 2) {
      for (i = 0; i < DCTSIZE2; i += 8) {
-        TRACEMS8(cinfo, 2, JTRC_QUANTVALS,
-                 quant_ptr->quantval[i],   quant_ptr->quantval[i+1],
-                 quant_ptr->quantval[i+2], quant_ptr->quantval[i+3],
-                 quant_ptr->quantval[i+4], quant_ptr->quantval[i+5],
-                 quant_ptr->quantval[i+6], quant_ptr->quantval[i+7]);
+	TRACEMS8(cinfo, 2, JTRC_QUANTVALS,
+		 quant_ptr->quantval[i],   quant_ptr->quantval[i+1],
+		 quant_ptr->quantval[i+2], quant_ptr->quantval[i+3],
+		 quant_ptr->quantval[i+4], quant_ptr->quantval[i+5],
+		 quant_ptr->quantval[i+6], quant_ptr->quantval[i+7]);
      }
    }

@@ -622,7 +627,7 @@ get_dri (j_decompress_ptr cinfo)
  INPUT_VARS(cinfo);

  INPUT_2BYTES(cinfo, length, return FALSE);
-
+  
  if (length != 4)
    ERREXIT(cinfo, JERR_BAD_LENGTH);

@@ -713,7 +718,7 @@ get_lse (j_decompress_ptr cinfo)

 LOCAL(void)
 examine_app0 (j_decompress_ptr cinfo, JOCTET FAR * data,
-              unsigned int datalen, INT32 remaining)
+	      unsigned int datalen, INT32 remaining)
 /* Examine first few bytes from an APP0.
 * Take appropriate action if it is a JFIF marker.
 * datalen is # of bytes at data[], remaining is length of rest of marker data.
@@ -735,25 +740,26 @@ examine_app0 (j_decompress_ptr cinfo, JOCTET FAR * data,
    cinfo->X_density = (GETJOCTET(data[8]) << 8) + GETJOCTET(data[9]);
    cinfo->Y_density = (GETJOCTET(data[10]) << 8) + GETJOCTET(data[11]);
    /* Check version.
-     * Major version must be 1, anything else signals an incompatible change.
+     * Major version must be 1 or 2, anything else signals an incompatible
+     * change.
     * (We used to treat this as an error, but now it's a nonfatal warning,
     * because some bozo at Hijaak couldn't read the spec.)
     * Minor version should be 0..2, but process anyway if newer.
     */
-    if (cinfo->JFIF_major_version != 1)
+    if (cinfo->JFIF_major_version != 1 && cinfo->JFIF_major_version != 2)
      WARNMS2(cinfo, JWRN_JFIF_MAJOR,
-              cinfo->JFIF_major_version, cinfo->JFIF_minor_version);
+	      cinfo->JFIF_major_version, cinfo->JFIF_minor_version);
    /* Generate trace messages */
    TRACEMS5(cinfo, 1, JTRC_JFIF,
-             cinfo->JFIF_major_version, cinfo->JFIF_minor_version,
-             cinfo->X_density, cinfo->Y_density, cinfo->density_unit);
+	     cinfo->JFIF_major_version, cinfo->JFIF_minor_version,
+	     cinfo->X_density, cinfo->Y_density, cinfo->density_unit);
    /* Validate thumbnail dimensions and issue appropriate messages */
    if (GETJOCTET(data[12]) | GETJOCTET(data[13]))
      TRACEMS2(cinfo, 1, JTRC_JFIF_THUMBNAIL,
-               GETJOCTET(data[12]), GETJOCTET(data[13]));
+	       GETJOCTET(data[12]), GETJOCTET(data[13]));
    totallen -= APP0_DATA_LEN;
    if (totallen !=
-        ((INT32)GETJOCTET(data[12]) * (INT32)GETJOCTET(data[13]) * (INT32) 3))
+	((INT32)GETJOCTET(data[12]) * (INT32)GETJOCTET(data[13]) * (INT32) 3))
      TRACEMS1(cinfo, 1, JTRC_JFIF_BADTHUMBNAILSIZE, (int) totallen);
  } else if (datalen >= 6 &&
      GETJOCTET(data[0]) == 0x4A &&
@@ -777,7 +783,7 @@ examine_app0 (j_decompress_ptr cinfo, JOCTET FAR * data,
      break;
    default:
      TRACEMS2(cinfo, 1, JTRC_JFIF_EXTENSION,
-               GETJOCTET(data[5]), (int) totallen);
+	       GETJOCTET(data[5]), (int) totallen);
      break;
    }
  } else {
@@ -789,7 +795,7 @@ examine_app0 (j_decompress_ptr cinfo, JOCTET FAR * data,

 LOCAL(void)
 examine_app14 (j_decompress_ptr cinfo, JOCTET FAR * data,
-               unsigned int datalen, INT32 remaining)
+	       unsigned int datalen, INT32 remaining)
 /* Examine first few bytes from an APP14.
 * Take appropriate action if it is an Adobe marker.
 * datalen is # of bytes at data[], remaining is length of rest of marker data.
@@ -885,15 +891,15 @@ save_marker (j_decompress_ptr cinfo)
      /* figure out how much we want to save */
      unsigned int limit;
      if (cinfo->unread_marker == (int) M_COM)
-        limit = marker->length_limit_COM;
+	limit = marker->length_limit_COM;
      else
-        limit = marker->length_limit_APPn[cinfo->unread_marker - (int) M_APP0];
+	limit = marker->length_limit_APPn[cinfo->unread_marker - (int) M_APP0];
      if ((unsigned int) length < limit)
-        limit = (unsigned int) length;
+	limit = (unsigned int) length;
      /* allocate and initialize the marker item */
      cur_marker = (jpeg_saved_marker_ptr)
-        (*cinfo->mem->alloc_large) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                    SIZEOF(struct jpeg_marker_struct) + limit);
+	(*cinfo->mem->alloc_large) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				    SIZEOF(struct jpeg_marker_struct) + limit);
      cur_marker->next = NULL;
      cur_marker->marker = (UINT8) cinfo->unread_marker;
      cur_marker->original_length = (unsigned int) length;
@@ -937,7 +943,7 @@ save_marker (j_decompress_ptr cinfo)
    } else {
      jpeg_saved_marker_ptr prev = cinfo->marker_list;
      while (prev->next != NULL)
-        prev = prev->next;
+	prev = prev->next;
      prev->next = cur_marker;
    }
    /* Reset pointer & calc remaining data length */
@@ -957,7 +963,7 @@ save_marker (j_decompress_ptr cinfo)
    break;
  default:
    TRACEMS2(cinfo, 1, JTRC_MISC_MARKER, cinfo->unread_marker,
-             (int) (data_length + length));
+	     (int) (data_length + length));
    break;
  }

@@ -981,7 +987,7 @@ skip_variable (j_decompress_ptr cinfo)

  INPUT_2BYTES(cinfo, length, return FALSE);
  length -= 2;
-
+  
  TRACEMS2(cinfo, 1, JTRC_MISC_MARKER, cinfo->unread_marker, (int) length);

  INPUT_SYNC(cinfo);		/* do before skip_input_data */
@@ -1093,11 +1099,11 @@ read_markers (j_decompress_ptr cinfo)
    /* NB: first_marker() enforces the requirement that SOI appear first. */
    if (cinfo->unread_marker == 0) {
      if (! cinfo->marker->saw_SOI) {
-        if (! first_marker(cinfo))
-          return JPEG_SUSPENDED;
+	if (! first_marker(cinfo))
+	  return JPEG_SUSPENDED;
      } else {
-        if (! next_marker(cinfo))
-          return JPEG_SUSPENDED;
+	if (! next_marker(cinfo))
+	  return JPEG_SUSPENDED;
      }
    }
    /* At this point cinfo->unread_marker contains the marker code and the
@@ -1107,32 +1113,32 @@ read_markers (j_decompress_ptr cinfo)
    switch (cinfo->unread_marker) {
    case M_SOI:
      if (! get_soi(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_SOF0:		/* Baseline */
      if (! get_sof(cinfo, TRUE, FALSE, FALSE))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_SOF1:		/* Extended sequential, Huffman */
      if (! get_sof(cinfo, FALSE, FALSE, FALSE))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_SOF2:		/* Progressive, Huffman */
      if (! get_sof(cinfo, FALSE, TRUE, FALSE))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_SOF9:		/* Extended sequential, arithmetic */
      if (! get_sof(cinfo, FALSE, FALSE, TRUE))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_SOF10:		/* Progressive, arithmetic */
      if (! get_sof(cinfo, FALSE, TRUE, TRUE))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    /* Currently unsupported SOFn types */
@@ -1150,7 +1156,7 @@ read_markers (j_decompress_ptr cinfo)

    case M_SOS:
      if (! get_sos(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      cinfo->unread_marker = 0;	/* processed the marker */
      return JPEG_REACHED_SOS;

@@ -1161,27 +1167,27 @@ read_markers (j_decompress_ptr cinfo)

    case M_DAC:
      if (! get_dac(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_DHT:
      if (! get_dht(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_DQT:
      if (! get_dqt(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_DRI:
      if (! get_dri(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_JPG8:
      if (! get_lse(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_APP0:
@@ -1201,13 +1207,13 @@ read_markers (j_decompress_ptr cinfo)
    case M_APP14:
    case M_APP15:
      if (! (*((my_marker_ptr) cinfo->marker)->process_APPn[
-                cinfo->unread_marker - (int) M_APP0]) (cinfo))
-        return JPEG_SUSPENDED;
+		cinfo->unread_marker - (int) M_APP0]) (cinfo))
+	return JPEG_SUSPENDED;
      break;

    case M_COM:
      if (! (*((my_marker_ptr) cinfo->marker)->process_COM) (cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    case M_RST0:		/* these are all parameterless */
@@ -1224,7 +1230,7 @@ read_markers (j_decompress_ptr cinfo)

    case M_DNL:			/* Ignore DNL ... perhaps the wrong thing */
      if (! skip_variable(cinfo))
-        return JPEG_SUSPENDED;
+	return JPEG_SUSPENDED;
      break;

    default:			/* must be DHP, EXP, JPGn, or RESn */
@@ -1273,7 +1279,7 @@ read_restart_marker (j_decompress_ptr cinfo)
    /* Uh-oh, the restart markers have been messed up. */
    /* Let the data source manager determine how to resync. */
    if (! (*cinfo->src->resync_to_restart) (cinfo,
-                                            cinfo->marker->next_restart_num))
+					    cinfo->marker->next_restart_num))
      return FALSE;
  }

@@ -1338,10 +1344,10 @@ jpeg_resync_to_restart (j_decompress_ptr cinfo, int desired)
 {
  int marker = cinfo->unread_marker;
  int action = 1;
-
+  
  /* Always put up a warning. */
  WARNMS2(cinfo, JWRN_MUST_RESYNC, marker, desired);
-
+  
  /* Outer loop handles repeated decision after scanning forward. */
  for (;;) {
    if (marker < (int) M_SOF0)
@@ -1350,13 +1356,13 @@ jpeg_resync_to_restart (j_decompress_ptr cinfo, int desired)
      action = 3;		/* valid non-restart marker */
    else {
      if (marker == ((int) M_RST0 + ((desired+1) & 7)) ||
-          marker == ((int) M_RST0 + ((desired+2) & 7)))
-        action = 3;		/* one of the next two expected restarts */
+	  marker == ((int) M_RST0 + ((desired+2) & 7)))
+	action = 3;		/* one of the next two expected restarts */
      else if (marker == ((int) M_RST0 + ((desired-1) & 7)) ||
-               marker == ((int) M_RST0 + ((desired-2) & 7)))
-        action = 2;		/* a prior restart, so advance */
+	       marker == ((int) M_RST0 + ((desired-2) & 7)))
+	action = 2;		/* a prior restart, so advance */
      else
-        action = 1;		/* desired restart or too far away */
+	action = 1;		/* desired restart or too far away */
    }
    TRACEMS2(cinfo, 4, JTRC_RECOVERY_ACTION, marker, action);
    switch (action) {
@@ -1367,7 +1373,7 @@ jpeg_resync_to_restart (j_decompress_ptr cinfo, int desired)
    case 2:
      /* Scan to the next marker, and repeat the decision loop. */
      if (! next_marker(cinfo))
-        return FALSE;
+	return FALSE;
      marker = cinfo->unread_marker;
      break;
    case 3:
@@ -1412,7 +1418,7 @@ jinit_marker_reader (j_decompress_ptr cinfo)
  /* Create subobject in permanent pool */
  marker = (my_marker_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT,
-                                SIZEOF(my_marker_reader));
+				SIZEOF(my_marker_reader));
  cinfo->marker = &marker->pub;
  /* Initialize public method pointers */
  marker->pub.reset_marker_reader = reset_marker_reader;
@@ -1443,7 +1449,7 @@ jinit_marker_reader (j_decompress_ptr cinfo)

 GLOBAL(void)
 jpeg_save_markers (j_decompress_ptr cinfo, int marker_code,
-                   unsigned int length_limit)
+		   unsigned int length_limit)
 {
  my_marker_ptr marker = (my_marker_ptr) cinfo->marker;
  long maxlength;
@@ -1492,7 +1498,7 @@ jpeg_save_markers (j_decompress_ptr cinfo, int marker_code,

 GLOBAL(void)
 jpeg_set_marker_processor (j_decompress_ptr cinfo, int marker_code,
-                           jpeg_marker_parser_method routine)
+			   jpeg_marker_parser_method routine)
 {
  my_marker_ptr marker = (my_marker_ptr) cinfo->marker;

@@ -2,7 +2,7 @@
 * jdmaster.c
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
- * Modified 2002-2011 by Guido Vollbeding.
+ * Modified 2002-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -45,13 +45,26 @@ LOCAL(boolean)
 use_merged_upsample (j_decompress_ptr cinfo)
 {
 #ifdef UPSAMPLE_MERGING_SUPPORTED
-  /* Merging is the equivalent of plain box-filter upsampling */
-  if (cinfo->do_fancy_upsampling || cinfo->CCIR601_sampling)
+  /* Merging is the equivalent of plain box-filter upsampling. */
+  /* The following condition is only needed if fancy shall select
+   * a different upsampling method.  In our current implementation
+   * fancy only affects the DCT scaling, thus we can use fancy
+   * upsampling and merged upsample simultaneously, in particular
+   * with scaled DCT sizes larger than the default DCTSIZE.
+   */
+#if 0
+  if (cinfo->do_fancy_upsampling)
+    return FALSE;
+#endif
+  if (cinfo->CCIR601_sampling)
    return FALSE;
  /* jdmerge.c only supports YCC=>RGB color conversion */
-  if (cinfo->jpeg_color_space != JCS_YCbCr || cinfo->num_components != 3 ||
+  if ((cinfo->jpeg_color_space != JCS_YCbCr &&
+       cinfo->jpeg_color_space != JCS_BG_YCC) ||
+      cinfo->num_components != 3 ||
      cinfo->out_color_space != JCS_RGB ||
-      cinfo->out_color_components != RGB_PIXELSIZE)
+      cinfo->out_color_components != RGB_PIXELSIZE ||
+      cinfo->color_transform)
    return FALSE;
  /* and it only handles 2h1v or 2h2v sampling ratios */
  if (cinfo->comp_info[0].h_samp_factor != 2 ||
@@ -113,24 +126,24 @@ jpeg_calc_output_dimensions (j_decompress_ptr cinfo)
       ci++, compptr++) {
    int ssize = 1;
    while (cinfo->min_DCT_h_scaled_size * ssize <=
-           (cinfo->do_fancy_upsampling ? DCTSIZE : DCTSIZE / 2) &&
-           (cinfo->max_h_samp_factor % (compptr->h_samp_factor * ssize * 2)) == 0) {
+	   (cinfo->do_fancy_upsampling ? DCTSIZE : DCTSIZE / 2) &&
+	   (cinfo->max_h_samp_factor % (compptr->h_samp_factor * ssize * 2)) == 0) {
      ssize = ssize * 2;
    }
    compptr->DCT_h_scaled_size = cinfo->min_DCT_h_scaled_size * ssize;
    ssize = 1;
    while (cinfo->min_DCT_v_scaled_size * ssize <=
-           (cinfo->do_fancy_upsampling ? DCTSIZE : DCTSIZE / 2) &&
-           (cinfo->max_v_samp_factor % (compptr->v_samp_factor * ssize * 2)) == 0) {
+	   (cinfo->do_fancy_upsampling ? DCTSIZE : DCTSIZE / 2) &&
+	   (cinfo->max_v_samp_factor % (compptr->v_samp_factor * ssize * 2)) == 0) {
      ssize = ssize * 2;
    }
    compptr->DCT_v_scaled_size = cinfo->min_DCT_v_scaled_size * ssize;

    /* We don't support IDCT ratios larger than 2. */
    if (compptr->DCT_h_scaled_size > compptr->DCT_v_scaled_size * 2)
-        compptr->DCT_h_scaled_size = compptr->DCT_v_scaled_size * 2;
+	compptr->DCT_h_scaled_size = compptr->DCT_v_scaled_size * 2;
    else if (compptr->DCT_v_scaled_size > compptr->DCT_h_scaled_size * 2)
-        compptr->DCT_v_scaled_size = compptr->DCT_h_scaled_size * 2;
+	compptr->DCT_v_scaled_size = compptr->DCT_h_scaled_size * 2;
  }

  /* Recompute downsampled dimensions of components;
@@ -141,12 +154,12 @@ jpeg_calc_output_dimensions (j_decompress_ptr cinfo)
    /* Size in samples, after IDCT scaling */
    compptr->downsampled_width = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_width *
-                    (long) (compptr->h_samp_factor * compptr->DCT_h_scaled_size),
-                    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
+		    (long) (compptr->h_samp_factor * compptr->DCT_h_scaled_size),
+		    (long) (cinfo->max_h_samp_factor * cinfo->block_size));
    compptr->downsampled_height = (JDIMENSION)
      jdiv_round_up((long) cinfo->image_height *
-                    (long) (compptr->v_samp_factor * compptr->DCT_v_scaled_size),
-                    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
+		    (long) (compptr->v_samp_factor * compptr->DCT_v_scaled_size),
+		    (long) (cinfo->max_v_samp_factor * cinfo->block_size));
  }

 #endif /* IDCT_SCALING_SUPPORTED */
@@ -158,9 +171,11 @@ jpeg_calc_output_dimensions (j_decompress_ptr cinfo)
    cinfo->out_color_components = 1;
    break;
  case JCS_RGB:
+  case JCS_BG_RGB:
    cinfo->out_color_components = RGB_PIXELSIZE;
    break;
  case JCS_YCbCr:
+  case JCS_BG_YCC:
    cinfo->out_color_components = 3;
    break;
  case JCS_CMYK:
@@ -172,7 +187,7 @@ jpeg_calc_output_dimensions (j_decompress_ptr cinfo)
    break;
  }
  cinfo->output_components = (cinfo->quantize_colors ? 1 :
-                              cinfo->out_color_components);
+			      cinfo->out_color_components);

  /* See if upsampler will want to emit more than one row at a time */
  if (use_merged_upsample(cinfo))
@@ -196,30 +211,20 @@ jpeg_calc_output_dimensions (j_decompress_ptr cinfo)
 * These processes all use a common table prepared by the routine below.
 *
 * For most steps we can mathematically guarantee that the initial value
- * of x is within MAXJSAMPLE+1 of the legal range, so a table running from
- * -(MAXJSAMPLE+1) to 2*MAXJSAMPLE+1 is sufficient.  But for the initial
- * limiting step (just after the IDCT), a wildly out-of-range value is
- * possible if the input data is corrupt.  To avoid any chance of indexing
+ * of x is within 2*(MAXJSAMPLE+1) of the legal range, so a table running
+ * from -2*(MAXJSAMPLE+1) to 3*MAXJSAMPLE+2 is sufficient.  But for the
+ * initial limiting step (just after the IDCT), a wildly out-of-range value
+ * is possible if the input data is corrupt.  To avoid any chance of indexing
 * off the end of memory and getting a bad-pointer trap, we perform the
 * post-IDCT limiting thus:
- *		x = range_limit[x & MASK];
+ *		x = (sample_range_limit - SUBSET)[(x + CENTER) & MASK];
 * where MASK is 2 bits wider than legal sample data, ie 10 bits for 8-bit
 * samples.  Under normal circumstances this is more than enough range and
 * a correct output will be generated; with bogus input data the mask will
 * cause wraparound, and we will safely generate a bogus-but-in-range output.
 * For the post-IDCT step, we want to convert the data from signed to unsigned
 * representation by adding CENTERJSAMPLE at the same time that we limit it.
- * So the post-IDCT limiting table ends up looking like this:
- *   CENTERJSAMPLE,CENTERJSAMPLE+1,...,MAXJSAMPLE,
- *   MAXJSAMPLE (repeat 2*(MAXJSAMPLE+1)-CENTERJSAMPLE times),
- *   0          (repeat 2*(MAXJSAMPLE+1)-CENTERJSAMPLE times),
- *   0,1,...,CENTERJSAMPLE-1
- * Negative inputs select values from the upper half of the table after
- * masking.
- *
- * We can save some space by overlapping the start of the post-IDCT table
- * with the simpler range limiting table.  The post-IDCT table begins at
- * sample_range_limit + CENTERJSAMPLE.
+ * This is accomplished with SUBSET = CENTER - CENTERJSAMPLE.
 *
 * Note that the table is allocated in near data space on PCs; it's small
 * enough and used often enough to justify this.
@@ -234,23 +239,17 @@ prepare_range_limit_table (j_decompress_ptr cinfo)

  table = (JSAMPLE *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                (5 * (MAXJSAMPLE+1) + CENTERJSAMPLE) * SIZEOF(JSAMPLE));
-  table += (MAXJSAMPLE+1);	/* allow negative subscripts of simple table */
+				5 * (MAXJSAMPLE+1) * SIZEOF(JSAMPLE));
+  /* First segment of range limit table: limit[x] = 0 for x < 0 */
+  MEMZERO(table, 2 * (MAXJSAMPLE+1) * SIZEOF(JSAMPLE));
+  table += 2 * (MAXJSAMPLE+1);	/* allow negative subscripts of table */
  cinfo->sample_range_limit = table;
-  /* First segment of "simple" table: limit[x] = 0 for x < 0 */
-  MEMZERO(table - (MAXJSAMPLE+1), (MAXJSAMPLE+1) * SIZEOF(JSAMPLE));
-  /* Main part of "simple" table: limit[x] = x */
+  /* Main part of range limit table: limit[x] = x */
  for (i = 0; i <= MAXJSAMPLE; i++)
    table[i] = (JSAMPLE) i;
-  table += CENTERJSAMPLE;	/* Point to where post-IDCT table starts */
-  /* End of simple table, rest of first half of post-IDCT table */
-  for (i = CENTERJSAMPLE; i < 2*(MAXJSAMPLE+1); i++)
+  /* End of range limit table: limit[x] = MAXJSAMPLE for x > MAXJSAMPLE */
+  for (; i < 3 * (MAXJSAMPLE+1); i++)
    table[i] = MAXJSAMPLE;
-  /* Second half of post-IDCT table */
-  MEMZERO(table + (2 * (MAXJSAMPLE+1)),
-          (2 * (MAXJSAMPLE+1) - CENTERJSAMPLE) * SIZEOF(JSAMPLE));
-  MEMCOPY(table + (4 * (MAXJSAMPLE+1) - CENTERJSAMPLE),
-          cinfo->sample_range_limit, CENTERJSAMPLE * SIZEOF(JSAMPLE));
 }


@@ -273,10 +272,19 @@ master_selection (j_decompress_ptr cinfo)
  long samplesperrow;
  JDIMENSION jd_samplesperrow;

+  /* For now, precision must match compiled-in value... */
+  if (cinfo->data_precision != BITS_IN_JSAMPLE)
+    ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision);
+
  /* Initialize dimensions and other stuff */
  jpeg_calc_output_dimensions(cinfo);
  prepare_range_limit_table(cinfo);

+  /* Sanity check on image dimensions */
+  if (cinfo->output_height <= 0 || cinfo->output_width <= 0 ||
+      cinfo->out_color_components <= 0)
+    ERREXIT(cinfo, JERR_EMPTY_IMAGE);
+
  /* Width of an output scanline must be representable as JDIMENSION. */
  samplesperrow = (long) cinfo->output_width * (long) cinfo->out_color_components;
  jd_samplesperrow = (JDIMENSION) samplesperrow;
@@ -427,24 +435,24 @@ prepare_for_output_pass (j_decompress_ptr cinfo)
    if (cinfo->quantize_colors && cinfo->colormap == NULL) {
      /* Select new quantization method */
      if (cinfo->two_pass_quantize && cinfo->enable_2pass_quant) {
-        cinfo->cquantize = master->quantizer_2pass;
-        master->pub.is_dummy_pass = TRUE;
+	cinfo->cquantize = master->quantizer_2pass;
+	master->pub.is_dummy_pass = TRUE;
      } else if (cinfo->enable_1pass_quant) {
-        cinfo->cquantize = master->quantizer_1pass;
+	cinfo->cquantize = master->quantizer_1pass;
      } else {
-        ERREXIT(cinfo, JERR_MODE_CHANGE);
+	ERREXIT(cinfo, JERR_MODE_CHANGE);
      }
    }
    (*cinfo->idct->start_pass) (cinfo);
    (*cinfo->coef->start_output_pass) (cinfo);
    if (! cinfo->raw_data_out) {
      if (! master->using_merged_upsample)
-        (*cinfo->cconvert->start_pass) (cinfo);
+	(*cinfo->cconvert->start_pass) (cinfo);
      (*cinfo->upsample->start_pass) (cinfo);
      if (cinfo->quantize_colors)
-        (*cinfo->cquantize->start_pass) (cinfo, master->pub.is_dummy_pass);
+	(*cinfo->cquantize->start_pass) (cinfo, master->pub.is_dummy_pass);
      (*cinfo->post->start_pass) (cinfo,
-            (master->pub.is_dummy_pass ? JBUF_SAVE_AND_PASS : JBUF_PASS_THRU));
+	    (master->pub.is_dummy_pass ? JBUF_SAVE_AND_PASS : JBUF_PASS_THRU));
      (*cinfo->main->start_pass) (cinfo, JBUF_PASS_THRU);
    }
  }
@@ -453,7 +461,7 @@ prepare_for_output_pass (j_decompress_ptr cinfo)
  if (cinfo->progress != NULL) {
    cinfo->progress->completed_passes = master->pass_number;
    cinfo->progress->total_passes = master->pass_number +
-                                    (master->pub.is_dummy_pass ? 2 : 1);
+				    (master->pub.is_dummy_pass ? 2 : 1);
    /* In buffered-image mode, we assume one more output pass if EOI not
     * yet reached, but no more passes if EOI has been reached.
     */
@@ -520,8 +528,8 @@ jinit_master_decompress (j_decompress_ptr cinfo)

  master = (my_master_ptr)
      (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                  SIZEOF(my_decomp_master));
-  cinfo->master = (struct jpeg_decomp_master *) master;
+				  SIZEOF(my_decomp_master));
+  cinfo->master = &master->pub;
  master->pub.prepare_for_output_pass = prepare_for_output_pass;
  master->pub.finish_output_pass = finish_output_pass;

@@ -2,6 +2,7 @@
 * jdmerge.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
+ * Modified 2013-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -23,7 +24,7 @@
 * multiplications needed for color conversion.
 *
 * This file currently provides implementations for the following cases:
- *	YCbCr => RGB color conversion only.
+ *	YCC => RGB color conversion only (YCbCr or BG_YCC).
 *	Sampling ratios of 2h1v or 2h2v.
 *	No scaling needed at upsample time.
 *	Corner-aligned (non-CCIR601) sampling alignment.
@@ -46,8 +47,8 @@ typedef struct {

  /* Pointer to routine to do actual upsampling/conversion of one row group */
  JMETHOD(void, upmethod, (j_decompress_ptr cinfo,
-                           JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr,
-                           JSAMPARRAY output_buf));
+			   JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr,
+			   JSAMPARRAY output_buf));

  /* Private state for YCC->RGB conversion */
  int * Cr_r_tab;		/* => table for Cr to R conversion */
@@ -75,12 +76,13 @@ typedef my_upsampler * my_upsample_ptr;


 /*
- * Initialize tables for YCC->RGB colorspace conversion.
+ * Initialize tables for YCbCr->RGB and BG_YCC->RGB colorspace conversion.
 * This is taken directly from jdcolor.c; see that file for more info.
 */

 LOCAL(void)
 build_ycc_rgb_table (j_decompress_ptr cinfo)
+/* Normal case, sYCC */
 {
  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;
  int i;
@@ -89,31 +91,71 @@ build_ycc_rgb_table (j_decompress_ptr cinfo)

  upsample->Cr_r_tab = (int *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(int));
+				(MAXJSAMPLE+1) * SIZEOF(int));
  upsample->Cb_b_tab = (int *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(int));
+				(MAXJSAMPLE+1) * SIZEOF(int));
  upsample->Cr_g_tab = (INT32 *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(INT32));
+				(MAXJSAMPLE+1) * SIZEOF(INT32));
  upsample->Cb_g_tab = (INT32 *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                (MAXJSAMPLE+1) * SIZEOF(INT32));
+				(MAXJSAMPLE+1) * SIZEOF(INT32));

  for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) {
    /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */
    /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */
-    /* Cr=>R value is nearest int to 1.40200 * x */
+    /* Cr=>R value is nearest int to 1.402 * x */
    upsample->Cr_r_tab[i] = (int)
-                    RIGHT_SHIFT(FIX(1.40200) * x + ONE_HALF, SCALEBITS);
-    /* Cb=>B value is nearest int to 1.77200 * x */
+		    RIGHT_SHIFT(FIX(1.402) * x + ONE_HALF, SCALEBITS);
+    /* Cb=>B value is nearest int to 1.772 * x */
    upsample->Cb_b_tab[i] = (int)
-                    RIGHT_SHIFT(FIX(1.77200) * x + ONE_HALF, SCALEBITS);
-    /* Cr=>G value is scaled-up -0.71414 * x */
-    upsample->Cr_g_tab[i] = (- FIX(0.71414)) * x;
-    /* Cb=>G value is scaled-up -0.34414 * x */
+		    RIGHT_SHIFT(FIX(1.772) * x + ONE_HALF, SCALEBITS);
+    /* Cr=>G value is scaled-up -0.714136286 * x */
+    upsample->Cr_g_tab[i] = (- FIX(0.714136286)) * x;
+    /* Cb=>G value is scaled-up -0.344136286 * x */
    /* We also add in ONE_HALF so that need not do it in inner loop */
-    upsample->Cb_g_tab[i] = (- FIX(0.34414)) * x + ONE_HALF;
+    upsample->Cb_g_tab[i] = (- FIX(0.344136286)) * x + ONE_HALF;
+  }
+}
+
+
+LOCAL(void)
+build_bg_ycc_rgb_table (j_decompress_ptr cinfo)
+/* Wide gamut case, bg-sYCC */
+{
+  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;
+  int i;
+  INT32 x;
+  SHIFT_TEMPS
+
+  upsample->Cr_r_tab = (int *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(int));
+  upsample->Cb_b_tab = (int *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(int));
+  upsample->Cr_g_tab = (INT32 *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(INT32));
+  upsample->Cb_g_tab = (INT32 *)
+    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
+				(MAXJSAMPLE+1) * SIZEOF(INT32));
+
+  for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) {
+    /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */
+    /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */
+    /* Cr=>R value is nearest int to 2.804 * x */
+    upsample->Cr_r_tab[i] = (int)
+		    RIGHT_SHIFT(FIX(2.804) * x + ONE_HALF, SCALEBITS);
+    /* Cb=>B value is nearest int to 3.544 * x */
+    upsample->Cb_b_tab[i] = (int)
+		    RIGHT_SHIFT(FIX(3.544) * x + ONE_HALF, SCALEBITS);
+    /* Cr=>G value is scaled-up -1.428272572 * x */
+    upsample->Cr_g_tab[i] = (- FIX(1.428272572)) * x;
+    /* Cb=>G value is scaled-up -0.688272572 * x */
+    /* We also add in ONE_HALF so that need not do it in inner loop */
+    upsample->Cb_g_tab[i] = (- FIX(0.688272572)) * x + ONE_HALF;
  }
 }

@@ -142,10 +184,10 @@ start_pass_merged_upsample (j_decompress_ptr cinfo)

 METHODDEF(void)
 merged_2v_upsample (j_decompress_ptr cinfo,
-                    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-                    JDIMENSION in_row_groups_avail,
-                    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                    JDIMENSION out_rows_avail)
+		    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+		    JDIMENSION in_row_groups_avail,
+		    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+		    JDIMENSION out_rows_avail)
 /* 2:1 vertical sampling case: may need a spare row. */
 {
  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;
@@ -155,7 +197,7 @@ merged_2v_upsample (j_decompress_ptr cinfo,
  if (upsample->spare_full) {
    /* If we have a spare row saved from a previous cycle, just return it. */
    jcopy_sample_rows(& upsample->spare_row, 0, output_buf + *out_row_ctr, 0,
-                      1, upsample->out_row_width);
+		      1, upsample->out_row_width);
    num_rows = 1;
    upsample->spare_full = FALSE;
  } else {
@@ -191,17 +233,17 @@ merged_2v_upsample (j_decompress_ptr cinfo,

 METHODDEF(void)
 merged_1v_upsample (j_decompress_ptr cinfo,
-                    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-                    JDIMENSION in_row_groups_avail,
-                    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                    JDIMENSION out_rows_avail)
+		    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+		    JDIMENSION in_row_groups_avail,
+		    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+		    JDIMENSION out_rows_avail)
 /* 1:1 vertical sampling case: much easier, never need a spare row. */
 {
  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;

  /* Just do the upsampling. */
  (*upsample->upmethod) (cinfo, input_buf, *in_row_group_ctr,
-                         output_buf + *out_row_ctr);
+			 output_buf + *out_row_ctr);
  /* Adjust counts */
  (*out_row_ctr)++;
  (*in_row_group_ctr)++;
@@ -224,8 +266,8 @@ merged_1v_upsample (j_decompress_ptr cinfo,

 METHODDEF(void)
 h2v1_merged_upsample (j_decompress_ptr cinfo,
-                      JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr,
-                      JSAMPARRAY output_buf)
+		      JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr,
+		      JSAMPARRAY output_buf)
 {
  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;
  register int y, cred, cgreen, cblue;
@@ -250,32 +292,32 @@ h2v1_merged_upsample (j_decompress_ptr cinfo,
    /* Do the chroma part of the calculation */
    cb = GETJSAMPLE(*inptr1++);
    cr = GETJSAMPLE(*inptr2++);
-    cred = Crrtab[cr];
+    cred   = Crrtab[cr];
    cgreen = (int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS);
-    cblue = Cbbtab[cb];
+    cblue  = Cbbtab[cb];
    /* Fetch 2 Y values and emit 2 pixels */
    y  = GETJSAMPLE(*inptr0++);
-    outptr[RGB_RED] =   range_limit[y + cred];
+    outptr[RGB_RED]   = range_limit[y + cred];
    outptr[RGB_GREEN] = range_limit[y + cgreen];
-    outptr[RGB_BLUE] =  range_limit[y + cblue];
+    outptr[RGB_BLUE]  = range_limit[y + cblue];
    outptr += RGB_PIXELSIZE;
    y  = GETJSAMPLE(*inptr0++);
-    outptr[RGB_RED] =   range_limit[y + cred];
+    outptr[RGB_RED]   = range_limit[y + cred];
    outptr[RGB_GREEN] = range_limit[y + cgreen];
-    outptr[RGB_BLUE] =  range_limit[y + cblue];
+    outptr[RGB_BLUE]  = range_limit[y + cblue];
    outptr += RGB_PIXELSIZE;
  }
  /* If image width is odd, do the last output column separately */
  if (cinfo->output_width & 1) {
    cb = GETJSAMPLE(*inptr1);
    cr = GETJSAMPLE(*inptr2);
-    cred = Crrtab[cr];
+    cred   = Crrtab[cr];
    cgreen = (int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS);
-    cblue = Cbbtab[cb];
+    cblue  = Cbbtab[cb];
    y  = GETJSAMPLE(*inptr0);
-    outptr[RGB_RED] =   range_limit[y + cred];
+    outptr[RGB_RED]   = range_limit[y + cred];
    outptr[RGB_GREEN] = range_limit[y + cgreen];
-    outptr[RGB_BLUE] =  range_limit[y + cblue];
+    outptr[RGB_BLUE]  = range_limit[y + cblue];
  }
 }

@@ -286,8 +328,8 @@ h2v1_merged_upsample (j_decompress_ptr cinfo,

 METHODDEF(void)
 h2v2_merged_upsample (j_decompress_ptr cinfo,
-                      JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr,
-                      JSAMPARRAY output_buf)
+		      JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr,
+		      JSAMPARRAY output_buf)
 {
  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;
  register int y, cred, cgreen, cblue;
@@ -314,46 +356,46 @@ h2v2_merged_upsample (j_decompress_ptr cinfo,
    /* Do the chroma part of the calculation */
    cb = GETJSAMPLE(*inptr1++);
    cr = GETJSAMPLE(*inptr2++);
-    cred = Crrtab[cr];
+    cred   = Crrtab[cr];
    cgreen = (int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS);
-    cblue = Cbbtab[cb];
+    cblue  = Cbbtab[cb];
    /* Fetch 4 Y values and emit 4 pixels */
    y  = GETJSAMPLE(*inptr00++);
-    outptr0[RGB_RED] =   range_limit[y + cred];
+    outptr0[RGB_RED]   = range_limit[y + cred];
    outptr0[RGB_GREEN] = range_limit[y + cgreen];
-    outptr0[RGB_BLUE] =  range_limit[y + cblue];
+    outptr0[RGB_BLUE]  = range_limit[y + cblue];
    outptr0 += RGB_PIXELSIZE;
    y  = GETJSAMPLE(*inptr00++);
-    outptr0[RGB_RED] =   range_limit[y + cred];
+    outptr0[RGB_RED]   = range_limit[y + cred];
    outptr0[RGB_GREEN] = range_limit[y + cgreen];
-    outptr0[RGB_BLUE] =  range_limit[y + cblue];
+    outptr0[RGB_BLUE]  = range_limit[y + cblue];
    outptr0 += RGB_PIXELSIZE;
    y  = GETJSAMPLE(*inptr01++);
-    outptr1[RGB_RED] =   range_limit[y + cred];
+    outptr1[RGB_RED]   = range_limit[y + cred];
    outptr1[RGB_GREEN] = range_limit[y + cgreen];
-    outptr1[RGB_BLUE] =  range_limit[y + cblue];
+    outptr1[RGB_BLUE]  = range_limit[y + cblue];
    outptr1 += RGB_PIXELSIZE;
    y  = GETJSAMPLE(*inptr01++);
-    outptr1[RGB_RED] =   range_limit[y + cred];
+    outptr1[RGB_RED]   = range_limit[y + cred];
    outptr1[RGB_GREEN] = range_limit[y + cgreen];
-    outptr1[RGB_BLUE] =  range_limit[y + cblue];
+    outptr1[RGB_BLUE]  = range_limit[y + cblue];
    outptr1 += RGB_PIXELSIZE;
  }
  /* If image width is odd, do the last output column separately */
  if (cinfo->output_width & 1) {
    cb = GETJSAMPLE(*inptr1);
    cr = GETJSAMPLE(*inptr2);
-    cred = Crrtab[cr];
+    cred   = Crrtab[cr];
    cgreen = (int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS);
-    cblue = Cbbtab[cb];
+    cblue  = Cbbtab[cb];
    y  = GETJSAMPLE(*inptr00);
-    outptr0[RGB_RED] =   range_limit[y + cred];
+    outptr0[RGB_RED]   = range_limit[y + cred];
    outptr0[RGB_GREEN] = range_limit[y + cgreen];
-    outptr0[RGB_BLUE] =  range_limit[y + cblue];
+    outptr0[RGB_BLUE]  = range_limit[y + cblue];
    y  = GETJSAMPLE(*inptr01);
-    outptr1[RGB_RED] =   range_limit[y + cred];
+    outptr1[RGB_RED]   = range_limit[y + cred];
    outptr1[RGB_GREEN] = range_limit[y + cgreen];
-    outptr1[RGB_BLUE] =  range_limit[y + cblue];
+    outptr1[RGB_BLUE]  = range_limit[y + cblue];
  }
 }

@@ -373,8 +415,8 @@ jinit_merged_upsampler (j_decompress_ptr cinfo)

  upsample = (my_upsample_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_upsampler));
-  cinfo->upsample = (struct jpeg_upsampler *) upsample;
+				SIZEOF(my_upsampler));
+  cinfo->upsample = &upsample->pub;
  upsample->pub.start_pass = start_pass_merged_upsample;
  upsample->pub.need_context_rows = FALSE;

@@ -386,7 +428,7 @@ jinit_merged_upsampler (j_decompress_ptr cinfo)
    /* Allocate a spare row buffer */
    upsample->spare_row = (JSAMPROW)
      (*cinfo->mem->alloc_large) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                (size_t) (upsample->out_row_width * SIZEOF(JSAMPLE)));
+		(size_t) (upsample->out_row_width * SIZEOF(JSAMPLE)));
  } else {
    upsample->pub.upsample = merged_1v_upsample;
    upsample->upmethod = h2v1_merged_upsample;
@@ -394,7 +436,10 @@ jinit_merged_upsampler (j_decompress_ptr cinfo)
    upsample->spare_row = NULL;
  }

-  build_ycc_rgb_table(cinfo);
+  if (cinfo->jpeg_color_space == JCS_BG_YCC)
+    build_bg_ycc_rgb_table(cinfo);
+  else
+    build_ycc_rgb_table(cinfo);
 }

 #endif /* UPSAMPLE_MERGING_SUPPORTED */
@@ -44,24 +44,24 @@ typedef my_post_controller * my_post_ptr;

 /* Forward declarations */
 METHODDEF(void) post_process_1pass
-        JPP((j_decompress_ptr cinfo,
-             JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-             JDIMENSION in_row_groups_avail,
-             JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-             JDIMENSION out_rows_avail));
+	JPP((j_decompress_ptr cinfo,
+	     JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+	     JDIMENSION in_row_groups_avail,
+	     JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+	     JDIMENSION out_rows_avail));
 #ifdef QUANT_2PASS_SUPPORTED
 METHODDEF(void) post_process_prepass
-        JPP((j_decompress_ptr cinfo,
-             JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-             JDIMENSION in_row_groups_avail,
-             JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-             JDIMENSION out_rows_avail));
+	JPP((j_decompress_ptr cinfo,
+	     JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+	     JDIMENSION in_row_groups_avail,
+	     JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+	     JDIMENSION out_rows_avail));
 METHODDEF(void) post_process_2pass
-        JPP((j_decompress_ptr cinfo,
-             JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-             JDIMENSION in_row_groups_avail,
-             JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-             JDIMENSION out_rows_avail));
+	JPP((j_decompress_ptr cinfo,
+	     JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+	     JDIMENSION in_row_groups_avail,
+	     JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+	     JDIMENSION out_rows_avail));
 #endif


@@ -84,9 +84,9 @@ start_pass_dpost (j_decompress_ptr cinfo, J_BUF_MODE pass_mode)
       * allocate a strip buffer.  Use the virtual-array buffer as workspace.
       */
      if (post->buffer == NULL) {
-        post->buffer = (*cinfo->mem->access_virt_sarray)
-          ((j_common_ptr) cinfo, post->whole_image,
-           (JDIMENSION) 0, post->strip_height, TRUE);
+	post->buffer = (*cinfo->mem->access_virt_sarray)
+	  ((j_common_ptr) cinfo, post->whole_image,
+	   (JDIMENSION) 0, post->strip_height, TRUE);
      }
    } else {
      /* For single-pass processing without color quantization,
@@ -124,10 +124,10 @@ start_pass_dpost (j_decompress_ptr cinfo, J_BUF_MODE pass_mode)

 METHODDEF(void)
 post_process_1pass (j_decompress_ptr cinfo,
-                    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-                    JDIMENSION in_row_groups_avail,
-                    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                    JDIMENSION out_rows_avail)
+		    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+		    JDIMENSION in_row_groups_avail,
+		    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+		    JDIMENSION out_rows_avail)
 {
  my_post_ptr post = (my_post_ptr) cinfo->post;
  JDIMENSION num_rows, max_rows;
@@ -139,11 +139,11 @@ post_process_1pass (j_decompress_ptr cinfo,
    max_rows = post->strip_height;
  num_rows = 0;
  (*cinfo->upsample->upsample) (cinfo,
-                input_buf, in_row_group_ctr, in_row_groups_avail,
-                post->buffer, &num_rows, max_rows);
+		input_buf, in_row_group_ctr, in_row_groups_avail,
+		post->buffer, &num_rows, max_rows);
  /* Quantize and emit data. */
  (*cinfo->cquantize->color_quantize) (cinfo,
-                post->buffer, output_buf + *out_row_ctr, (int) num_rows);
+		post->buffer, output_buf + *out_row_ctr, (int) num_rows);
  *out_row_ctr += num_rows;
 }

@@ -156,10 +156,10 @@ post_process_1pass (j_decompress_ptr cinfo,

 METHODDEF(void)
 post_process_prepass (j_decompress_ptr cinfo,
-                      JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-                      JDIMENSION in_row_groups_avail,
-                      JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                      JDIMENSION out_rows_avail)
+		      JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+		      JDIMENSION in_row_groups_avail,
+		      JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+		      JDIMENSION out_rows_avail)
 {
  my_post_ptr post = (my_post_ptr) cinfo->post;
  JDIMENSION old_next_row, num_rows;
@@ -167,22 +167,22 @@ post_process_prepass (j_decompress_ptr cinfo,
  /* Reposition virtual buffer if at start of strip. */
  if (post->next_row == 0) {
    post->buffer = (*cinfo->mem->access_virt_sarray)
-        ((j_common_ptr) cinfo, post->whole_image,
-         post->starting_row, post->strip_height, TRUE);
+	((j_common_ptr) cinfo, post->whole_image,
+	 post->starting_row, post->strip_height, TRUE);
  }

  /* Upsample some data (up to a strip height's worth). */
  old_next_row = post->next_row;
  (*cinfo->upsample->upsample) (cinfo,
-                input_buf, in_row_group_ctr, in_row_groups_avail,
-                post->buffer, &post->next_row, post->strip_height);
+		input_buf, in_row_group_ctr, in_row_groups_avail,
+		post->buffer, &post->next_row, post->strip_height);

  /* Allow quantizer to scan new data.  No data is emitted, */
  /* but we advance out_row_ctr so outer loop can tell when we're done. */
  if (post->next_row > old_next_row) {
    num_rows = post->next_row - old_next_row;
    (*cinfo->cquantize->color_quantize) (cinfo, post->buffer + old_next_row,
-                                         (JSAMPARRAY) NULL, (int) num_rows);
+					 (JSAMPARRAY) NULL, (int) num_rows);
    *out_row_ctr += num_rows;
  }

@@ -200,10 +200,10 @@ post_process_prepass (j_decompress_ptr cinfo,

 METHODDEF(void)
 post_process_2pass (j_decompress_ptr cinfo,
-                    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-                    JDIMENSION in_row_groups_avail,
-                    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                    JDIMENSION out_rows_avail)
+		    JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+		    JDIMENSION in_row_groups_avail,
+		    JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+		    JDIMENSION out_rows_avail)
 {
  my_post_ptr post = (my_post_ptr) cinfo->post;
  JDIMENSION num_rows, max_rows;
@@ -211,8 +211,8 @@ post_process_2pass (j_decompress_ptr cinfo,
  /* Reposition virtual buffer if at start of strip. */
  if (post->next_row == 0) {
    post->buffer = (*cinfo->mem->access_virt_sarray)
-        ((j_common_ptr) cinfo, post->whole_image,
-         post->starting_row, post->strip_height, FALSE);
+	((j_common_ptr) cinfo, post->whole_image,
+	 post->starting_row, post->strip_height, FALSE);
  }

  /* Determine number of rows to emit. */
@@ -227,8 +227,8 @@ post_process_2pass (j_decompress_ptr cinfo,

  /* Quantize and emit data. */
  (*cinfo->cquantize->color_quantize) (cinfo,
-                post->buffer + post->next_row, output_buf + *out_row_ctr,
-                (int) num_rows);
+		post->buffer + post->next_row, output_buf + *out_row_ctr,
+		(int) num_rows);
  *out_row_ctr += num_rows;

  /* Advance if we filled the strip. */
@@ -253,7 +253,7 @@ jinit_d_post_controller (j_decompress_ptr cinfo, boolean need_full_buffer)

  post = (my_post_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_post_controller));
+				SIZEOF(my_post_controller));
  cinfo->post = (struct jpeg_d_post_controller *) post;
  post->pub.start_pass = start_pass_dpost;
  post->whole_image = NULL;	/* flag for no virtual arrays */
@@ -271,20 +271,20 @@ jinit_d_post_controller (j_decompress_ptr cinfo, boolean need_full_buffer)
      /* We round up the number of rows to a multiple of the strip height. */
 #ifdef QUANT_2PASS_SUPPORTED
      post->whole_image = (*cinfo->mem->request_virt_sarray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE,
-         cinfo->output_width * cinfo->out_color_components,
-         (JDIMENSION) jround_up((long) cinfo->output_height,
-                                (long) post->strip_height),
-         post->strip_height);
+	((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE,
+	 cinfo->output_width * cinfo->out_color_components,
+	 (JDIMENSION) jround_up((long) cinfo->output_height,
+				(long) post->strip_height),
+	 post->strip_height);
 #else
      ERREXIT(cinfo, JERR_BAD_BUFFER_MODE);
 #endif /* QUANT_2PASS_SUPPORTED */
    } else {
      /* One-pass color quantization: just make a strip buffer. */
      post->buffer = (*cinfo->mem->alloc_sarray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE,
-         cinfo->output_width * cinfo->out_color_components,
-         post->strip_height);
+	((j_common_ptr) cinfo, JPOOL_IMAGE,
+	 cinfo->output_width * cinfo->out_color_components,
+	 post->strip_height);
    }
  }
 }
@@ -2,7 +2,7 @@
 * jdsample.c
 *
 * Copyright (C) 1991-1996, Thomas G. Lane.
- * Modified 2002-2008 by Guido Vollbeding.
+ * Modified 2002-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -26,8 +26,8 @@

 /* Pointer to routine to upsample a single component */
 typedef JMETHOD(void, upsample1_ptr,
-                (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-                 JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr));
+		(j_decompress_ptr cinfo, jpeg_component_info * compptr,
+		 JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr));

 /* Private subobject */

@@ -88,10 +88,10 @@ start_pass_upsample (j_decompress_ptr cinfo)

 METHODDEF(void)
 sep_upsample (j_decompress_ptr cinfo,
-              JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
-              JDIMENSION in_row_groups_avail,
-              JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-              JDIMENSION out_rows_avail)
+	      JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr,
+	      JDIMENSION in_row_groups_avail,
+	      JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+	      JDIMENSION out_rows_avail)
 {
  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;
  int ci;
@@ -101,13 +101,13 @@ sep_upsample (j_decompress_ptr cinfo,
  /* Fill the conversion buffer, if it's empty */
  if (upsample->next_row_out >= cinfo->max_v_samp_factor) {
    for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
-         ci++, compptr++) {
+	 ci++, compptr++) {
      /* Invoke per-component upsample method.  Notice we pass a POINTER
       * to color_buf[ci], so that fullsize_upsample can change it.
       */
      (*upsample->methods[ci]) (cinfo, compptr,
-        input_buf[ci] + (*in_row_group_ctr * upsample->rowgroup_height[ci]),
-        upsample->color_buf + ci);
+	input_buf[ci] + (*in_row_group_ctr * upsample->rowgroup_height[ci]),
+	upsample->color_buf + ci);
    }
    upsample->next_row_out = 0;
  }
@@ -119,7 +119,7 @@ sep_upsample (j_decompress_ptr cinfo,
  /* Not more than the distance to the end of the image.  Need this test
   * in case the image height is not a multiple of max_v_samp_factor:
   */
-  if (num_rows > upsample->rows_to_go)
+  if (num_rows > upsample->rows_to_go) 
    num_rows = upsample->rows_to_go;
  /* And not more than what the client can accept: */
  out_rows_avail -= *out_row_ctr;
@@ -127,9 +127,9 @@ sep_upsample (j_decompress_ptr cinfo,
    num_rows = out_rows_avail;

  (*cinfo->cconvert->color_convert) (cinfo, upsample->color_buf,
-                                     (JDIMENSION) upsample->next_row_out,
-                                     output_buf + *out_row_ctr,
-                                     (int) num_rows);
+				     (JDIMENSION) upsample->next_row_out,
+				     output_buf + *out_row_ctr,
+				     (int) num_rows);

  /* Adjust counts */
  *out_row_ctr += num_rows;
@@ -156,7 +156,7 @@ sep_upsample (j_decompress_ptr cinfo,

 METHODDEF(void)
 fullsize_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-                   JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
+		   JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
 {
  *output_data_ptr = input_data;
 }
@@ -169,7 +169,7 @@ fullsize_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 noop_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-               JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
+	       JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
 {
  *output_data_ptr = NULL;	/* safety check */
 }
@@ -188,7 +188,7 @@ noop_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 int_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-              JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
+	      JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
 {
  my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample;
  JSAMPARRAY output_data = *output_data_ptr;
@@ -211,13 +211,13 @@ int_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
    while (outptr < outend) {
      invalue = *inptr++;	/* don't need GETJSAMPLE() here */
      for (h = h_expand; h > 0; h--) {
-        *outptr++ = invalue;
+	*outptr++ = invalue;
      }
    }
    /* Generate any additional output rows by duplicating the first one */
    if (v_expand > 1) {
      jcopy_sample_rows(output_data, outrow, output_data, outrow+1,
-                        v_expand-1, cinfo->output_width);
+			v_expand-1, cinfo->output_width);
    }
    inrow++;
    outrow += v_expand;
@@ -232,7 +232,7 @@ int_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 h2v1_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-               JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
+	       JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
 {
  JSAMPARRAY output_data = *output_data_ptr;
  register JSAMPROW inptr, outptr;
@@ -260,7 +260,7 @@ h2v1_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,

 METHODDEF(void)
 h2v2_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-               JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
+	       JSAMPARRAY input_data, JSAMPARRAY * output_data_ptr)
 {
  JSAMPARRAY output_data = *output_data_ptr;
  register JSAMPROW inptr, outptr;
@@ -279,7 +279,7 @@ h2v2_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr,
      *outptr++ = invalue;
    }
    jcopy_sample_rows(output_data, outrow, output_data, outrow+1,
-                      1, cinfo->output_width);
+		      1, cinfo->output_width);
    inrow++;
    outrow += 2;
  }
@@ -296,13 +296,12 @@ jinit_upsampler (j_decompress_ptr cinfo)
  my_upsample_ptr upsample;
  int ci;
  jpeg_component_info * compptr;
-  boolean need_buffer;
  int h_in_group, v_in_group, h_out_group, v_out_group;

  upsample = (my_upsample_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_upsampler));
-  cinfo->upsample = (struct jpeg_upsampler *) upsample;
+				SIZEOF(my_upsampler));
+  cinfo->upsample = &upsample->pub;
  upsample->pub.start_pass = start_pass_upsample;
  upsample->pub.upsample = sep_upsample;
  upsample->pub.need_context_rows = FALSE; /* until we find out differently */
@@ -319,43 +318,41 @@ jinit_upsampler (j_decompress_ptr cinfo)
     * are to be converted to max_h_samp_factor * max_v_samp_factor pixels.
     */
    h_in_group = (compptr->h_samp_factor * compptr->DCT_h_scaled_size) /
-                 cinfo->min_DCT_h_scaled_size;
+		 cinfo->min_DCT_h_scaled_size;
    v_in_group = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) /
-                 cinfo->min_DCT_v_scaled_size;
+		 cinfo->min_DCT_v_scaled_size;
    h_out_group = cinfo->max_h_samp_factor;
    v_out_group = cinfo->max_v_samp_factor;
    upsample->rowgroup_height[ci] = v_in_group; /* save for use later */
-    need_buffer = TRUE;
    if (! compptr->component_needed) {
      /* Don't bother to upsample an uninteresting component. */
      upsample->methods[ci] = noop_upsample;
-      need_buffer = FALSE;
-    } else if (h_in_group == h_out_group && v_in_group == v_out_group) {
+      continue;		/* don't need to allocate buffer */
+    }
+    if (h_in_group == h_out_group && v_in_group == v_out_group) {
      /* Fullsize components can be processed without any work. */
      upsample->methods[ci] = fullsize_upsample;
-      need_buffer = FALSE;
-    } else if (h_in_group * 2 == h_out_group &&
-               v_in_group == v_out_group) {
+      continue;		/* don't need to allocate buffer */
+    }
+    if (h_in_group * 2 == h_out_group && v_in_group == v_out_group) {
      /* Special case for 2h1v upsampling */
      upsample->methods[ci] = h2v1_upsample;
    } else if (h_in_group * 2 == h_out_group &&
-               v_in_group * 2 == v_out_group) {
+	       v_in_group * 2 == v_out_group) {
      /* Special case for 2h2v upsampling */
      upsample->methods[ci] = h2v2_upsample;
    } else if ((h_out_group % h_in_group) == 0 &&
-               (v_out_group % v_in_group) == 0) {
+	       (v_out_group % v_in_group) == 0) {
      /* Generic integral-factors upsampling method */
      upsample->methods[ci] = int_upsample;
      upsample->h_expand[ci] = (UINT8) (h_out_group / h_in_group);
      upsample->v_expand[ci] = (UINT8) (v_out_group / v_in_group);
    } else
      ERREXIT(cinfo, JERR_FRACT_SAMPLE_NOTIMPL);
-    if (need_buffer) {
-      upsample->color_buf[ci] = (*cinfo->mem->alloc_sarray)
-        ((j_common_ptr) cinfo, JPOOL_IMAGE,
-         (JDIMENSION) jround_up((long) cinfo->output_width,
-                                (long) cinfo->max_h_samp_factor),
-         (JDIMENSION) cinfo->max_v_samp_factor);
-    }
+    upsample->color_buf[ci] = (*cinfo->mem->alloc_sarray)
+      ((j_common_ptr) cinfo, JPOOL_IMAGE,
+       (JDIMENSION) jround_up((long) cinfo->output_width,
+			      (long) cinfo->max_h_samp_factor),
+       (JDIMENSION) cinfo->max_v_samp_factor);
  }
 }
@@ -56,20 +56,20 @@ jpeg_read_coefficients (j_decompress_ptr cinfo)
      int retcode;
      /* Call progress monitor hook if present */
      if (cinfo->progress != NULL)
-        (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
+	(*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo);
      /* Absorb some more input */
      retcode = (*cinfo->inputctl->consume_input) (cinfo);
      if (retcode == JPEG_SUSPENDED)
-        return NULL;
+	return NULL;
      if (retcode == JPEG_REACHED_EOI)
-        break;
+	break;
      /* Advance progress counter if appropriate */
      if (cinfo->progress != NULL &&
-          (retcode == JPEG_ROW_COMPLETED || retcode == JPEG_REACHED_SOS)) {
-        if (++cinfo->progress->pass_counter >= cinfo->progress->pass_limit) {
-          /* startup underestimated number of scans; ratchet up one scan */
-          cinfo->progress->pass_limit += (long) cinfo->total_iMCU_rows;
-        }
+	  (retcode == JPEG_ROW_COMPLETED || retcode == JPEG_REACHED_SOS)) {
+	if (++cinfo->progress->pass_counter >= cinfo->progress->pass_limit) {
+	  /* startup underestimated number of scans; ratchet up one scan */
+	  cinfo->progress->pass_limit += (long) cinfo->total_iMCU_rows;
+	}
      }
    }
    /* Set state so that jpeg_finish_decompress does the right thing */
@@ -2,7 +2,7 @@
 * jerror.c
 *
 * Copyright (C) 1991-1998, Thomas G. Lane.
- * Modified 2012 by Guido Vollbeding.
+ * Modified 2012-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -19,16 +19,16 @@
 * These routines are used by both the compression and decompression code.
 */

+#ifdef USE_WINDOWS_MESSAGEBOX
+#include <windows.h>
+#endif
+
 /* this is not a core library module, so it doesn't define JPEG_INTERNALS */
 #include "jinclude.h"
 #include "jpeglib.h"
 #include "jversion.h"
 #include "jerror.h"

-#ifdef USE_WINDOWS_MESSAGEBOX
-#include <windows.h>
-#endif
-
 #ifndef EXIT_FAILURE		/* define exit() codes if not provided */
 #define EXIT_FAILURE  1
 #endif
@@ -106,7 +106,7 @@ output_message (j_common_ptr cinfo)
 #ifdef USE_WINDOWS_MESSAGEBOX
  /* Display it in a message dialog box */
  MessageBox(GetActiveWindow(), buffer, "JPEG Library Error",
-             MB_OK | MB_ICONERROR);
+	     MB_OK | MB_ICONERROR);
 #else
  /* Send it to stderr, adding a newline */
  fprintf(stderr, "%s\n", buffer);
@@ -168,8 +168,8 @@ format_message (j_common_ptr cinfo, char * buffer)
  if (msg_code > 0 && msg_code <= err->last_jpeg_message) {
    msgtext = err->jpeg_message_table[msg_code];
  } else if (err->addon_message_table != NULL &&
-             msg_code >= err->first_addon_message &&
-             msg_code <= err->last_addon_message) {
+	     msg_code >= err->first_addon_message &&
+	     msg_code <= err->last_addon_message) {
    msgtext = err->addon_message_table[msg_code - err->first_addon_message];
  }

@@ -194,10 +194,10 @@ format_message (j_common_ptr cinfo, char * buffer)
    sprintf(buffer, msgtext, err->msg_parm.s);
  else
    sprintf(buffer, msgtext,
-            err->msg_parm.i[0], err->msg_parm.i[1],
-            err->msg_parm.i[2], err->msg_parm.i[3],
-            err->msg_parm.i[4], err->msg_parm.i[5],
-            err->msg_parm.i[6], err->msg_parm.i[7]);
+	    err->msg_parm.i[0], err->msg_parm.i[1],
+	    err->msg_parm.i[2], err->msg_parm.i[3],
+	    err->msg_parm.i[4], err->msg_parm.i[5],
+	    err->msg_parm.i[6], err->msg_parm.i[7]);
 }


@@ -48,25 +48,25 @@ JMESSAGE(JERR_BAD_CROP_SPEC, "Invalid crop request")
 JMESSAGE(JERR_BAD_DCT_COEF, "DCT coefficient out of range")
 JMESSAGE(JERR_BAD_DCTSIZE, "DCT scaled block size %dx%d not supported")
 JMESSAGE(JERR_BAD_DROP_SAMPLING,
-         "Component index %d: mismatching sampling ratio %d:%d, %d:%d, %c")
+	 "Component index %d: mismatching sampling ratio %d:%d, %d:%d, %c")
 JMESSAGE(JERR_BAD_HUFF_TABLE, "Bogus Huffman table definition")
 JMESSAGE(JERR_BAD_IN_COLORSPACE, "Bogus input colorspace")
 JMESSAGE(JERR_BAD_J_COLORSPACE, "Bogus JPEG colorspace")
 JMESSAGE(JERR_BAD_LENGTH, "Bogus marker length")
 JMESSAGE(JERR_BAD_LIB_VERSION,
-         "Wrong JPEG library version: library is %d, caller expects %d")
+	 "Wrong JPEG library version: library is %d, caller expects %d")
 JMESSAGE(JERR_BAD_MCU_SIZE, "Sampling factors too large for interleaved scan")
 JMESSAGE(JERR_BAD_POOL_ID, "Invalid memory pool code %d")
 JMESSAGE(JERR_BAD_PRECISION, "Unsupported JPEG data precision %d")
 JMESSAGE(JERR_BAD_PROGRESSION,
-         "Invalid progressive parameters Ss=%d Se=%d Ah=%d Al=%d")
+	 "Invalid progressive parameters Ss=%d Se=%d Ah=%d Al=%d")
 JMESSAGE(JERR_BAD_PROG_SCRIPT,
-         "Invalid progressive parameters at scan script entry %d")
+	 "Invalid progressive parameters at scan script entry %d")
 JMESSAGE(JERR_BAD_SAMPLING, "Bogus sampling factors")
 JMESSAGE(JERR_BAD_SCAN_SCRIPT, "Invalid scan script at entry %d")
 JMESSAGE(JERR_BAD_STATE, "Improper call to JPEG library in state %d")
 JMESSAGE(JERR_BAD_STRUCT_SIZE,
-         "JPEG parameter struct mismatch: library thinks size is %u, caller expects %u")
+	 "JPEG parameter struct mismatch: library thinks size is %u, caller expects %u")
 JMESSAGE(JERR_BAD_VIRTUAL_ACCESS, "Bogus virtual array access")
 JMESSAGE(JERR_BUFFER_SIZE, "Buffer passed to JPEG library is too small")
 JMESSAGE(JERR_CANT_SUSPEND, "Suspension not allowed here")
@@ -90,7 +90,7 @@ JMESSAGE(JERR_IMAGE_TOO_BIG, "Maximum supported image dimension is %u pixels")
 JMESSAGE(JERR_INPUT_EMPTY, "Empty input file")
 JMESSAGE(JERR_INPUT_EOF, "Premature end of input file")
 JMESSAGE(JERR_MISMATCHED_QUANT_TABLE,
-         "Cannot transcode due to multiple use of quantization table %d")
+	 "Cannot transcode due to multiple use of quantization table %d")
 JMESSAGE(JERR_MISSING_DATA, "Scan script does not transmit all data")
 JMESSAGE(JERR_MODE_CHANGE, "Invalid color quantization mode change")
 JMESSAGE(JERR_NOTIMPL, "Not implemented yet")
@@ -103,7 +103,7 @@ JMESSAGE(JERR_NO_QUANT_TABLE, "Quantization table 0x%02x was not defined")
 JMESSAGE(JERR_NO_SOI, "Not a JPEG file: starts with 0x%02x 0x%02x")
 JMESSAGE(JERR_OUT_OF_MEMORY, "Insufficient memory (case %d)")
 JMESSAGE(JERR_QUANT_COMPONENTS,
-         "Cannot quantize more than %d color components")
+	 "Cannot quantize more than %d color components")
 JMESSAGE(JERR_QUANT_FEW_COLORS, "Cannot quantize to fewer than %d colors")
 JMESSAGE(JERR_QUANT_MANY_COLORS, "Cannot quantize to more than %d colors")
 JMESSAGE(JERR_SOF_BEFORE, "Invalid JPEG file structure: %s before SOF")
@@ -115,7 +115,7 @@ JMESSAGE(JERR_TFILE_CREATE, "Failed to create temporary file %s")
 JMESSAGE(JERR_TFILE_READ, "Read failed on temporary file")
 JMESSAGE(JERR_TFILE_SEEK, "Seek failed on temporary file")
 JMESSAGE(JERR_TFILE_WRITE,
-         "Write failed on temporary file --- out of disk space?")
+	 "Write failed on temporary file --- out of disk space?")
 JMESSAGE(JERR_TOO_LITTLE_DATA, "Application transferred too few scanlines")
 JMESSAGE(JERR_UNKNOWN_MARKER, "Unsupported marker type 0x%02x")
 JMESSAGE(JERR_VIRTUAL_BUG, "Virtual array controller messed up")
@@ -125,9 +125,9 @@ JMESSAGE(JERR_XMS_WRITE, "Write to XMS failed")
 JMESSAGE(JMSG_COPYRIGHT, JCOPYRIGHT)
 JMESSAGE(JMSG_VERSION, JVERSION)
 JMESSAGE(JTRC_16BIT_TABLES,
-         "Caution: quantization tables are too coarse for baseline JPEG")
+	 "Caution: quantization tables are too coarse for baseline JPEG")
 JMESSAGE(JTRC_ADOBE,
-         "Adobe APP14 marker: version %d, flags 0x%04x 0x%04x, transform %d")
+	 "Adobe APP14 marker: version %d, flags 0x%04x 0x%04x, transform %d")
 JMESSAGE(JTRC_APP0, "Unknown APP0 marker (not JFIF), length %u")
 JMESSAGE(JTRC_APP14, "Unknown APP14 marker (not Adobe), length %u")
 JMESSAGE(JTRC_DAC, "Define Arithmetic Table 0x%02x: 0x%02x")
@@ -140,9 +140,9 @@ JMESSAGE(JTRC_EOI, "End Of Image")
 JMESSAGE(JTRC_HUFFBITS, "        %3d %3d %3d %3d %3d %3d %3d %3d")
 JMESSAGE(JTRC_JFIF, "JFIF APP0 marker: version %d.%02d, density %dx%d  %d")
 JMESSAGE(JTRC_JFIF_BADTHUMBNAILSIZE,
-         "Warning: thumbnail image size does not match data length %u")
+	 "Warning: thumbnail image size does not match data length %u")
 JMESSAGE(JTRC_JFIF_EXTENSION,
-         "JFIF extension marker: type 0x%02x, length %u")
+	 "JFIF extension marker: type 0x%02x, length %u")
 JMESSAGE(JTRC_JFIF_THUMBNAIL, "    with %d x %d thumbnail image")
 JMESSAGE(JTRC_MISC_MARKER, "Miscellaneous marker 0x%02x, length %u")
 JMESSAGE(JTRC_PARMLESS_MARKER, "Unexpected marker 0x%02x")
@@ -153,7 +153,7 @@ JMESSAGE(JTRC_QUANT_SELECTED, "Selected %d colors for quantization")
 JMESSAGE(JTRC_RECOVERY_ACTION, "At marker 0x%02x, recovery action %d")
 JMESSAGE(JTRC_RST, "RST%d")
 JMESSAGE(JTRC_SMOOTH_NOTIMPL,
-         "Smoothing not supported with nonstandard sampling ratios")
+	 "Smoothing not supported with nonstandard sampling ratios")
 JMESSAGE(JTRC_SOF, "Start Of Frame 0x%02x: width=%u, height=%u, components=%d")
 JMESSAGE(JTRC_SOF_COMPONENT, "    Component %d: %dhx%dv q=%d")
 JMESSAGE(JTRC_SOI, "Start of Image")
@@ -163,27 +163,27 @@ JMESSAGE(JTRC_SOS_PARAMS, "  Ss=%d, Se=%d, Ah=%d, Al=%d")
 JMESSAGE(JTRC_TFILE_CLOSE, "Closed temporary file %s")
 JMESSAGE(JTRC_TFILE_OPEN, "Opened temporary file %s")
 JMESSAGE(JTRC_THUMB_JPEG,
-         "JFIF extension marker: JPEG-compressed thumbnail image, length %u")
+	 "JFIF extension marker: JPEG-compressed thumbnail image, length %u")
 JMESSAGE(JTRC_THUMB_PALETTE,
-         "JFIF extension marker: palette thumbnail image, length %u")
+	 "JFIF extension marker: palette thumbnail image, length %u")
 JMESSAGE(JTRC_THUMB_RGB,
-         "JFIF extension marker: RGB thumbnail image, length %u")
+	 "JFIF extension marker: RGB thumbnail image, length %u")
 JMESSAGE(JTRC_UNKNOWN_IDS,
-         "Unrecognized component IDs %d %d %d, assuming YCbCr")
+	 "Unrecognized component IDs %d %d %d, assuming YCbCr")
 JMESSAGE(JTRC_XMS_CLOSE, "Freed XMS handle %u")
 JMESSAGE(JTRC_XMS_OPEN, "Obtained XMS handle %u")
 JMESSAGE(JWRN_ADOBE_XFORM, "Unknown Adobe color transform code %d")
 JMESSAGE(JWRN_ARITH_BAD_CODE, "Corrupt JPEG data: bad arithmetic code")
 JMESSAGE(JWRN_BOGUS_PROGRESSION,
-         "Inconsistent progression sequence for component %d coefficient %d")
+	 "Inconsistent progression sequence for component %d coefficient %d")
 JMESSAGE(JWRN_EXTRANEOUS_DATA,
-         "Corrupt JPEG data: %u extraneous bytes before marker 0x%02x")
+	 "Corrupt JPEG data: %u extraneous bytes before marker 0x%02x")
 JMESSAGE(JWRN_HIT_MARKER, "Corrupt JPEG data: premature end of data segment")
 JMESSAGE(JWRN_HUFF_BAD_CODE, "Corrupt JPEG data: bad Huffman code")
 JMESSAGE(JWRN_JFIF_MAJOR, "Warning: unknown JFIF revision number %d.%02d")
 JMESSAGE(JWRN_JPEG_EOF, "Premature end of JPEG file")
 JMESSAGE(JWRN_MUST_RESYNC,
-         "Corrupt JPEG data: found marker 0x%02x instead of RST%d")
+	 "Corrupt JPEG data: found marker 0x%02x instead of RST%d")
 JMESSAGE(JWRN_NOT_SEQUENTIAL, "Invalid SOS parameters for sequential JPEG")
 JMESSAGE(JWRN_TOO_MUCH_DATA, "Application transferred too many scanlines")

@@ -276,26 +276,26 @@ JMESSAGE(JWRN_TOO_MUCH_DATA, "Application transferred too many scanlines")
   (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)))
 #define TRACEMS3(cinfo,lvl,code,p1,p2,p3)  \
  MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \
-           _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); \
-           (cinfo)->err->msg_code = (code); \
-           (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
+	   _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); \
+	   (cinfo)->err->msg_code = (code); \
+	   (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
 #define TRACEMS4(cinfo,lvl,code,p1,p2,p3,p4)  \
  MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \
-           _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \
-           (cinfo)->err->msg_code = (code); \
-           (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
+	   _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \
+	   (cinfo)->err->msg_code = (code); \
+	   (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
 #define TRACEMS5(cinfo,lvl,code,p1,p2,p3,p4,p5)  \
  MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \
-           _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \
-           _mp[4] = (p5); \
-           (cinfo)->err->msg_code = (code); \
-           (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
+	   _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \
+	   _mp[4] = (p5); \
+	   (cinfo)->err->msg_code = (code); \
+	   (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
 #define TRACEMS8(cinfo,lvl,code,p1,p2,p3,p4,p5,p6,p7,p8)  \
  MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \
-           _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \
-           _mp[4] = (p5); _mp[5] = (p6); _mp[6] = (p7); _mp[7] = (p8); \
-           (cinfo)->err->msg_code = (code); \
-           (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
+	   _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \
+	   _mp[4] = (p5); _mp[5] = (p6); _mp[6] = (p7); _mp[7] = (p8); \
+	   (cinfo)->err->msg_code = (code); \
+	   (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); )
 #define TRACEMSS(cinfo,lvl,code,str)  \
  ((cinfo)->err->msg_code = (code), \
   strncpy((cinfo)->err->msg_parm.s, (str), JMSG_STR_PARM_MAX), \
@@ -2,7 +2,7 @@
 * jfdctflt.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
- * Modified 2003-2009 by Guido Vollbeding.
+ * Modified 2003-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -54,6 +54,8 @@

 /*
 * Perform the forward DCT on one block of samples.
+ *
+ * cK represents cos(K*pi/16).
 */

 GLOBAL(void)
@@ -89,7 +91,7 @@ jpeg_fdct_float (FAST_FLOAT * data, JSAMPARRAY sample_data, JDIMENSION start_col
    tmp11 = tmp1 + tmp2;
    tmp12 = tmp1 - tmp2;

-    /* Apply unsigned->signed conversion */
+    /* Apply unsigned->signed conversion. */
    dataptr[0] = tmp10 + tmp11 - 8 * CENTERJSAMPLE; /* phase 3 */
    dataptr[4] = tmp10 - tmp11;

@@ -2,7 +2,7 @@
 * jfdctfst.c
 *
 * Copyright (C) 1994-1996, Thomas G. Lane.
- * Modified 2003-2009 by Guido Vollbeding.
+ * Modified 2003-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -109,6 +109,8 @@

 /*
 * Perform the forward DCT on one block of samples.
+ *
+ * cK represents cos(K*pi/16).
 */

 GLOBAL(void)
@@ -145,7 +147,7 @@ jpeg_fdct_ifast (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)
    tmp11 = tmp1 + tmp2;
    tmp12 = tmp1 - tmp2;

-    /* Apply unsigned->signed conversion */
+    /* Apply unsigned->signed conversion. */
    dataptr[0] = tmp10 + tmp11 - 8 * CENTERJSAMPLE; /* phase 3 */
    dataptr[4] = tmp10 - tmp11;

@@ -2,7 +2,7 @@
 * jidctflt.c
 *
 * Copyright (C) 1994-1998, Thomas G. Lane.
- * Modified 2010 by Guido Vollbeding.
+ * Modified 2010-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -63,12 +63,14 @@

 /*
 * Perform dequantization and inverse DCT on one block of coefficients.
+ *
+ * cK represents cos(K*pi/16).
 */

 GLOBAL(void)
 jpeg_idct_float (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-                 JCOEFPTR coef_block,
-                 JSAMPARRAY output_buf, JDIMENSION output_col)
+		 JCOEFPTR coef_block,
+		 JSAMPARRAY output_buf, JDIMENSION output_col)
 {
  FAST_FLOAT tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
  FAST_FLOAT tmp10, tmp11, tmp12, tmp13;
@@ -77,7 +79,7 @@ jpeg_idct_float (j_decompress_ptr cinfo, jpeg_component_info * compptr,
  FLOAT_MULT_TYPE * quantptr;
  FAST_FLOAT * wsptr;
  JSAMPROW outptr;
-  JSAMPLE *range_limit = cinfo->sample_range_limit;
+  JSAMPLE *range_limit = IDCT_range_limit(cinfo);
  int ctr;
  FAST_FLOAT workspace[DCTSIZE2]; /* buffers data between passes */

@@ -97,9 +99,9 @@ jpeg_idct_float (j_decompress_ptr cinfo, jpeg_component_info * compptr,
     */

    if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 &&
-        inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 &&
-        inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 &&
-        inptr[DCTSIZE*7] == 0) {
+	inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 &&
+	inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 &&
+	inptr[DCTSIZE*7] == 0) {
      /* AC terms all zero */
      FAST_FLOAT dcval = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]);

@@ -186,13 +188,14 @@ jpeg_idct_float (j_decompress_ptr cinfo, jpeg_component_info * compptr,

    /* Even part */

-    /* Apply signed->unsigned and prepare float->int conversion */
-    z5 = wsptr[0] + ((FAST_FLOAT) CENTERJSAMPLE + (FAST_FLOAT) 0.5);
+    /* Prepare range-limit and float->int conversion */
+    z5 = wsptr[0] + (((FAST_FLOAT) RANGE_CENTER) + ((FAST_FLOAT) 0.5));
    tmp10 = z5 + wsptr[4];
    tmp11 = z5 - wsptr[4];

    tmp13 = wsptr[2] + wsptr[6];
-    tmp12 = (wsptr[2] - wsptr[6]) * ((FAST_FLOAT) 1.414213562) - tmp13;
+    tmp12 = (wsptr[2] - wsptr[6]) *
+	      ((FAST_FLOAT) 1.414213562) - tmp13; /* 2*c4 */

    tmp0 = tmp10 + tmp13;
    tmp3 = tmp10 - tmp13;
@@ -206,27 +209,27 @@ jpeg_idct_float (j_decompress_ptr cinfo, jpeg_component_info * compptr,
    z11 = wsptr[1] + wsptr[7];
    z12 = wsptr[1] - wsptr[7];

-    tmp7 = z11 + z13;
-    tmp11 = (z11 - z13) * ((FAST_FLOAT) 1.414213562);
+    tmp7 = z11 + z13;		/* phase 5 */
+    tmp11 = (z11 - z13) * ((FAST_FLOAT) 1.414213562); /* 2*c4 */

    z5 = (z10 + z12) * ((FAST_FLOAT) 1.847759065); /* 2*c2 */
    tmp10 = z5 - z12 * ((FAST_FLOAT) 1.082392200); /* 2*(c2-c6) */
    tmp12 = z5 - z10 * ((FAST_FLOAT) 2.613125930); /* 2*(c2+c6) */

-    tmp6 = tmp12 - tmp7;
+    tmp6 = tmp12 - tmp7;	/* phase 2 */
    tmp5 = tmp11 - tmp6;
    tmp4 = tmp10 - tmp5;

    /* Final output stage: float->int conversion and range-limit */

-    outptr[0] = range_limit[((int) (tmp0 + tmp7)) & RANGE_MASK];
-    outptr[7] = range_limit[((int) (tmp0 - tmp7)) & RANGE_MASK];
-    outptr[1] = range_limit[((int) (tmp1 + tmp6)) & RANGE_MASK];
-    outptr[6] = range_limit[((int) (tmp1 - tmp6)) & RANGE_MASK];
-    outptr[2] = range_limit[((int) (tmp2 + tmp5)) & RANGE_MASK];
-    outptr[5] = range_limit[((int) (tmp2 - tmp5)) & RANGE_MASK];
-    outptr[3] = range_limit[((int) (tmp3 + tmp4)) & RANGE_MASK];
-    outptr[4] = range_limit[((int) (tmp3 - tmp4)) & RANGE_MASK];
+    outptr[0] = range_limit[(int) (tmp0 + tmp7) & RANGE_MASK];
+    outptr[7] = range_limit[(int) (tmp0 - tmp7) & RANGE_MASK];
+    outptr[1] = range_limit[(int) (tmp1 + tmp6) & RANGE_MASK];
+    outptr[6] = range_limit[(int) (tmp1 - tmp6) & RANGE_MASK];
+    outptr[2] = range_limit[(int) (tmp2 + tmp5) & RANGE_MASK];
+    outptr[5] = range_limit[(int) (tmp2 - tmp5) & RANGE_MASK];
+    outptr[3] = range_limit[(int) (tmp3 + tmp4) & RANGE_MASK];
+    outptr[4] = range_limit[(int) (tmp3 - tmp4) & RANGE_MASK];

    wsptr += DCTSIZE;		/* advance pointer to next row */
  }
@@ -2,6 +2,7 @@
 * jidctfst.c
 *
 * Copyright (C) 1994-1998, Thomas G. Lane.
+ * Modified 2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -129,45 +130,20 @@
 #define DEQUANTIZE(coef,quantval)  (((IFAST_MULT_TYPE) (coef)) * (quantval))
 #else
 #define DEQUANTIZE(coef,quantval)  \
-        DESCALE((coef)*(quantval), IFAST_SCALE_BITS-PASS1_BITS)
-#endif
-
-
-/* Like DESCALE, but applies to a DCTELEM and produces an int.
- * We assume that int right shift is unsigned if INT32 right shift is.
- */
-
-#ifdef RIGHT_SHIFT_IS_UNSIGNED
-#define ISHIFT_TEMPS	DCTELEM ishift_temp;
-#if BITS_IN_JSAMPLE == 8
-#define DCTELEMBITS  16		/* DCTELEM may be 16 or 32 bits */
-#else
-#define DCTELEMBITS  32		/* DCTELEM must be 32 bits */
-#endif
-#define IRIGHT_SHIFT(x,shft)  \
-    ((ishift_temp = (x)) < 0 ? \
-     (ishift_temp >> (shft)) | ((~((DCTELEM) 0)) << (DCTELEMBITS-(shft))) : \
-     (ishift_temp >> (shft)))
-#else
-#define ISHIFT_TEMPS
-#define IRIGHT_SHIFT(x,shft)	((x) >> (shft))
-#endif
-
-#ifdef USE_ACCURATE_ROUNDING
-#define IDESCALE(x,n)  ((int) IRIGHT_SHIFT((x) + (1 << ((n)-1)), n))
-#else
-#define IDESCALE(x,n)  ((int) IRIGHT_SHIFT(x, n))
+	DESCALE((coef)*(quantval), IFAST_SCALE_BITS-PASS1_BITS)
 #endif


 /*
 * Perform dequantization and inverse DCT on one block of coefficients.
+ *
+ * cK represents cos(K*pi/16).
 */

 GLOBAL(void)
 jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-                 JCOEFPTR coef_block,
-                 JSAMPARRAY output_buf, JDIMENSION output_col)
+		 JCOEFPTR coef_block,
+		 JSAMPARRAY output_buf, JDIMENSION output_col)
 {
  DCTELEM tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
  DCTELEM tmp10, tmp11, tmp12, tmp13;
@@ -180,7 +156,7 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
  int ctr;
  int workspace[DCTSIZE2];	/* buffers data between passes */
  SHIFT_TEMPS			/* for DESCALE */
-  ISHIFT_TEMPS			/* for IDESCALE */
+  ISHIFT_TEMPS			/* for IRIGHT_SHIFT */

  /* Pass 1: process columns from input, store into work array. */

@@ -196,11 +172,11 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
     * With typical images and quantization tables, half or more of the
     * column DCT calculations can be simplified this way.
     */
-
+    
    if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 &&
-        inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 &&
-        inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 &&
-        inptr[DCTSIZE*7] == 0) {
+	inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 &&
+	inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 &&
+	inptr[DCTSIZE*7] == 0) {
      /* AC terms all zero */
      int dcval = (int) DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]);

@@ -212,13 +188,13 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
      wsptr[DCTSIZE*5] = dcval;
      wsptr[DCTSIZE*6] = dcval;
      wsptr[DCTSIZE*7] = dcval;
-
+      
      inptr++;			/* advance pointers to next column */
      quantptr++;
      wsptr++;
      continue;
    }
-
+    
    /* Even part */

    tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]);
@@ -236,7 +212,7 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
    tmp3 = tmp10 - tmp13;
    tmp1 = tmp11 + tmp12;
    tmp2 = tmp11 - tmp12;
-
+    
    /* Odd part */

    tmp4 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]);
@@ -253,12 +229,12 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
    tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562); /* 2*c4 */

    z5 = MULTIPLY(z10 + z12, FIX_1_847759065); /* 2*c2 */
-    tmp10 = MULTIPLY(z12, FIX_1_082392200) - z5; /* 2*(c2-c6) */
-    tmp12 = MULTIPLY(z10, - FIX_2_613125930) + z5; /* -2*(c2+c6) */
+    tmp10 = z5 - MULTIPLY(z12, FIX_1_082392200); /* 2*(c2-c6) */
+    tmp12 = z5 - MULTIPLY(z10, FIX_2_613125930); /* 2*(c2+c6) */

    tmp6 = tmp12 - tmp7;	/* phase 2 */
    tmp5 = tmp11 - tmp6;
-    tmp4 = tmp10 + tmp5;
+    tmp4 = tmp10 - tmp5;

    wsptr[DCTSIZE*0] = (int) (tmp0 + tmp7);
    wsptr[DCTSIZE*7] = (int) (tmp0 - tmp7);
@@ -266,21 +242,28 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
    wsptr[DCTSIZE*6] = (int) (tmp1 - tmp6);
    wsptr[DCTSIZE*2] = (int) (tmp2 + tmp5);
    wsptr[DCTSIZE*5] = (int) (tmp2 - tmp5);
-    wsptr[DCTSIZE*4] = (int) (tmp3 + tmp4);
-    wsptr[DCTSIZE*3] = (int) (tmp3 - tmp4);
+    wsptr[DCTSIZE*3] = (int) (tmp3 + tmp4);
+    wsptr[DCTSIZE*4] = (int) (tmp3 - tmp4);

    inptr++;			/* advance pointers to next column */
    quantptr++;
    wsptr++;
  }
-
-  /* Pass 2: process rows from work array, store into output array. */
-  /* Note that we must descale the results by a factor of 8 == 2**3, */
-  /* and also undo the PASS1_BITS scaling. */
+  
+  /* Pass 2: process rows from work array, store into output array.
+   * Note that we must descale the results by a factor of 8 == 2**3,
+   * and also undo the PASS1_BITS scaling.
+   */

  wsptr = workspace;
  for (ctr = 0; ctr < DCTSIZE; ctr++) {
    outptr = output_buf[ctr] + output_col;
+
+    /* Add range center and fudge factor for final descale and range-limit. */
+    z5 = (DCTELEM) wsptr[0] +
+	   ((((DCTELEM) RANGE_CENTER) << (PASS1_BITS+3)) +
+	    (1 << (PASS1_BITS+2)));
+
    /* Rows of zeroes can be exploited in the same way as we did with columns.
     * However, the column calculation has created many nonzero AC terms, so
     * the simplification applies less often (typically 5% to 10% of the time).
@@ -288,14 +271,14 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
     * test takes more time than it's worth.  In that case this section
     * may be commented out.
     */
-
+    
 #ifndef NO_ZERO_ROW_TEST
    if (wsptr[1] == 0 && wsptr[2] == 0 && wsptr[3] == 0 && wsptr[4] == 0 &&
-        wsptr[5] == 0 && wsptr[6] == 0 && wsptr[7] == 0) {
+	wsptr[5] == 0 && wsptr[6] == 0 && wsptr[7] == 0) {
      /* AC terms all zero */
-      JSAMPLE dcval = range_limit[IDESCALE(wsptr[0], PASS1_BITS+3)
-                                  & RANGE_MASK];
-
+      JSAMPLE dcval = range_limit[(int) IRIGHT_SHIFT(z5, PASS1_BITS+3)
+				  & RANGE_MASK];
+      
      outptr[0] = dcval;
      outptr[1] = dcval;
      outptr[2] = dcval;
@@ -309,15 +292,15 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
      continue;
    }
 #endif
-
+    
    /* Even part */

-    tmp10 = ((DCTELEM) wsptr[0] + (DCTELEM) wsptr[4]);
-    tmp11 = ((DCTELEM) wsptr[0] - (DCTELEM) wsptr[4]);
+    tmp10 = z5 + (DCTELEM) wsptr[4];
+    tmp11 = z5 - (DCTELEM) wsptr[4];

-    tmp13 = ((DCTELEM) wsptr[2] + (DCTELEM) wsptr[6]);
-    tmp12 = MULTIPLY((DCTELEM) wsptr[2] - (DCTELEM) wsptr[6], FIX_1_414213562)
-            - tmp13;
+    tmp13 = (DCTELEM) wsptr[2] + (DCTELEM) wsptr[6];
+    tmp12 = MULTIPLY((DCTELEM) wsptr[2] - (DCTELEM) wsptr[6],
+		     FIX_1_414213562) - tmp13; /* 2*c4 */

    tmp0 = tmp10 + tmp13;
    tmp3 = tmp10 - tmp13;
@@ -335,31 +318,31 @@ jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
    tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562); /* 2*c4 */

    z5 = MULTIPLY(z10 + z12, FIX_1_847759065); /* 2*c2 */
-    tmp10 = MULTIPLY(z12, FIX_1_082392200) - z5; /* 2*(c2-c6) */
-    tmp12 = MULTIPLY(z10, - FIX_2_613125930) + z5; /* -2*(c2+c6) */
+    tmp10 = z5 - MULTIPLY(z12, FIX_1_082392200); /* 2*(c2-c6) */
+    tmp12 = z5 - MULTIPLY(z10, FIX_2_613125930); /* 2*(c2+c6) */

    tmp6 = tmp12 - tmp7;	/* phase 2 */
    tmp5 = tmp11 - tmp6;
-    tmp4 = tmp10 + tmp5;
+    tmp4 = tmp10 - tmp5;

    /* Final output stage: scale down by a factor of 8 and range-limit */

-    outptr[0] = range_limit[IDESCALE(tmp0 + tmp7, PASS1_BITS+3)
-                            & RANGE_MASK];
-    outptr[7] = range_limit[IDESCALE(tmp0 - tmp7, PASS1_BITS+3)
-                            & RANGE_MASK];
-    outptr[1] = range_limit[IDESCALE(tmp1 + tmp6, PASS1_BITS+3)
-                            & RANGE_MASK];
-    outptr[6] = range_limit[IDESCALE(tmp1 - tmp6, PASS1_BITS+3)
-                            & RANGE_MASK];
-    outptr[2] = range_limit[IDESCALE(tmp2 + tmp5, PASS1_BITS+3)
-                            & RANGE_MASK];
-    outptr[5] = range_limit[IDESCALE(tmp2 - tmp5, PASS1_BITS+3)
-                            & RANGE_MASK];
-    outptr[4] = range_limit[IDESCALE(tmp3 + tmp4, PASS1_BITS+3)
-                            & RANGE_MASK];
-    outptr[3] = range_limit[IDESCALE(tmp3 - tmp4, PASS1_BITS+3)
-                            & RANGE_MASK];
+    outptr[0] = range_limit[(int) IRIGHT_SHIFT(tmp0 + tmp7, PASS1_BITS+3)
+			    & RANGE_MASK];
+    outptr[7] = range_limit[(int) IRIGHT_SHIFT(tmp0 - tmp7, PASS1_BITS+3)
+			    & RANGE_MASK];
+    outptr[1] = range_limit[(int) IRIGHT_SHIFT(tmp1 + tmp6, PASS1_BITS+3)
+			    & RANGE_MASK];
+    outptr[6] = range_limit[(int) IRIGHT_SHIFT(tmp1 - tmp6, PASS1_BITS+3)
+			    & RANGE_MASK];
+    outptr[2] = range_limit[(int) IRIGHT_SHIFT(tmp2 + tmp5, PASS1_BITS+3)
+			    & RANGE_MASK];
+    outptr[5] = range_limit[(int) IRIGHT_SHIFT(tmp2 - tmp5, PASS1_BITS+3)
+			    & RANGE_MASK];
+    outptr[3] = range_limit[(int) IRIGHT_SHIFT(tmp3 + tmp4, PASS1_BITS+3)
+			    & RANGE_MASK];
+    outptr[4] = range_limit[(int) IRIGHT_SHIFT(tmp3 - tmp4, PASS1_BITS+3)
+			    & RANGE_MASK];

    wsptr += DCTSIZE;		/* advance pointer to next row */
  }
@@ -79,7 +79,7 @@ jpeg_free_large (j_common_ptr cinfo, void FAR * object, size_t sizeofobject)

 GLOBAL(long)
 jpeg_mem_available (j_common_ptr cinfo, long min_bytes_needed,
-                    long max_bytes_needed, long already_allocated)
+		    long max_bytes_needed, long already_allocated)
 {
  return cinfo->mem->max_memory_to_use - already_allocated;
 }
@@ -95,8 +95,8 @@ jpeg_mem_available (j_common_ptr cinfo, long min_bytes_needed,

 METHODDEF(void)
 read_backing_store (j_common_ptr cinfo, backing_store_ptr info,
-                    void FAR * buffer_address,
-                    long file_offset, long byte_count)
+		    void FAR * buffer_address,
+		    long file_offset, long byte_count)
 {
  if (fseek(info->temp_file, file_offset, SEEK_SET))
    ERREXIT(cinfo, JERR_TFILE_SEEK);
@@ -108,8 +108,8 @@ read_backing_store (j_common_ptr cinfo, backing_store_ptr info,

 METHODDEF(void)
 write_backing_store (j_common_ptr cinfo, backing_store_ptr info,
-                     void FAR * buffer_address,
-                     long file_offset, long byte_count)
+		     void FAR * buffer_address,
+		     long file_offset, long byte_count)
 {
  if (fseek(info->temp_file, file_offset, SEEK_SET))
    ERREXIT(cinfo, JERR_TFILE_SEEK);
@@ -139,7 +139,7 @@ close_backing_store (j_common_ptr cinfo, backing_store_ptr info)

 GLOBAL(void)
 jpeg_open_backing_store (j_common_ptr cinfo, backing_store_ptr info,
-                         long total_bytes_needed)
+			 long total_bytes_needed)
 {
  if ((info->temp_file = tmpfile()) == NULL)
    ERREXITS(cinfo, JERR_TFILE_CREATE, "");
@@ -195,19 +195,19 @@ print_mem_stats (j_common_ptr cinfo, int pool_id)
   * This is helpful because message parm array can't handle longs.
   */
  fprintf(stderr, "Freeing pool %d, total space = %ld\n",
-          pool_id, mem->total_space_allocated);
+	  pool_id, mem->total_space_allocated);

  for (lhdr_ptr = mem->large_list[pool_id]; lhdr_ptr != NULL;
       lhdr_ptr = lhdr_ptr->hdr.next) {
    fprintf(stderr, "  Large chunk used %ld\n",
-            (long) lhdr_ptr->hdr.bytes_used);
+	    (long) lhdr_ptr->hdr.bytes_used);
  }

  for (shdr_ptr = mem->small_list[pool_id]; shdr_ptr != NULL;
       shdr_ptr = shdr_ptr->hdr.next) {
    fprintf(stderr, "  Small chunk used %ld free %ld\n",
-            (long) shdr_ptr->hdr.bytes_used,
-            (long) shdr_ptr->hdr.bytes_left);
+	    (long) shdr_ptr->hdr.bytes_used,
+	    (long) shdr_ptr->hdr.bytes_left);
  }
 }

@@ -239,16 +239,16 @@ out_of_memory (j_common_ptr cinfo, int which)
 * machines, but may be too small if longs are 64 bits or more.
 */

-static const size_t first_pool_slop[JPOOL_NUMPOOLS] =
+static const size_t first_pool_slop[JPOOL_NUMPOOLS] = 
 {
-        1600,			/* first PERMANENT pool */
-        16000			/* first IMAGE pool */
+	1600,			/* first PERMANENT pool */
+	16000			/* first IMAGE pool */
 };

-static const size_t extra_pool_slop[JPOOL_NUMPOOLS] =
+static const size_t extra_pool_slop[JPOOL_NUMPOOLS] = 
 {
-        0,			/* additional PERMANENT pools */
-        5000			/* additional IMAGE pools */
+	0,			/* additional PERMANENT pools */
+	5000			/* additional IMAGE pools */
 };

 #define MIN_SLOP  50		/* greater than 0 to avoid futile looping */
@@ -299,10 +299,10 @@ alloc_small (j_common_ptr cinfo, int pool_id, size_t sizeofobject)
    for (;;) {
      hdr_ptr = (small_pool_ptr) jpeg_get_small(cinfo, min_request + slop);
      if (hdr_ptr != NULL)
-        break;
+	break;
      slop /= 2;
      if (slop < MIN_SLOP)	/* give up when it gets real small */
-        out_of_memory(cinfo, 2); /* jpeg_get_small failed */
+	out_of_memory(cinfo, 2); /* jpeg_get_small failed */
    }
    mem->total_space_allocated += min_request + slop;
    /* Success, initialize the new pool header and add to end of list */
@@ -361,7 +361,7 @@ alloc_large (j_common_ptr cinfo, int pool_id, size_t sizeofobject)
    ERREXIT1(cinfo, JERR_BAD_POOL_ID, pool_id);	/* safety check */

  hdr_ptr = (large_pool_ptr) jpeg_get_large(cinfo, sizeofobject +
-                                            SIZEOF(large_pool_hdr));
+					    SIZEOF(large_pool_hdr));
  if (hdr_ptr == NULL)
    out_of_memory(cinfo, 4);	/* jpeg_get_large failed */
  mem->total_space_allocated += sizeofobject + SIZEOF(large_pool_hdr);
@@ -394,7 +394,7 @@ alloc_large (j_common_ptr cinfo, int pool_id, size_t sizeofobject)

 METHODDEF(JSAMPARRAY)
 alloc_sarray (j_common_ptr cinfo, int pool_id,
-              JDIMENSION samplesperrow, JDIMENSION numrows)
+	      JDIMENSION samplesperrow, JDIMENSION numrows)
 /* Allocate a 2-D sample array */
 {
  my_mem_ptr mem = (my_mem_ptr) cinfo->mem;
@@ -405,7 +405,7 @@ alloc_sarray (j_common_ptr cinfo, int pool_id,

  /* Calculate max # of rows allowed in one allocation chunk */
  ltemp = (MAX_ALLOC_CHUNK-SIZEOF(large_pool_hdr)) /
-          ((long) samplesperrow * SIZEOF(JSAMPLE));
+	  ((long) samplesperrow * SIZEOF(JSAMPLE));
  if (ltemp <= 0)
    ERREXIT(cinfo, JERR_WIDTH_OVERFLOW);
  if (ltemp < (long) numrows)
@@ -416,15 +416,15 @@ alloc_sarray (j_common_ptr cinfo, int pool_id,

  /* Get space for row pointers (small object) */
  result = (JSAMPARRAY) alloc_small(cinfo, pool_id,
-                                    (size_t) (numrows * SIZEOF(JSAMPROW)));
+				    (size_t) (numrows * SIZEOF(JSAMPROW)));

  /* Get the rows themselves (large objects) */
  currow = 0;
  while (currow < numrows) {
    rowsperchunk = MIN(rowsperchunk, numrows - currow);
    workspace = (JSAMPROW) alloc_large(cinfo, pool_id,
-        (size_t) ((size_t) rowsperchunk * (size_t) samplesperrow
-                  * SIZEOF(JSAMPLE)));
+	(size_t) ((size_t) rowsperchunk * (size_t) samplesperrow
+		  * SIZEOF(JSAMPLE)));
    for (i = rowsperchunk; i > 0; i--) {
      result[currow++] = workspace;
      workspace += samplesperrow;
@@ -442,7 +442,7 @@ alloc_sarray (j_common_ptr cinfo, int pool_id,

 METHODDEF(JBLOCKARRAY)
 alloc_barray (j_common_ptr cinfo, int pool_id,
-              JDIMENSION blocksperrow, JDIMENSION numrows)
+	      JDIMENSION blocksperrow, JDIMENSION numrows)
 /* Allocate a 2-D coefficient-block array */
 {
  my_mem_ptr mem = (my_mem_ptr) cinfo->mem;
@@ -453,7 +453,7 @@ alloc_barray (j_common_ptr cinfo, int pool_id,

  /* Calculate max # of rows allowed in one allocation chunk */
  ltemp = (MAX_ALLOC_CHUNK-SIZEOF(large_pool_hdr)) /
-          ((long) blocksperrow * SIZEOF(JBLOCK));
+	  ((long) blocksperrow * SIZEOF(JBLOCK));
  if (ltemp <= 0)
    ERREXIT(cinfo, JERR_WIDTH_OVERFLOW);
  if (ltemp < (long) numrows)
@@ -464,15 +464,15 @@ alloc_barray (j_common_ptr cinfo, int pool_id,

  /* Get space for row pointers (small object) */
  result = (JBLOCKARRAY) alloc_small(cinfo, pool_id,
-                                     (size_t) (numrows * SIZEOF(JBLOCKROW)));
+				     (size_t) (numrows * SIZEOF(JBLOCKROW)));

  /* Get the rows themselves (large objects) */
  currow = 0;
  while (currow < numrows) {
    rowsperchunk = MIN(rowsperchunk, numrows - currow);
    workspace = (JBLOCKROW) alloc_large(cinfo, pool_id,
-        (size_t) ((size_t) rowsperchunk * (size_t) blocksperrow
-                  * SIZEOF(JBLOCK)));
+	(size_t) ((size_t) rowsperchunk * (size_t) blocksperrow
+		  * SIZEOF(JBLOCK)));
    for (i = rowsperchunk; i > 0; i--) {
      result[currow++] = workspace;
      workspace += blocksperrow;
@@ -522,8 +522,8 @@ alloc_barray (j_common_ptr cinfo, int pool_id,

 METHODDEF(jvirt_sarray_ptr)
 request_virt_sarray (j_common_ptr cinfo, int pool_id, boolean pre_zero,
-                     JDIMENSION samplesperrow, JDIMENSION numrows,
-                     JDIMENSION maxaccess)
+		     JDIMENSION samplesperrow, JDIMENSION numrows,
+		     JDIMENSION maxaccess)
 /* Request a virtual 2-D sample array */
 {
  my_mem_ptr mem = (my_mem_ptr) cinfo->mem;
@@ -535,7 +535,7 @@ request_virt_sarray (j_common_ptr cinfo, int pool_id, boolean pre_zero,

  /* get control block */
  result = (jvirt_sarray_ptr) alloc_small(cinfo, pool_id,
-                                          SIZEOF(struct jvirt_sarray_control));
+					  SIZEOF(struct jvirt_sarray_control));

  result->mem_buffer = NULL;	/* marks array not yet realized */
  result->rows_in_array = numrows;
@@ -552,8 +552,8 @@ request_virt_sarray (j_common_ptr cinfo, int pool_id, boolean pre_zero,

 METHODDEF(jvirt_barray_ptr)
 request_virt_barray (j_common_ptr cinfo, int pool_id, boolean pre_zero,
-                     JDIMENSION blocksperrow, JDIMENSION numrows,
-                     JDIMENSION maxaccess)
+		     JDIMENSION blocksperrow, JDIMENSION numrows,
+		     JDIMENSION maxaccess)
 /* Request a virtual 2-D coefficient-block array */
 {
  my_mem_ptr mem = (my_mem_ptr) cinfo->mem;
@@ -565,7 +565,7 @@ request_virt_barray (j_common_ptr cinfo, int pool_id, boolean pre_zero,

  /* get control block */
  result = (jvirt_barray_ptr) alloc_small(cinfo, pool_id,
-                                          SIZEOF(struct jvirt_barray_control));
+					  SIZEOF(struct jvirt_barray_control));

  result->mem_buffer = NULL;	/* marks array not yet realized */
  result->rows_in_array = numrows;
@@ -599,17 +599,17 @@ realize_virt_arrays (j_common_ptr cinfo)
  for (sptr = mem->virt_sarray_list; sptr != NULL; sptr = sptr->next) {
    if (sptr->mem_buffer == NULL) { /* if not realized yet */
      space_per_minheight += (long) sptr->maxaccess *
-                             (long) sptr->samplesperrow * SIZEOF(JSAMPLE);
+			     (long) sptr->samplesperrow * SIZEOF(JSAMPLE);
      maximum_space += (long) sptr->rows_in_array *
-                       (long) sptr->samplesperrow * SIZEOF(JSAMPLE);
+		       (long) sptr->samplesperrow * SIZEOF(JSAMPLE);
    }
  }
  for (bptr = mem->virt_barray_list; bptr != NULL; bptr = bptr->next) {
    if (bptr->mem_buffer == NULL) { /* if not realized yet */
      space_per_minheight += (long) bptr->maxaccess *
-                             (long) bptr->blocksperrow * SIZEOF(JBLOCK);
+			     (long) bptr->blocksperrow * SIZEOF(JBLOCK);
      maximum_space += (long) bptr->rows_in_array *
-                       (long) bptr->blocksperrow * SIZEOF(JBLOCK);
+		       (long) bptr->blocksperrow * SIZEOF(JBLOCK);
    }
  }

@@ -618,7 +618,7 @@ realize_virt_arrays (j_common_ptr cinfo)

  /* Determine amount of memory to actually use; this is system-dependent. */
  avail_mem = jpeg_mem_available(cinfo, space_per_minheight, maximum_space,
-                                 mem->total_space_allocated);
+				 mem->total_space_allocated);

  /* If the maximum space needed is available, make all the buffers full
   * height; otherwise parcel it out with the same number of minheights
@@ -641,19 +641,19 @@ realize_virt_arrays (j_common_ptr cinfo)
    if (sptr->mem_buffer == NULL) { /* if not realized yet */
      minheights = ((long) sptr->rows_in_array - 1L) / sptr->maxaccess + 1L;
      if (minheights <= max_minheights) {
-        /* This buffer fits in memory */
-        sptr->rows_in_mem = sptr->rows_in_array;
+	/* This buffer fits in memory */
+	sptr->rows_in_mem = sptr->rows_in_array;
      } else {
-        /* It doesn't fit in memory, create backing store. */
-        sptr->rows_in_mem = (JDIMENSION) (max_minheights * sptr->maxaccess);
-        jpeg_open_backing_store(cinfo, & sptr->b_s_info,
-                                (long) sptr->rows_in_array *
-                                (long) sptr->samplesperrow *
-                                (long) SIZEOF(JSAMPLE));
-        sptr->b_s_open = TRUE;
+	/* It doesn't fit in memory, create backing store. */
+	sptr->rows_in_mem = (JDIMENSION) (max_minheights * sptr->maxaccess);
+	jpeg_open_backing_store(cinfo, & sptr->b_s_info,
+				(long) sptr->rows_in_array *
+				(long) sptr->samplesperrow *
+				(long) SIZEOF(JSAMPLE));
+	sptr->b_s_open = TRUE;
      }
      sptr->mem_buffer = alloc_sarray(cinfo, JPOOL_IMAGE,
-                                      sptr->samplesperrow, sptr->rows_in_mem);
+				      sptr->samplesperrow, sptr->rows_in_mem);
      sptr->rowsperchunk = mem->last_rowsperchunk;
      sptr->cur_start_row = 0;
      sptr->first_undef_row = 0;
@@ -665,19 +665,19 @@ realize_virt_arrays (j_common_ptr cinfo)
    if (bptr->mem_buffer == NULL) { /* if not realized yet */
      minheights = ((long) bptr->rows_in_array - 1L) / bptr->maxaccess + 1L;
      if (minheights <= max_minheights) {
-        /* This buffer fits in memory */
-        bptr->rows_in_mem = bptr->rows_in_array;
+	/* This buffer fits in memory */
+	bptr->rows_in_mem = bptr->rows_in_array;
      } else {
-        /* It doesn't fit in memory, create backing store. */
-        bptr->rows_in_mem = (JDIMENSION) (max_minheights * bptr->maxaccess);
-        jpeg_open_backing_store(cinfo, & bptr->b_s_info,
-                                (long) bptr->rows_in_array *
-                                (long) bptr->blocksperrow *
-                                (long) SIZEOF(JBLOCK));
-        bptr->b_s_open = TRUE;
+	/* It doesn't fit in memory, create backing store. */
+	bptr->rows_in_mem = (JDIMENSION) (max_minheights * bptr->maxaccess);
+	jpeg_open_backing_store(cinfo, & bptr->b_s_info,
+				(long) bptr->rows_in_array *
+				(long) bptr->blocksperrow *
+				(long) SIZEOF(JBLOCK));
+	bptr->b_s_open = TRUE;
      }
      bptr->mem_buffer = alloc_barray(cinfo, JPOOL_IMAGE,
-                                      bptr->blocksperrow, bptr->rows_in_mem);
+				      bptr->blocksperrow, bptr->rows_in_mem);
      bptr->rowsperchunk = mem->last_rowsperchunk;
      bptr->cur_start_row = 0;
      bptr->first_undef_row = 0;
@@ -709,12 +709,12 @@ do_sarray_io (j_common_ptr cinfo, jvirt_sarray_ptr ptr, boolean writing)
    byte_count = rows * bytesperrow;
    if (writing)
      (*ptr->b_s_info.write_backing_store) (cinfo, & ptr->b_s_info,
-                                            (void FAR *) ptr->mem_buffer[i],
-                                            file_offset, byte_count);
+					    (void FAR *) ptr->mem_buffer[i],
+					    file_offset, byte_count);
    else
      (*ptr->b_s_info.read_backing_store) (cinfo, & ptr->b_s_info,
-                                           (void FAR *) ptr->mem_buffer[i],
-                                           file_offset, byte_count);
+					   (void FAR *) ptr->mem_buffer[i],
+					   file_offset, byte_count);
    file_offset += byte_count;
  }
 }
@@ -742,12 +742,12 @@ do_barray_io (j_common_ptr cinfo, jvirt_barray_ptr ptr, boolean writing)
    byte_count = rows * bytesperrow;
    if (writing)
      (*ptr->b_s_info.write_backing_store) (cinfo, & ptr->b_s_info,
-                                            (void FAR *) ptr->mem_buffer[i],
-                                            file_offset, byte_count);
+					    (void FAR *) ptr->mem_buffer[i],
+					    file_offset, byte_count);
    else
      (*ptr->b_s_info.read_backing_store) (cinfo, & ptr->b_s_info,
-                                           (void FAR *) ptr->mem_buffer[i],
-                                           file_offset, byte_count);
+					   (void FAR *) ptr->mem_buffer[i],
+					   file_offset, byte_count);
    file_offset += byte_count;
  }
 }
@@ -755,8 +755,8 @@ do_barray_io (j_common_ptr cinfo, jvirt_barray_ptr ptr, boolean writing)

 METHODDEF(JSAMPARRAY)
 access_virt_sarray (j_common_ptr cinfo, jvirt_sarray_ptr ptr,
-                    JDIMENSION start_row, JDIMENSION num_rows,
-                    boolean writable)
+		    JDIMENSION start_row, JDIMENSION num_rows,
+		    boolean writable)
 /* Access the part of a virtual sample array starting at start_row */
 /* and extending for num_rows rows.  writable is true if  */
 /* caller intends to modify the accessed area. */
@@ -794,7 +794,7 @@ access_virt_sarray (j_common_ptr cinfo, jvirt_sarray_ptr ptr,

      ltemp = (long) end_row - (long) ptr->rows_in_mem;
      if (ltemp < 0)
-        ltemp = 0;		/* don't fall off front end of file */
+	ltemp = 0;		/* don't fall off front end of file */
      ptr->cur_start_row = (JDIMENSION) ltemp;
    }
    /* Read in the selected part of the array.
@@ -810,7 +810,7 @@ access_virt_sarray (j_common_ptr cinfo, jvirt_sarray_ptr ptr,
  if (ptr->first_undef_row < end_row) {
    if (ptr->first_undef_row < start_row) {
      if (writable)		/* writer skipped over a section of array */
-        ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
+	ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
      undef_row = start_row;	/* but reader is allowed to read ahead */
    } else {
      undef_row = ptr->first_undef_row;
@@ -822,12 +822,12 @@ access_virt_sarray (j_common_ptr cinfo, jvirt_sarray_ptr ptr,
      undef_row -= ptr->cur_start_row; /* make indexes relative to buffer */
      end_row -= ptr->cur_start_row;
      while (undef_row < end_row) {
-        FMEMZERO((void FAR *) ptr->mem_buffer[undef_row], bytesperrow);
-        undef_row++;
+	FMEMZERO((void FAR *) ptr->mem_buffer[undef_row], bytesperrow);
+	undef_row++;
      }
    } else {
      if (! writable)		/* reader looking at undefined data */
-        ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
+	ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
    }
  }
  /* Flag the buffer dirty if caller will write in it */
@@ -840,8 +840,8 @@ access_virt_sarray (j_common_ptr cinfo, jvirt_sarray_ptr ptr,

 METHODDEF(JBLOCKARRAY)
 access_virt_barray (j_common_ptr cinfo, jvirt_barray_ptr ptr,
-                    JDIMENSION start_row, JDIMENSION num_rows,
-                    boolean writable)
+		    JDIMENSION start_row, JDIMENSION num_rows,
+		    boolean writable)
 /* Access the part of a virtual block array starting at start_row */
 /* and extending for num_rows rows.  writable is true if  */
 /* caller intends to modify the accessed area. */
@@ -879,7 +879,7 @@ access_virt_barray (j_common_ptr cinfo, jvirt_barray_ptr ptr,

      ltemp = (long) end_row - (long) ptr->rows_in_mem;
      if (ltemp < 0)
-        ltemp = 0;		/* don't fall off front end of file */
+	ltemp = 0;		/* don't fall off front end of file */
      ptr->cur_start_row = (JDIMENSION) ltemp;
    }
    /* Read in the selected part of the array.
@@ -895,7 +895,7 @@ access_virt_barray (j_common_ptr cinfo, jvirt_barray_ptr ptr,
  if (ptr->first_undef_row < end_row) {
    if (ptr->first_undef_row < start_row) {
      if (writable)		/* writer skipped over a section of array */
-        ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
+	ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
      undef_row = start_row;	/* but reader is allowed to read ahead */
    } else {
      undef_row = ptr->first_undef_row;
@@ -907,12 +907,12 @@ access_virt_barray (j_common_ptr cinfo, jvirt_barray_ptr ptr,
      undef_row -= ptr->cur_start_row; /* make indexes relative to buffer */
      end_row -= ptr->cur_start_row;
      while (undef_row < end_row) {
-        FMEMZERO((void FAR *) ptr->mem_buffer[undef_row], bytesperrow);
-        undef_row++;
+	FMEMZERO((void FAR *) ptr->mem_buffer[undef_row], bytesperrow);
+	undef_row++;
      }
    } else {
      if (! writable)		/* reader looking at undefined data */
-        ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
+	ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS);
    }
  }
  /* Flag the buffer dirty if caller will write in it */
@@ -950,15 +950,15 @@ free_pool (j_common_ptr cinfo, int pool_id)

    for (sptr = mem->virt_sarray_list; sptr != NULL; sptr = sptr->next) {
      if (sptr->b_s_open) {	/* there may be no backing store */
-        sptr->b_s_open = FALSE;	/* prevent recursive close if error */
-        (*sptr->b_s_info.close_backing_store) (cinfo, & sptr->b_s_info);
+	sptr->b_s_open = FALSE;	/* prevent recursive close if error */
+	(*sptr->b_s_info.close_backing_store) (cinfo, & sptr->b_s_info);
      }
    }
    mem->virt_sarray_list = NULL;
    for (bptr = mem->virt_barray_list; bptr != NULL; bptr = bptr->next) {
      if (bptr->b_s_open) {	/* there may be no backing store */
-        bptr->b_s_open = FALSE;	/* prevent recursive close if error */
-        (*bptr->b_s_info.close_backing_store) (cinfo, & bptr->b_s_info);
+	bptr->b_s_open = FALSE;	/* prevent recursive close if error */
+	(*bptr->b_s_info.close_backing_store) (cinfo, & bptr->b_s_info);
      }
    }
    mem->virt_barray_list = NULL;
@@ -971,8 +971,8 @@ free_pool (j_common_ptr cinfo, int pool_id)
  while (lhdr_ptr != NULL) {
    large_pool_ptr next_lhdr_ptr = lhdr_ptr->hdr.next;
    space_freed = lhdr_ptr->hdr.bytes_used +
-                  lhdr_ptr->hdr.bytes_left +
-                  SIZEOF(large_pool_hdr);
+		  lhdr_ptr->hdr.bytes_left +
+		  SIZEOF(large_pool_hdr);
    jpeg_free_large(cinfo, (void FAR *) lhdr_ptr, space_freed);
    mem->total_space_allocated -= space_freed;
    lhdr_ptr = next_lhdr_ptr;
@@ -985,8 +985,8 @@ free_pool (j_common_ptr cinfo, int pool_id)
  while (shdr_ptr != NULL) {
    small_pool_ptr next_shdr_ptr = shdr_ptr->hdr.next;
    space_freed = shdr_ptr->hdr.bytes_used +
-                  shdr_ptr->hdr.bytes_left +
-                  SIZEOF(small_pool_hdr);
+		  shdr_ptr->hdr.bytes_left +
+		  SIZEOF(small_pool_hdr);
    jpeg_free_small(cinfo, (void *) shdr_ptr, space_freed);
    mem->total_space_allocated -= space_freed;
    shdr_ptr = next_shdr_ptr;
@@ -1108,9 +1108,9 @@ jinit_memory_mgr (j_common_ptr cinfo)
      char ch = 'x';

      if (sscanf(memenv, "%ld%c", &max_to_use, &ch) > 0) {
-        if (ch == 'm' || ch == 'M')
-          max_to_use *= 1000L;
-        mem->pub.max_memory_to_use = max_to_use * 1000L;
+	if (ch == 'm' || ch == 'M')
+	  max_to_use *= 1000L;
+	mem->pub.max_memory_to_use = max_to_use * 1000L;
      }
    }
  }
@@ -71,7 +71,7 @@ jpeg_free_large (j_common_ptr cinfo, void FAR * object, size_t sizeofobject)

 GLOBAL(long)
 jpeg_mem_available (j_common_ptr cinfo, long min_bytes_needed,
-                    long max_bytes_needed, long already_allocated)
+		    long max_bytes_needed, long already_allocated)
 {
  return max_bytes_needed;
 }
@@ -85,7 +85,7 @@ jpeg_mem_available (j_common_ptr cinfo, long min_bytes_needed,

 GLOBAL(void)
 jpeg_open_backing_store (j_common_ptr cinfo, backing_store_ptr info,
-                         long total_bytes_needed)
+			 long total_bytes_needed)
 {
  ERREXIT(cinfo, JERR_NO_BACKING_STORE);
 }
@@ -46,7 +46,7 @@

 EXTERN(void *) jpeg_get_small JPP((j_common_ptr cinfo, size_t sizeofobject));
 EXTERN(void) jpeg_free_small JPP((j_common_ptr cinfo, void * object,
-                                  size_t sizeofobject));
+				  size_t sizeofobject));

 /*
 * These two functions are used to allocate and release large chunks of
@@ -58,9 +58,9 @@ EXTERN(void) jpeg_free_small JPP((j_common_ptr cinfo, void * object,
 */

 EXTERN(void FAR *) jpeg_get_large JPP((j_common_ptr cinfo,
-                                       size_t sizeofobject));
+				       size_t sizeofobject));
 EXTERN(void) jpeg_free_large JPP((j_common_ptr cinfo, void FAR * object,
-                                  size_t sizeofobject));
+				  size_t sizeofobject));

 /*
 * The macro MAX_ALLOC_CHUNK designates the maximum number of bytes that may
@@ -101,9 +101,9 @@ EXTERN(void) jpeg_free_large JPP((j_common_ptr cinfo, void FAR * object,
 */

 EXTERN(long) jpeg_mem_available JPP((j_common_ptr cinfo,
-                                     long min_bytes_needed,
-                                     long max_bytes_needed,
-                                     long already_allocated));
+				     long min_bytes_needed,
+				     long max_bytes_needed,
+				     long already_allocated));


 /*
@@ -139,15 +139,15 @@ typedef struct backing_store_struct * backing_store_ptr;
 typedef struct backing_store_struct {
  /* Methods for reading/writing/closing this backing-store object */
  JMETHOD(void, read_backing_store, (j_common_ptr cinfo,
-                                     backing_store_ptr info,
-                                     void FAR * buffer_address,
-                                     long file_offset, long byte_count));
+				     backing_store_ptr info,
+				     void FAR * buffer_address,
+				     long file_offset, long byte_count));
  JMETHOD(void, write_backing_store, (j_common_ptr cinfo,
-                                      backing_store_ptr info,
-                                      void FAR * buffer_address,
-                                      long file_offset, long byte_count));
+				      backing_store_ptr info,
+				      void FAR * buffer_address,
+				      long file_offset, long byte_count));
  JMETHOD(void, close_backing_store, (j_common_ptr cinfo,
-                                      backing_store_ptr info));
+				      backing_store_ptr info));

  /* Private fields for system-dependent backing-store management */
 #ifdef USE_MSDOS_MEMMGR
@@ -178,8 +178,8 @@ typedef struct backing_store_struct {
 */

 EXTERN(void) jpeg_open_backing_store JPP((j_common_ptr cinfo,
-                                          backing_store_ptr info,
-                                          long total_bytes_needed));
+					  backing_store_ptr info,
+					  long total_bytes_needed));


 /*
@@ -2,7 +2,7 @@
 * jmorecfg.h
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
- * Modified 1997-2012 by Guido Vollbeding.
+ * Modified 1997-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -15,13 +15,22 @@
 /*
 * Define BITS_IN_JSAMPLE as either
 *   8   for 8-bit sample values (the usual setting)
+ *   9   for 9-bit sample values
+ *   10  for 10-bit sample values
+ *   11  for 11-bit sample values
 *   12  for 12-bit sample values
- * Only 8 and 12 are legal data precisions for lossy JPEG according to the
- * JPEG standard, and the IJG code does not support anything else!
- * We do not support run-time selection of data precision, sorry.
+ * Only 8, 9, 10, 11, and 12 bits sample data precision are supported for
+ * full-feature DCT processing.  Further depths up to 16-bit may be added
+ * later for the lossless modes of operation.
+ * Run-time selection and conversion of data precision will be added later
+ * and are currently not supported, sorry.
+ * Exception:  The transcoding part (jpegtran) supports all settings in a
+ * single instance, since it operates on the level of DCT coefficients and
+ * not sample values.  The DCT coefficients are of the same type (16 bits)
+ * in all cases (see below).
 */

-#define BITS_IN_JSAMPLE  8	/* use 8 or 12 */
+#define BITS_IN_JSAMPLE  8	/* use 8, 9, 10, 11, or 12 */


 /*
@@ -77,6 +86,48 @@ typedef char JSAMPLE;
 #endif /* BITS_IN_JSAMPLE == 8 */


+#if BITS_IN_JSAMPLE == 9
+/* JSAMPLE should be the smallest type that will hold the values 0..511.
+ * On nearly all machines "short" will do nicely.
+ */
+
+typedef short JSAMPLE;
+#define GETJSAMPLE(value)  ((int) (value))
+
+#define MAXJSAMPLE	511
+#define CENTERJSAMPLE	256
+
+#endif /* BITS_IN_JSAMPLE == 9 */
+
+
+#if BITS_IN_JSAMPLE == 10
+/* JSAMPLE should be the smallest type that will hold the values 0..1023.
+ * On nearly all machines "short" will do nicely.
+ */
+
+typedef short JSAMPLE;
+#define GETJSAMPLE(value)  ((int) (value))
+
+#define MAXJSAMPLE	1023
+#define CENTERJSAMPLE	512
+
+#endif /* BITS_IN_JSAMPLE == 10 */
+
+
+#if BITS_IN_JSAMPLE == 11
+/* JSAMPLE should be the smallest type that will hold the values 0..2047.
+ * On nearly all machines "short" will do nicely.
+ */
+
+typedef short JSAMPLE;
+#define GETJSAMPLE(value)  ((int) (value))
+
+#define MAXJSAMPLE	2047
+#define CENTERJSAMPLE	1024
+
+#endif /* BITS_IN_JSAMPLE == 11 */
+
+
 #if BITS_IN_JSAMPLE == 12
 /* JSAMPLE should be the smallest type that will hold the values 0..4095.
 * On nearly all machines "short" will do nicely.
@@ -252,7 +303,10 @@ typedef void noreturn_t;
 * Defining HAVE_BOOLEAN before including jpeglib.h should make it work.
 */

-#ifdef HAVE_BOOLEAN
+#ifndef HAVE_BOOLEAN
+#if defined FALSE || defined TRUE || defined QGLOBAL_H
+/* Qt3 defines FALSE and TRUE as "const" variables in qglobal.h */
+typedef int boolean;
 #ifndef FALSE			/* in case these macros already exist */
 #define FALSE	0		/* values of boolean */
 #endif
@@ -262,6 +316,7 @@ typedef void noreturn_t;
 #else
 typedef enum { FALSE = 0, TRUE = 1 } boolean;
 #endif
+#endif


 /*
@@ -299,11 +354,12 @@ typedef enum { FALSE = 0, TRUE = 1 } boolean;
 #define C_PROGRESSIVE_SUPPORTED	    /* Progressive JPEG? (Requires MULTISCAN)*/
 #define DCT_SCALING_SUPPORTED	    /* Input rescaling via DCT? (Requires DCT_ISLOW)*/
 #define ENTROPY_OPT_SUPPORTED	    /* Optimization of entropy coding parms? */
-/* Note: if you selected 12-bit data precision, it is dangerous to turn off
- * ENTROPY_OPT_SUPPORTED.  The standard Huffman tables are only good for 8-bit
- * precision, so jchuff.c normally uses entropy optimization to compute
- * usable tables for higher precision.  If you don't want to do optimization,
- * you'll have to supply different default Huffman tables.
+/* Note: if you selected more than 8-bit data precision, it is dangerous to
+ * turn off ENTROPY_OPT_SUPPORTED.  The standard Huffman tables are only
+ * good for 8-bit precision, so arithmetic coding is recommended for higher
+ * precision.  The Huffman encoder normally uses entropy optimization to
+ * compute usable tables for higher precision.  Otherwise, you'll have to
+ * supply different default Huffman tables.
 * The exact same statements apply for progressive JPEG: the default tables
 * don't work for progressive mode.  (This may get fixed, however.)
 */
@@ -314,7 +370,7 @@ typedef enum { FALSE = 0, TRUE = 1 } boolean;
 #define D_ARITH_CODING_SUPPORTED    /* Arithmetic coding back end? */
 #define D_MULTISCAN_FILES_SUPPORTED /* Multiple-scan JPEG files? */
 #define D_PROGRESSIVE_SUPPORTED	    /* Progressive JPEG? (Requires MULTISCAN)*/
-#define IDCT_SCALING_SUPPORTED	    /* Output rescaling via IDCT? */
+#define IDCT_SCALING_SUPPORTED	    /* Output rescaling via IDCT? (Requires DCT_ISLOW)*/
 #define SAVE_MARKERS_SUPPORTED	    /* jpeg_save_markers() needed? */
 #define BLOCK_SMOOTHING_SUPPORTED   /* Block smoothing? (Progressive only) */
 #undef  UPSAMPLE_SCALING_SUPPORTED  /* Output rescaling at upsample stage? */
@@ -2,7 +2,7 @@
 * jpegint.h
 *
 * Copyright (C) 1991-1997, Thomas G. Lane.
- * Modified 1997-2011 by Guido Vollbeding.
+ * Modified 1997-2013 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -15,11 +15,11 @@
 /* Declarations for both compression & decompression */

 typedef enum {			/* Operating modes for buffer controllers */
-        JBUF_PASS_THRU,		/* Plain stripwise operation */
-        /* Remaining modes require a full-image buffer to have been created */
-        JBUF_SAVE_SOURCE,	/* Run source subobject only, save output */
-        JBUF_CRANK_DEST,	/* Run dest subobject only, using saved data */
-        JBUF_SAVE_AND_PASS	/* Run both subobjects, save output */
+	JBUF_PASS_THRU,		/* Plain stripwise operation */
+	/* Remaining modes require a full-image buffer to have been created */
+	JBUF_SAVE_SOURCE,	/* Run source subobject only, save output */
+	JBUF_CRANK_DEST,	/* Run dest subobject only, using saved data */
+	JBUF_SAVE_AND_PASS	/* Run both subobjects, save output */
 } J_BUF_MODE;

 /* Values of global_state field (jdapi.c has some dependencies on ordering!) */
@@ -57,54 +57,54 @@ struct jpeg_comp_master {
 struct jpeg_c_main_controller {
  JMETHOD(void, start_pass, (j_compress_ptr cinfo, J_BUF_MODE pass_mode));
  JMETHOD(void, process_data, (j_compress_ptr cinfo,
-                               JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
-                               JDIMENSION in_rows_avail));
+			       JSAMPARRAY input_buf, JDIMENSION *in_row_ctr,
+			       JDIMENSION in_rows_avail));
 };

 /* Compression preprocessing (downsampling input buffer control) */
 struct jpeg_c_prep_controller {
  JMETHOD(void, start_pass, (j_compress_ptr cinfo, J_BUF_MODE pass_mode));
  JMETHOD(void, pre_process_data, (j_compress_ptr cinfo,
-                                   JSAMPARRAY input_buf,
-                                   JDIMENSION *in_row_ctr,
-                                   JDIMENSION in_rows_avail,
-                                   JSAMPIMAGE output_buf,
-                                   JDIMENSION *out_row_group_ctr,
-                                   JDIMENSION out_row_groups_avail));
+				   JSAMPARRAY input_buf,
+				   JDIMENSION *in_row_ctr,
+				   JDIMENSION in_rows_avail,
+				   JSAMPIMAGE output_buf,
+				   JDIMENSION *out_row_group_ctr,
+				   JDIMENSION out_row_groups_avail));
 };

 /* Coefficient buffer control */
 struct jpeg_c_coef_controller {
  JMETHOD(void, start_pass, (j_compress_ptr cinfo, J_BUF_MODE pass_mode));
  JMETHOD(boolean, compress_data, (j_compress_ptr cinfo,
-                                   JSAMPIMAGE input_buf));
+				   JSAMPIMAGE input_buf));
 };

 /* Colorspace conversion */
 struct jpeg_color_converter {
  JMETHOD(void, start_pass, (j_compress_ptr cinfo));
  JMETHOD(void, color_convert, (j_compress_ptr cinfo,
-                                JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
-                                JDIMENSION output_row, int num_rows));
+				JSAMPARRAY input_buf, JSAMPIMAGE output_buf,
+				JDIMENSION output_row, int num_rows));
 };

 /* Downsampling */
 struct jpeg_downsampler {
  JMETHOD(void, start_pass, (j_compress_ptr cinfo));
  JMETHOD(void, downsample, (j_compress_ptr cinfo,
-                             JSAMPIMAGE input_buf, JDIMENSION in_row_index,
-                             JSAMPIMAGE output_buf,
-                             JDIMENSION out_row_group_index));
+			     JSAMPIMAGE input_buf, JDIMENSION in_row_index,
+			     JSAMPIMAGE output_buf,
+			     JDIMENSION out_row_group_index));

  boolean need_context_rows;	/* TRUE if need rows above & below */
 };

 /* Forward DCT (also controls coefficient quantization) */
 typedef JMETHOD(void, forward_DCT_ptr,
-                (j_compress_ptr cinfo, jpeg_component_info * compptr,
-                 JSAMPARRAY sample_data, JBLOCKROW coef_blocks,
-                 JDIMENSION start_row, JDIMENSION start_col,
-                 JDIMENSION num_blocks));
+		(j_compress_ptr cinfo, jpeg_component_info * compptr,
+		 JSAMPARRAY sample_data, JBLOCKROW coef_blocks,
+		 JDIMENSION start_row, JDIMENSION start_col,
+		 JDIMENSION num_blocks));

 struct jpeg_forward_dct {
  JMETHOD(void, start_pass, (j_compress_ptr cinfo));
@@ -129,7 +129,7 @@ struct jpeg_marker_writer {
  /* These routines are exported to allow insertion of extra markers */
  /* Probably only COM and APPn markers should be written this way */
  JMETHOD(void, write_marker_header, (j_compress_ptr cinfo, int marker,
-                                      unsigned int datalen));
+				      unsigned int datalen));
  JMETHOD(void, write_marker_byte, (j_compress_ptr cinfo, int val));
 };

@@ -161,8 +161,8 @@ struct jpeg_input_controller {
 struct jpeg_d_main_controller {
  JMETHOD(void, start_pass, (j_decompress_ptr cinfo, J_BUF_MODE pass_mode));
  JMETHOD(void, process_data, (j_decompress_ptr cinfo,
-                               JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
-                               JDIMENSION out_rows_avail));
+			       JSAMPARRAY output_buf, JDIMENSION *out_row_ctr,
+			       JDIMENSION out_rows_avail));
 };

 /* Coefficient buffer control */
@@ -171,7 +171,7 @@ struct jpeg_d_coef_controller {
  JMETHOD(int, consume_data, (j_decompress_ptr cinfo));
  JMETHOD(void, start_output_pass, (j_decompress_ptr cinfo));
  JMETHOD(int, decompress_data, (j_decompress_ptr cinfo,
-                                 JSAMPIMAGE output_buf));
+				 JSAMPIMAGE output_buf));
  /* Pointer to array of coefficient virtual arrays, or NULL if none */
  jvirt_barray_ptr *coef_arrays;
 };
@@ -180,12 +180,12 @@ struct jpeg_d_coef_controller {
 struct jpeg_d_post_controller {
  JMETHOD(void, start_pass, (j_decompress_ptr cinfo, J_BUF_MODE pass_mode));
  JMETHOD(void, post_process_data, (j_decompress_ptr cinfo,
-                                    JSAMPIMAGE input_buf,
-                                    JDIMENSION *in_row_group_ctr,
-                                    JDIMENSION in_row_groups_avail,
-                                    JSAMPARRAY output_buf,
-                                    JDIMENSION *out_row_ctr,
-                                    JDIMENSION out_rows_avail));
+				    JSAMPIMAGE input_buf,
+				    JDIMENSION *in_row_group_ctr,
+				    JDIMENSION in_row_groups_avail,
+				    JSAMPARRAY output_buf,
+				    JDIMENSION *out_row_ctr,
+				    JDIMENSION out_rows_avail));
 };

 /* Marker reading & parsing */
@@ -211,15 +211,15 @@ struct jpeg_marker_reader {
 /* Entropy decoding */
 struct jpeg_entropy_decoder {
  JMETHOD(void, start_pass, (j_decompress_ptr cinfo));
-  JMETHOD(boolean, decode_mcu, (j_decompress_ptr cinfo,
-                                JBLOCKROW *MCU_data));
+  JMETHOD(boolean, decode_mcu, (j_decompress_ptr cinfo, JBLOCKROW *MCU_data));
+  JMETHOD(void, finish_pass, (j_decompress_ptr cinfo));
 };

 /* Inverse DCT (also performs dequantization) */
 typedef JMETHOD(void, inverse_DCT_method_ptr,
-                (j_decompress_ptr cinfo, jpeg_component_info * compptr,
-                 JCOEFPTR coef_block,
-                 JSAMPARRAY output_buf, JDIMENSION output_col));
+		(j_decompress_ptr cinfo, jpeg_component_info * compptr,
+		 JCOEFPTR coef_block,
+		 JSAMPARRAY output_buf, JDIMENSION output_col));

 struct jpeg_inverse_dct {
  JMETHOD(void, start_pass, (j_decompress_ptr cinfo));
@@ -231,12 +231,12 @@ struct jpeg_inverse_dct {
 struct jpeg_upsampler {
  JMETHOD(void, start_pass, (j_decompress_ptr cinfo));
  JMETHOD(void, upsample, (j_decompress_ptr cinfo,
-                           JSAMPIMAGE input_buf,
-                           JDIMENSION *in_row_group_ctr,
-                           JDIMENSION in_row_groups_avail,
-                           JSAMPARRAY output_buf,
-                           JDIMENSION *out_row_ctr,
-                           JDIMENSION out_rows_avail));
+			   JSAMPIMAGE input_buf,
+			   JDIMENSION *in_row_group_ctr,
+			   JDIMENSION in_row_groups_avail,
+			   JSAMPARRAY output_buf,
+			   JDIMENSION *out_row_ctr,
+			   JDIMENSION out_rows_avail));

  boolean need_context_rows;	/* TRUE if need rows above & below */
 };
@@ -245,16 +245,16 @@ struct jpeg_upsampler {
 struct jpeg_color_deconverter {
  JMETHOD(void, start_pass, (j_decompress_ptr cinfo));
  JMETHOD(void, color_convert, (j_decompress_ptr cinfo,
-                                JSAMPIMAGE input_buf, JDIMENSION input_row,
-                                JSAMPARRAY output_buf, int num_rows));
+				JSAMPIMAGE input_buf, JDIMENSION input_row,
+				JSAMPARRAY output_buf, int num_rows));
 };

 /* Color quantization or color precision reduction */
 struct jpeg_color_quantizer {
  JMETHOD(void, start_pass, (j_decompress_ptr cinfo, boolean is_pre_scan));
  JMETHOD(void, color_quantize, (j_decompress_ptr cinfo,
-                                 JSAMPARRAY input_buf, JSAMPARRAY output_buf,
-                                 int num_rows));
+				 JSAMPARRAY input_buf, JSAMPARRAY output_buf,
+				 int num_rows));
  JMETHOD(void, finish_pass, (j_decompress_ptr cinfo));
  JMETHOD(void, new_color_map, (j_decompress_ptr cinfo));
 };
@@ -281,9 +281,9 @@ struct jpeg_color_quantizer {
 #ifdef RIGHT_SHIFT_IS_UNSIGNED
 #define SHIFT_TEMPS	INT32 shift_temp;
 #define RIGHT_SHIFT(x,shft)  \
-        ((shift_temp = (x)) < 0 ? \
-         (shift_temp >> (shft)) | ((~((INT32) 0)) << (32-(shft))) : \
-         (shift_temp >> (shft)))
+	((shift_temp = (x)) < 0 ? \
+	 (shift_temp >> (shft)) | ((~((INT32) 0)) << (32-(shft))) : \
+	 (shift_temp >> (shft)))
 #else
 #define SHIFT_TEMPS
 #define RIGHT_SHIFT(x,shft)	((x) >> (shft))
@@ -359,13 +359,13 @@ EXTERN(void) jzero_far JPP((void FAR * target, size_t bytestozero));
 /* Compression module initialization routines */
 EXTERN(void) jinit_compress_master JPP((j_compress_ptr cinfo));
 EXTERN(void) jinit_c_master_control JPP((j_compress_ptr cinfo,
-                                         boolean transcode_only));
+					 boolean transcode_only));
 EXTERN(void) jinit_c_main_controller JPP((j_compress_ptr cinfo,
-                                          boolean need_full_buffer));
+					  boolean need_full_buffer));
 EXTERN(void) jinit_c_prep_controller JPP((j_compress_ptr cinfo,
-                                          boolean need_full_buffer));
+					  boolean need_full_buffer));
 EXTERN(void) jinit_c_coef_controller JPP((j_compress_ptr cinfo,
-                                          boolean need_full_buffer));
+					  boolean need_full_buffer));
 EXTERN(void) jinit_color_converter JPP((j_compress_ptr cinfo));
 EXTERN(void) jinit_downsampler JPP((j_compress_ptr cinfo));
 EXTERN(void) jinit_forward_dct JPP((j_compress_ptr cinfo));
@@ -375,11 +375,11 @@ EXTERN(void) jinit_marker_writer JPP((j_compress_ptr cinfo));
 /* Decompression module initialization routines */
 EXTERN(void) jinit_master_decompress JPP((j_decompress_ptr cinfo));
 EXTERN(void) jinit_d_main_controller JPP((j_decompress_ptr cinfo,
-                                          boolean need_full_buffer));
+					  boolean need_full_buffer));
 EXTERN(void) jinit_d_coef_controller JPP((j_decompress_ptr cinfo,
-                                          boolean need_full_buffer));
+					  boolean need_full_buffer));
 EXTERN(void) jinit_d_post_controller JPP((j_decompress_ptr cinfo,
-                                          boolean need_full_buffer));
+					  boolean need_full_buffer));
 EXTERN(void) jinit_input_controller JPP((j_decompress_ptr cinfo));
 EXTERN(void) jinit_marker_reader JPP((j_decompress_ptr cinfo));
 EXTERN(void) jinit_huff_decoder JPP((j_decompress_ptr cinfo));
@@ -397,10 +397,10 @@ EXTERN(void) jinit_memory_mgr JPP((j_common_ptr cinfo));
 EXTERN(long) jdiv_round_up JPP((long a, long b));
 EXTERN(long) jround_up JPP((long a, long b));
 EXTERN(void) jcopy_sample_rows JPP((JSAMPARRAY input_array, int source_row,
-                                    JSAMPARRAY output_array, int dest_row,
-                                    int num_rows, JDIMENSION num_cols));
+				    JSAMPARRAY output_array, int dest_row,
+				    int num_rows, JDIMENSION num_cols));
 EXTERN(void) jcopy_block_row JPP((JBLOCKROW input_row, JBLOCKROW output_row,
-                                  JDIMENSION num_blocks));
+				  JDIMENSION num_blocks));
 /* Constant tables in jutils.c */
 #if 0				/* This table is not actually needed in v6a */
 extern const int jpeg_zigzag_order[]; /* natural coef order to zigzag order */
@@ -2,7 +2,7 @@
 * jpeglib.h
 *
 * Copyright (C) 1991-1998, Thomas G. Lane.
- * Modified 2002-2012 by Guido Vollbeding.
+ * Modified 2002-2015 by Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -39,12 +39,12 @@ extern "C" {

 #define JPEG_LIB_VERSION        90	/* Compatibility version 9.0 */
 #define JPEG_LIB_VERSION_MAJOR  9
-#define JPEG_LIB_VERSION_MINOR  0
+#define JPEG_LIB_VERSION_MINOR  2


 /* Various constants determining the sizes of things.
- * All of these are specified by the JPEG standard, so don't change them
- * if you want to be compatible.
+ * All of these are specified by the JPEG standard,
+ * so don't change them if you want to be compatible.
 */

 #define DCTSIZE		    8	/* The basic DCT block is 8x8 coefficients */
@@ -109,7 +109,7 @@ typedef struct {
 typedef struct {
  /* These two fields directly represent the contents of a JPEG DHT marker */
  UINT8 bits[17];		/* bits[k] = # of symbols with codes of */
-                                /* length k bits; bits[0] is unused */
+				/* length k bits; bits[0] is unused */
  UINT8 huffval[256];		/* The symbols, in order of incr code length */
  /* This field is used only during compression.  It's initialized FALSE when
   * the table is created, and set TRUE when it's been output to the file.
@@ -137,9 +137,9 @@ typedef struct {
  /* The decompressor output side may not use these variables. */
  int dc_tbl_no;		/* DC entropy table selector (0..3) */
  int ac_tbl_no;		/* AC entropy table selector (0..3) */
-
+  
  /* Remaining fields should be treated as private by applications. */
-
+  
  /* These values are computed during compression or decompression startup: */
  /* Component's size in DCT blocks.
   * Any dummy blocks added to complete an MCU are not counted; therefore
@@ -157,16 +157,21 @@ typedef struct {
  /* The downsampled dimensions are the component's actual, unpadded number
   * of samples at the main buffer (preprocessing/compression interface);
   * DCT scaling is included, so
-   * downsampled_width = ceil(image_width * Hi/Hmax * DCT_h_scaled_size/DCTSIZE)
+   * downsampled_width =
+   *   ceil(image_width * Hi/Hmax * DCT_h_scaled_size/block_size)
   * and similarly for height.
   */
  JDIMENSION downsampled_width;	 /* actual width in samples */
  JDIMENSION downsampled_height; /* actual height in samples */
-  /* This flag is used only for decompression.  In cases where some of the
-   * components will be ignored (eg grayscale output from YCbCr image),
-   * we can skip most computations for the unused components.
+  /* For decompression, in cases where some of the components will be
+   * ignored (eg grayscale output from YCbCr image), we can skip most
+   * computations for the unused components.
+   * For compression, some of the components will need further quantization
+   * scale by factor of 2 after DCT (eg BG_YCC output from normal RGB input).
+   * The field is first set TRUE for decompression, FALSE for compression
+   * in initial_setup, and then adapted in color conversion setup.
   */
-  boolean component_needed;	/* do we need the value of this component? */
+  boolean component_needed;

  /* These values are computed before starting a scan of the component. */
  /* The decompressor output side may not use these variables. */
@@ -213,27 +218,29 @@ struct jpeg_marker_struct {
 /* Known color spaces. */

 typedef enum {
-        JCS_UNKNOWN,		/* error/unspecified */
-        JCS_GRAYSCALE,		/* monochrome */
-        JCS_RGB,		/* red/green/blue */
-        JCS_YCbCr,		/* Y/Cb/Cr (also known as YUV) */
-        JCS_CMYK,		/* C/M/Y/K */
-        JCS_YCCK		/* Y/Cb/Cr/K */
+	JCS_UNKNOWN,		/* error/unspecified */
+	JCS_GRAYSCALE,		/* monochrome */
+	JCS_RGB,		/* red/green/blue, standard RGB (sRGB) */
+	JCS_YCbCr,		/* Y/Cb/Cr (also known as YUV), standard YCC */
+	JCS_CMYK,		/* C/M/Y/K */
+	JCS_YCCK,		/* Y/Cb/Cr/K */
+	JCS_BG_RGB,		/* big gamut red/green/blue, bg-sRGB */
+	JCS_BG_YCC		/* big gamut Y/Cb/Cr, bg-sYCC */
 } J_COLOR_SPACE;

 /* Supported color transforms. */

 typedef enum {
-        JCT_NONE           = 0,
-        JCT_SUBTRACT_GREEN = 1
+	JCT_NONE           = 0,
+	JCT_SUBTRACT_GREEN = 1
 } J_COLOR_TRANSFORM;

 /* DCT/IDCT algorithm options. */

 typedef enum {
-        JDCT_ISLOW,		/* slow but accurate integer algorithm */
-        JDCT_IFAST,		/* faster, less accurate integer method */
-        JDCT_FLOAT		/* floating-point: accurate, fast on fast HW */
+	JDCT_ISLOW,		/* slow but accurate integer algorithm */
+	JDCT_IFAST,		/* faster, less accurate integer method */
+	JDCT_FLOAT		/* floating-point: accurate, fast on fast HW */
 } J_DCT_METHOD;

 #ifndef JDCT_DEFAULT		/* may be overridden in jconfig.h */
@@ -246,9 +253,9 @@ typedef enum {
 /* Dithering options for decompression. */

 typedef enum {
-        JDITHER_NONE,		/* no dithering */
-        JDITHER_ORDERED,	/* simple ordered dither */
-        JDITHER_FS		/* Floyd-Steinberg error diffusion dither */
+	JDITHER_NONE,		/* no dithering */
+	JDITHER_ORDERED,	/* simple ordered dither */
+	JDITHER_FS		/* Floyd-Steinberg error diffusion dither */
 } J_DITHER_MODE;


@@ -407,7 +414,7 @@ struct jpeg_compress_struct {
   * There are v_samp_factor * DCTSIZE sample rows of each component in an
   * "iMCU" (interleaved MCU) row.
   */
-
+  
  /*
   * These fields are valid during any one scan.
   * They describe the components and MCUs actually appearing in the scan.
@@ -415,10 +422,10 @@ struct jpeg_compress_struct {
  int comps_in_scan;		/* # of JPEG components in this scan */
  jpeg_component_info * cur_comp_info[MAX_COMPS_IN_SCAN];
  /* *cur_comp_info[i] describes component that appears i'th in SOS */
-
+  
  JDIMENSION MCUs_per_row;	/* # of MCUs across the image */
  JDIMENSION MCU_rows_in_scan;	/* # of MCU rows in the image */
-
+  
  int blocks_in_MCU;		/* # of DCT blocks per MCU */
  int MCU_membership[C_MAX_BLOCKS_IN_MCU];
  /* MCU_membership[i] is index in cur_comp_info of component owning */
@@ -704,7 +711,7 @@ struct jpeg_error_mgr {
 #define JMSG_LENGTH_MAX  200	/* recommended size of format_message buffer */
  /* Reset error state variables at start of a new image */
  JMETHOD(void, reset_error_mgr, (j_common_ptr cinfo));
-
+  
  /* The message ID code and any parameters are saved here.
   * A message can have one string parameter or up to 8 int parameters.
   */
@@ -714,11 +721,11 @@ struct jpeg_error_mgr {
    int i[8];
    char s[JMSG_STR_PARM_MAX];
  } msg_parm;
-
+  
  /* Standard state variables for error facility */
-
+  
  int trace_level;		/* max msg_level that will be displayed */
-
+  
  /* For recoverable corrupt-data errors, we emit a warning message,
   * but keep going unless emit_message chooses to abort.  emit_message
   * should count warnings in num_warnings.  The surrounding application
@@ -808,38 +815,38 @@ typedef struct jvirt_barray_control * jvirt_barray_ptr;
 struct jpeg_memory_mgr {
  /* Method pointers */
  JMETHOD(void *, alloc_small, (j_common_ptr cinfo, int pool_id,
-                                size_t sizeofobject));
+				size_t sizeofobject));
  JMETHOD(void FAR *, alloc_large, (j_common_ptr cinfo, int pool_id,
-                                     size_t sizeofobject));
+				     size_t sizeofobject));
  JMETHOD(JSAMPARRAY, alloc_sarray, (j_common_ptr cinfo, int pool_id,
-                                     JDIMENSION samplesperrow,
-                                     JDIMENSION numrows));
+				     JDIMENSION samplesperrow,
+				     JDIMENSION numrows));
  JMETHOD(JBLOCKARRAY, alloc_barray, (j_common_ptr cinfo, int pool_id,
-                                      JDIMENSION blocksperrow,
-                                      JDIMENSION numrows));
+				      JDIMENSION blocksperrow,
+				      JDIMENSION numrows));
  JMETHOD(jvirt_sarray_ptr, request_virt_sarray, (j_common_ptr cinfo,
-                                                  int pool_id,
-                                                  boolean pre_zero,
-                                                  JDIMENSION samplesperrow,
-                                                  JDIMENSION numrows,
-                                                  JDIMENSION maxaccess));
+						  int pool_id,
+						  boolean pre_zero,
+						  JDIMENSION samplesperrow,
+						  JDIMENSION numrows,
+						  JDIMENSION maxaccess));
  JMETHOD(jvirt_barray_ptr, request_virt_barray, (j_common_ptr cinfo,
-                                                  int pool_id,
-                                                  boolean pre_zero,
-                                                  JDIMENSION blocksperrow,
-                                                  JDIMENSION numrows,
-                                                  JDIMENSION maxaccess));
+						  int pool_id,
+						  boolean pre_zero,
+						  JDIMENSION blocksperrow,
+						  JDIMENSION numrows,
+						  JDIMENSION maxaccess));
  JMETHOD(void, realize_virt_arrays, (j_common_ptr cinfo));
  JMETHOD(JSAMPARRAY, access_virt_sarray, (j_common_ptr cinfo,
-                                           jvirt_sarray_ptr ptr,
-                                           JDIMENSION start_row,
-                                           JDIMENSION num_rows,
-                                           boolean writable));
+					   jvirt_sarray_ptr ptr,
+					   JDIMENSION start_row,
+					   JDIMENSION num_rows,
+					   boolean writable));
  JMETHOD(JBLOCKARRAY, access_virt_barray, (j_common_ptr cinfo,
-                                            jvirt_barray_ptr ptr,
-                                            JDIMENSION start_row,
-                                            JDIMENSION num_rows,
-                                            boolean writable));
+					    jvirt_barray_ptr ptr,
+					    JDIMENSION start_row,
+					    JDIMENSION num_rows,
+					    boolean writable));
  JMETHOD(void, free_pool, (j_common_ptr cinfo, int pool_id));
  JMETHOD(void, self_destruct, (j_common_ptr cinfo));

@@ -876,7 +883,7 @@ typedef JMETHOD(boolean, jpeg_marker_parser_method, (j_decompress_ptr cinfo));
 /* Short forms of external names for systems with brain-damaged linkers.
 * We shorten external names to be unique in the first six letters, which
 * is good enough for all known systems.
- * (If your compiler itself needs names to be unique in less than 15
+ * (If your compiler itself needs names to be unique in less than 15 
 * characters, you are out of luck.  Get a better compiler.)
 */

@@ -939,7 +946,7 @@ typedef JMETHOD(boolean, jpeg_marker_parser_method, (j_decompress_ptr cinfo));

 /* Default error-management setup */
 EXTERN(struct jpeg_error_mgr *) jpeg_std_error
-        JPP((struct jpeg_error_mgr * err));
+	JPP((struct jpeg_error_mgr * err));

 /* Initialization of JPEG compression objects.
 * jpeg_create_compress() and jpeg_create_decompress() are the exported
@@ -950,14 +957,14 @@ EXTERN(struct jpeg_error_mgr *) jpeg_std_error
 */
 #define jpeg_create_compress(cinfo) \
    jpeg_CreateCompress((cinfo), JPEG_LIB_VERSION, \
-                        (size_t) sizeof(struct jpeg_compress_struct))
+			(size_t) sizeof(struct jpeg_compress_struct))
 #define jpeg_create_decompress(cinfo) \
    jpeg_CreateDecompress((cinfo), JPEG_LIB_VERSION, \
-                          (size_t) sizeof(struct jpeg_decompress_struct))
+			  (size_t) sizeof(struct jpeg_decompress_struct))
 EXTERN(void) jpeg_CreateCompress JPP((j_compress_ptr cinfo,
-                                      int version, size_t structsize));
+				      int version, size_t structsize));
 EXTERN(void) jpeg_CreateDecompress JPP((j_decompress_ptr cinfo,
-                                        int version, size_t structsize));
+					int version, size_t structsize));
 /* Destruction of JPEG compression objects */
 EXTERN(void) jpeg_destroy_compress JPP((j_compress_ptr cinfo));
 EXTERN(void) jpeg_destroy_decompress JPP((j_decompress_ptr cinfo));
@@ -969,42 +976,42 @@ EXTERN(void) jpeg_stdio_src JPP((j_decompress_ptr cinfo, FILE * infile));

 /* Data source and destination managers: memory buffers. */
 EXTERN(void) jpeg_mem_dest JPP((j_compress_ptr cinfo,
-                               unsigned char ** outbuffer,
-                               unsigned long * outsize));
+			       unsigned char ** outbuffer,
+			       unsigned long * outsize));
 EXTERN(void) jpeg_mem_src JPP((j_decompress_ptr cinfo,
-                              unsigned char * inbuffer,
-                              unsigned long insize));
+			      const unsigned char * inbuffer,
+			      unsigned long insize));

 /* Default parameter setup for compression */
 EXTERN(void) jpeg_set_defaults JPP((j_compress_ptr cinfo));
 /* Compression parameter setup aids */
 EXTERN(void) jpeg_set_colorspace JPP((j_compress_ptr cinfo,
-                                      J_COLOR_SPACE colorspace));
+				      J_COLOR_SPACE colorspace));
 EXTERN(void) jpeg_default_colorspace JPP((j_compress_ptr cinfo));
 EXTERN(void) jpeg_set_quality JPP((j_compress_ptr cinfo, int quality,
-                                   boolean force_baseline));
+				   boolean force_baseline));
 EXTERN(void) jpeg_set_linear_quality JPP((j_compress_ptr cinfo,
-                                          int scale_factor,
-                                          boolean force_baseline));
+					  int scale_factor,
+					  boolean force_baseline));
 EXTERN(void) jpeg_default_qtables JPP((j_compress_ptr cinfo,
-                                       boolean force_baseline));
+				       boolean force_baseline));
 EXTERN(void) jpeg_add_quant_table JPP((j_compress_ptr cinfo, int which_tbl,
-                                       const unsigned int *basic_table,
-                                       int scale_factor,
-                                       boolean force_baseline));
+				       const unsigned int *basic_table,
+				       int scale_factor,
+				       boolean force_baseline));
 EXTERN(int) jpeg_quality_scaling JPP((int quality));
 EXTERN(void) jpeg_simple_progression JPP((j_compress_ptr cinfo));
 EXTERN(void) jpeg_suppress_tables JPP((j_compress_ptr cinfo,
-                                       boolean suppress));
+				       boolean suppress));
 EXTERN(JQUANT_TBL *) jpeg_alloc_quant_table JPP((j_common_ptr cinfo));
 EXTERN(JHUFF_TBL *) jpeg_alloc_huff_table JPP((j_common_ptr cinfo));

 /* Main entry points for compression */
 EXTERN(void) jpeg_start_compress JPP((j_compress_ptr cinfo,
-                                      boolean write_all_tables));
+				      boolean write_all_tables));
 EXTERN(JDIMENSION) jpeg_write_scanlines JPP((j_compress_ptr cinfo,
-                                             JSAMPARRAY scanlines,
-                                             JDIMENSION num_lines));
+					     JSAMPARRAY scanlines,
+					     JDIMENSION num_lines));
 EXTERN(void) jpeg_finish_compress JPP((j_compress_ptr cinfo));

 /* Precalculate JPEG dimensions for current compression parameters. */
@@ -1012,25 +1019,25 @@ EXTERN(void) jpeg_calc_jpeg_dimensions JPP((j_compress_ptr cinfo));

 /* Replaces jpeg_write_scanlines when writing raw downsampled data. */
 EXTERN(JDIMENSION) jpeg_write_raw_data JPP((j_compress_ptr cinfo,
-                                            JSAMPIMAGE data,
-                                            JDIMENSION num_lines));
+					    JSAMPIMAGE data,
+					    JDIMENSION num_lines));

 /* Write a special marker.  See libjpeg.txt concerning safe usage. */
 EXTERN(void) jpeg_write_marker
-        JPP((j_compress_ptr cinfo, int marker,
-             const JOCTET * dataptr, unsigned int datalen));
+	JPP((j_compress_ptr cinfo, int marker,
+	     const JOCTET * dataptr, unsigned int datalen));
 /* Same, but piecemeal. */
 EXTERN(void) jpeg_write_m_header
-        JPP((j_compress_ptr cinfo, int marker, unsigned int datalen));
+	JPP((j_compress_ptr cinfo, int marker, unsigned int datalen));
 EXTERN(void) jpeg_write_m_byte
-        JPP((j_compress_ptr cinfo, int val));
+	JPP((j_compress_ptr cinfo, int val));

 /* Alternate compression function: just write an abbreviated table file */
 EXTERN(void) jpeg_write_tables JPP((j_compress_ptr cinfo));

 /* Decompression startup: read start of JPEG datastream to see what's there */
 EXTERN(int) jpeg_read_header JPP((j_decompress_ptr cinfo,
-                                  boolean require_image));
+				  boolean require_image));
 /* Return value is one of: */
 #define JPEG_SUSPENDED		0 /* Suspended due to lack of input data */
 #define JPEG_HEADER_OK		1 /* Found valid image datastream */
@@ -1044,19 +1051,19 @@ EXTERN(int) jpeg_read_header JPP((j_decompress_ptr cinfo,
 /* Main entry points for decompression */
 EXTERN(boolean) jpeg_start_decompress JPP((j_decompress_ptr cinfo));
 EXTERN(JDIMENSION) jpeg_read_scanlines JPP((j_decompress_ptr cinfo,
-                                            JSAMPARRAY scanlines,
-                                            JDIMENSION max_lines));
+					    JSAMPARRAY scanlines,
+					    JDIMENSION max_lines));
 EXTERN(boolean) jpeg_finish_decompress JPP((j_decompress_ptr cinfo));

 /* Replaces jpeg_read_scanlines when reading raw downsampled data. */
 EXTERN(JDIMENSION) jpeg_read_raw_data JPP((j_decompress_ptr cinfo,
-                                           JSAMPIMAGE data,
-                                           JDIMENSION max_lines));
+					   JSAMPIMAGE data,
+					   JDIMENSION max_lines));

 /* Additional entry points for buffered-image mode. */
 EXTERN(boolean) jpeg_has_multiple_scans JPP((j_decompress_ptr cinfo));
 EXTERN(boolean) jpeg_start_output JPP((j_decompress_ptr cinfo,
-                                       int scan_number));
+				       int scan_number));
 EXTERN(boolean) jpeg_finish_output JPP((j_decompress_ptr cinfo));
 EXTERN(boolean) jpeg_input_complete JPP((j_decompress_ptr cinfo));
 EXTERN(void) jpeg_new_colormap JPP((j_decompress_ptr cinfo));
@@ -1074,20 +1081,20 @@ EXTERN(void) jpeg_calc_output_dimensions JPP((j_decompress_ptr cinfo));

 /* Control saving of COM and APPn markers into marker_list. */
 EXTERN(void) jpeg_save_markers
-        JPP((j_decompress_ptr cinfo, int marker_code,
-             unsigned int length_limit));
+	JPP((j_decompress_ptr cinfo, int marker_code,
+	     unsigned int length_limit));

 /* Install a special processing method for COM or APPn markers. */
 EXTERN(void) jpeg_set_marker_processor
-        JPP((j_decompress_ptr cinfo, int marker_code,
-             jpeg_marker_parser_method routine));
+	JPP((j_decompress_ptr cinfo, int marker_code,
+	     jpeg_marker_parser_method routine));

 /* Read or write raw DCT coefficients --- useful for lossless transcoding. */
 EXTERN(jvirt_barray_ptr *) jpeg_read_coefficients JPP((j_decompress_ptr cinfo));
 EXTERN(void) jpeg_write_coefficients JPP((j_compress_ptr cinfo,
-                                          jvirt_barray_ptr * coef_arrays));
+					  jvirt_barray_ptr * coef_arrays));
 EXTERN(void) jpeg_copy_critical_parameters JPP((j_decompress_ptr srcinfo,
-                                                j_compress_ptr dstinfo));
+						j_compress_ptr dstinfo));

 /* If you choose to abort compression or decompression before completing
 * jpeg_finish_(de)compress, then you need to clean up to release memory,
@@ -1106,7 +1113,7 @@ EXTERN(void) jpeg_destroy JPP((j_common_ptr cinfo));

 /* Default restart-marker-resync procedure for use by data source modules */
 EXTERN(boolean) jpeg_resync_to_restart JPP((j_decompress_ptr cinfo,
-                                            int desired));
+					    int desired));


 /* These marker codes are exported since applications and data source modules
@@ -231,7 +231,7 @@ select_ncolors (j_decompress_ptr cinfo, int Ncolors[])
      temp = total_colors / Ncolors[j];
      temp *= Ncolors[j]+1;	/* done in long arith to avoid oflo */
      if (temp > (long) max_colors)
-        break;			/* won't fit, done with this pass */
+	break;			/* won't fit, done with this pass */
      Ncolors[j]++;		/* OK, apply the increment */
      total_colors = (int) temp;
      changed = TRUE;
@@ -284,8 +284,8 @@ create_colormap (j_decompress_ptr cinfo)
  /* Report selected color counts */
  if (cinfo->out_color_components == 3)
    TRACEMS4(cinfo, 1, JTRC_QUANT_3_NCOLORS,
-             total_colors, cquantize->Ncolors[0],
-             cquantize->Ncolors[1], cquantize->Ncolors[2]);
+	     total_colors, cquantize->Ncolors[0],
+	     cquantize->Ncolors[1], cquantize->Ncolors[2]);
  else
    TRACEMS1(cinfo, 1, JTRC_QUANT_NCOLORS, total_colors);

@@ -310,9 +310,9 @@ create_colormap (j_decompress_ptr cinfo)
      val = output_value(cinfo, i, j, nci-1);
      /* Fill in all colormap entries that have this value of this component */
      for (ptr = j * blksize; ptr < total_colors; ptr += blkdist) {
-        /* fill in blksize entries beginning at ptr */
-        for (k = 0; k < blksize; k++)
-          colormap[i][ptr+k] = (JSAMPLE) val;
+	/* fill in blksize entries beginning at ptr */
+	for (k = 0; k < blksize; k++)
+	  colormap[i][ptr+k] = (JSAMPLE) val;
      }
    }
    blkdist = blksize;		/* blksize of this color is blkdist of next */
@@ -374,15 +374,15 @@ create_colorindex (j_decompress_ptr cinfo)
    k = largest_input_value(cinfo, i, 0, nci-1);
    for (j = 0; j <= MAXJSAMPLE; j++) {
      while (j > k)		/* advance val if past boundary */
-        k = largest_input_value(cinfo, i, ++val, nci-1);
+	k = largest_input_value(cinfo, i, ++val, nci-1);
      /* premultiply so that no multiplication needed in main processing */
      indexptr[j] = (JSAMPLE) (val * blksize);
    }
    /* Pad at both ends if necessary */
    if (pad)
      for (j = 1; j <= MAXJSAMPLE; j++) {
-        indexptr[-j] = indexptr[0];
-        indexptr[MAXJSAMPLE+j] = indexptr[MAXJSAMPLE];
+	indexptr[-j] = indexptr[0];
+	indexptr[MAXJSAMPLE+j] = indexptr[MAXJSAMPLE];
      }
  }
 }
@@ -402,7 +402,7 @@ make_odither_array (j_decompress_ptr cinfo, int ncolors)

  odither = (ODITHER_MATRIX_PTR)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(ODITHER_MATRIX));
+				SIZEOF(ODITHER_MATRIX));
  /* The inter-value distance for this color is MAXJSAMPLE/(ncolors-1).
   * Hence the dither value for the matrix cell with fill order f
   * (f=0..N-1) should be (N-1-2*f)/(2*N) * MAXJSAMPLE/(ncolors-1).
@@ -412,7 +412,7 @@ make_odither_array (j_decompress_ptr cinfo, int ncolors)
  for (j = 0; j < ODITHER_SIZE; j++) {
    for (k = 0; k < ODITHER_SIZE; k++) {
      num = ((INT32) (ODITHER_CELLS-1 - 2*((int)base_dither_matrix[j][k])))
-            * MAXJSAMPLE;
+	    * MAXJSAMPLE;
      /* Ensure round towards zero despite C's lack of consistency
       * about rounding negative values in integer division...
       */
@@ -425,7 +425,7 @@ make_odither_array (j_decompress_ptr cinfo, int ncolors)

 /*
 * Create the ordered-dither tables.
- * Components having the same number of representative colors may
+ * Components having the same number of representative colors may 
 * share a dither table.
 */

@@ -441,8 +441,8 @@ create_odither_tables (j_decompress_ptr cinfo)
    odither = NULL;		/* search for matching prior component */
    for (j = 0; j < i; j++) {
      if (nci == cquantize->Ncolors[j]) {
-        odither = cquantize->odither[j];
-        break;
+	odither = cquantize->odither[j];
+	break;
      }
    }
    if (odither == NULL)	/* need a new table? */
@@ -458,7 +458,7 @@ create_odither_tables (j_decompress_ptr cinfo)

 METHODDEF(void)
 color_quantize (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
-                JSAMPARRAY output_buf, int num_rows)
+		JSAMPARRAY output_buf, int num_rows)
 /* General case, no dithering */
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
@@ -476,7 +476,7 @@ color_quantize (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
    for (col = width; col > 0; col--) {
      pixcode = 0;
      for (ci = 0; ci < nc; ci++) {
-        pixcode += GETJSAMPLE(colorindex[ci][GETJSAMPLE(*ptrin++)]);
+	pixcode += GETJSAMPLE(colorindex[ci][GETJSAMPLE(*ptrin++)]);
      }
      *ptrout++ = (JSAMPLE) pixcode;
    }
@@ -486,7 +486,7 @@ color_quantize (j_decompress_ptr cinfo, JSAMPARRAY input_buf,

 METHODDEF(void)
 color_quantize3 (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
-                 JSAMPARRAY output_buf, int num_rows)
+		 JSAMPARRAY output_buf, int num_rows)
 /* Fast path for out_color_components==3, no dithering */
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
@@ -514,7 +514,7 @@ color_quantize3 (j_decompress_ptr cinfo, JSAMPARRAY input_buf,

 METHODDEF(void)
 quantize_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
-                     JSAMPARRAY output_buf, int num_rows)
+		     JSAMPARRAY output_buf, int num_rows)
 /* General case, with ordered dithering */
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
@@ -532,7 +532,7 @@ quantize_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
  for (row = 0; row < num_rows; row++) {
    /* Initialize output values to 0 so can process components separately */
    FMEMZERO((void FAR *) output_buf[row],
-             (size_t) (width * SIZEOF(JSAMPLE)));
+	     (size_t) (width * SIZEOF(JSAMPLE)));
    row_index = cquantize->row_index;
    for (ci = 0; ci < nc; ci++) {
      input_ptr = input_buf[row] + ci;
@@ -542,17 +542,17 @@ quantize_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
      col_index = 0;

      for (col = width; col > 0; col--) {
-        /* Form pixel value + dither, range-limit to 0..MAXJSAMPLE,
-         * select output value, accumulate into output code for this pixel.
-         * Range-limiting need not be done explicitly, as we have extended
-         * the colorindex table to produce the right answers for out-of-range
-         * inputs.  The maximum dither is +- MAXJSAMPLE; this sets the
-         * required amount of padding.
-         */
-        *output_ptr += colorindex_ci[GETJSAMPLE(*input_ptr)+dither[col_index]];
-        input_ptr += nc;
-        output_ptr++;
-        col_index = (col_index + 1) & ODITHER_MASK;
+	/* Form pixel value + dither, range-limit to 0..MAXJSAMPLE,
+	 * select output value, accumulate into output code for this pixel.
+	 * Range-limiting need not be done explicitly, as we have extended
+	 * the colorindex table to produce the right answers for out-of-range
+	 * inputs.  The maximum dither is +- MAXJSAMPLE; this sets the
+	 * required amount of padding.
+	 */
+	*output_ptr += colorindex_ci[GETJSAMPLE(*input_ptr)+dither[col_index]];
+	input_ptr += nc;
+	output_ptr++;
+	col_index = (col_index + 1) & ODITHER_MASK;
      }
    }
    /* Advance row index for next row */
@@ -564,7 +564,7 @@ quantize_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,

 METHODDEF(void)
 quantize3_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
-                      JSAMPARRAY output_buf, int num_rows)
+		      JSAMPARRAY output_buf, int num_rows)
 /* Fast path for out_color_components==3, with ordered dithering */
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
@@ -593,11 +593,11 @@ quantize3_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,

    for (col = width; col > 0; col--) {
      pixcode  = GETJSAMPLE(colorindex0[GETJSAMPLE(*input_ptr++) +
-                                        dither0[col_index]]);
+					dither0[col_index]]);
      pixcode += GETJSAMPLE(colorindex1[GETJSAMPLE(*input_ptr++) +
-                                        dither1[col_index]]);
+					dither1[col_index]]);
      pixcode += GETJSAMPLE(colorindex2[GETJSAMPLE(*input_ptr++) +
-                                        dither2[col_index]]);
+					dither2[col_index]]);
      *output_ptr++ = (JSAMPLE) pixcode;
      col_index = (col_index + 1) & ODITHER_MASK;
    }
@@ -609,7 +609,7 @@ quantize3_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,

 METHODDEF(void)
 quantize_fs_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
-                    JSAMPARRAY output_buf, int num_rows)
+		    JSAMPARRAY output_buf, int num_rows)
 /* General case, with Floyd-Steinberg dithering */
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
@@ -637,22 +637,22 @@ quantize_fs_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
  for (row = 0; row < num_rows; row++) {
    /* Initialize output values to 0 so can process components separately */
    FMEMZERO((void FAR *) output_buf[row],
-             (size_t) (width * SIZEOF(JSAMPLE)));
+	     (size_t) (width * SIZEOF(JSAMPLE)));
    for (ci = 0; ci < nc; ci++) {
      input_ptr = input_buf[row] + ci;
      output_ptr = output_buf[row];
      if (cquantize->on_odd_row) {
-        /* work right to left in this row */
-        input_ptr += (width-1) * nc; /* so point to rightmost pixel */
-        output_ptr += width-1;
-        dir = -1;
-        dirnc = -nc;
-        errorptr = cquantize->fserrors[ci] + (width+1); /* => entry after last column */
+	/* work right to left in this row */
+	input_ptr += (width-1) * nc; /* so point to rightmost pixel */
+	output_ptr += width-1;
+	dir = -1;
+	dirnc = -nc;
+	errorptr = cquantize->fserrors[ci] + (width+1); /* => entry after last column */
      } else {
-        /* work left to right in this row */
-        dir = 1;
-        dirnc = nc;
-        errorptr = cquantize->fserrors[ci]; /* => entry before first column */
+	/* work left to right in this row */
+	dir = 1;
+	dirnc = nc;
+	errorptr = cquantize->fserrors[ci]; /* => entry before first column */
      }
      colorindex_ci = cquantize->colorindex[ci];
      colormap_ci = cquantize->sv_colormap[ci];
@@ -662,47 +662,47 @@ quantize_fs_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
      belowerr = bpreverr = 0;

      for (col = width; col > 0; col--) {
-        /* cur holds the error propagated from the previous pixel on the
-         * current line.  Add the error propagated from the previous line
-         * to form the complete error correction term for this pixel, and
-         * round the error term (which is expressed * 16) to an integer.
-         * RIGHT_SHIFT rounds towards minus infinity, so adding 8 is correct
-         * for either sign of the error value.
-         * Note: errorptr points to *previous* column's array entry.
-         */
-        cur = RIGHT_SHIFT(cur + errorptr[dir] + 8, 4);
-        /* Form pixel value + error, and range-limit to 0..MAXJSAMPLE.
-         * The maximum error is +- MAXJSAMPLE; this sets the required size
-         * of the range_limit array.
-         */
-        cur += GETJSAMPLE(*input_ptr);
-        cur = GETJSAMPLE(range_limit[cur]);
-        /* Select output value, accumulate into output code for this pixel */
-        pixcode = GETJSAMPLE(colorindex_ci[cur]);
-        *output_ptr += (JSAMPLE) pixcode;
-        /* Compute actual representation error at this pixel */
-        /* Note: we can do this even though we don't have the final */
-        /* pixel code, because the colormap is orthogonal. */
-        cur -= GETJSAMPLE(colormap_ci[pixcode]);
-        /* Compute error fractions to be propagated to adjacent pixels.
-         * Add these into the running sums, and simultaneously shift the
-         * next-line error sums left by 1 column.
-         */
-        bnexterr = cur;
-        delta = cur * 2;
-        cur += delta;		/* form error * 3 */
-        errorptr[0] = (FSERROR) (bpreverr + cur);
-        cur += delta;		/* form error * 5 */
-        bpreverr = belowerr + cur;
-        belowerr = bnexterr;
-        cur += delta;		/* form error * 7 */
-        /* At this point cur contains the 7/16 error value to be propagated
-         * to the next pixel on the current line, and all the errors for the
-         * next line have been shifted over. We are therefore ready to move on.
-         */
-        input_ptr += dirnc;	/* advance input ptr to next column */
-        output_ptr += dir;	/* advance output ptr to next column */
-        errorptr += dir;	/* advance errorptr to current column */
+	/* cur holds the error propagated from the previous pixel on the
+	 * current line.  Add the error propagated from the previous line
+	 * to form the complete error correction term for this pixel, and
+	 * round the error term (which is expressed * 16) to an integer.
+	 * RIGHT_SHIFT rounds towards minus infinity, so adding 8 is correct
+	 * for either sign of the error value.
+	 * Note: errorptr points to *previous* column's array entry.
+	 */
+	cur = RIGHT_SHIFT(cur + errorptr[dir] + 8, 4);
+	/* Form pixel value + error, and range-limit to 0..MAXJSAMPLE.
+	 * The maximum error is +- MAXJSAMPLE; this sets the required size
+	 * of the range_limit array.
+	 */
+	cur += GETJSAMPLE(*input_ptr);
+	cur = GETJSAMPLE(range_limit[cur]);
+	/* Select output value, accumulate into output code for this pixel */
+	pixcode = GETJSAMPLE(colorindex_ci[cur]);
+	*output_ptr += (JSAMPLE) pixcode;
+	/* Compute actual representation error at this pixel */
+	/* Note: we can do this even though we don't have the final */
+	/* pixel code, because the colormap is orthogonal. */
+	cur -= GETJSAMPLE(colormap_ci[pixcode]);
+	/* Compute error fractions to be propagated to adjacent pixels.
+	 * Add these into the running sums, and simultaneously shift the
+	 * next-line error sums left by 1 column.
+	 */
+	bnexterr = cur;
+	delta = cur * 2;
+	cur += delta;		/* form error * 3 */
+	errorptr[0] = (FSERROR) (bpreverr + cur);
+	cur += delta;		/* form error * 5 */
+	bpreverr = belowerr + cur;
+	belowerr = bnexterr;
+	cur += delta;		/* form error * 7 */
+	/* At this point cur contains the 7/16 error value to be propagated
+	 * to the next pixel on the current line, and all the errors for the
+	 * next line have been shifted over. We are therefore ready to move on.
+	 */
+	input_ptr += dirnc;	/* advance input ptr to next column */
+	output_ptr += dir;	/* advance output ptr to next column */
+	errorptr += dir;	/* advance errorptr to current column */
      }
      /* Post-loop cleanup: we must unload the final error value into the
       * final fserrors[] entry.  Note we need not unload belowerr because
@@ -825,7 +825,7 @@ jinit_1pass_quantizer (j_decompress_ptr cinfo)

  cquantize = (my_cquantize_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_cquantizer));
+				SIZEOF(my_cquantizer));
  cinfo->cquantize = (struct jpeg_color_quantizer *) cquantize;
  cquantize->pub.start_pass = start_pass_1_quant;
  cquantize->pub.finish_pass = finish_pass_1_quant;
@@ -42,7 +42,7 @@
 * color space, and repeatedly splits the "largest" remaining box until we
 * have as many boxes as desired colors.  Then the mean color in each
 * remaining box becomes one of the possible output colors.
- *
+ * 
 * The second pass over the image maps each input pixel to the closest output
 * color (optionally after applying a Floyd-Steinberg dithering correction).
 * This mapping is logically trivial, but making it go fast enough requires
@@ -223,7 +223,7 @@ typedef my_cquantizer * my_cquantize_ptr;

 METHODDEF(void)
 prescan_quantize (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
-                  JSAMPARRAY output_buf, int num_rows)
+		  JSAMPARRAY output_buf, int num_rows)
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
  register JSAMPROW ptr;
@@ -238,11 +238,11 @@ prescan_quantize (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
    for (col = width; col > 0; col--) {
      /* get pixel value and index into the histogram */
      histp = & histogram[GETJSAMPLE(ptr[0]) >> C0_SHIFT]
-                         [GETJSAMPLE(ptr[1]) >> C1_SHIFT]
-                         [GETJSAMPLE(ptr[2]) >> C2_SHIFT];
+			 [GETJSAMPLE(ptr[1]) >> C1_SHIFT]
+			 [GETJSAMPLE(ptr[2]) >> C2_SHIFT];
      /* increment, check for overflow and undo increment if so. */
      if (++(*histp) <= 0)
-        (*histp)--;
+	(*histp)--;
      ptr += 3;
    }
  }
@@ -265,9 +265,9 @@ typedef struct {
  INT32 volume;
  /* The number of nonzero histogram cells within this box */
  long colorcount;
-} _box;
+} box;

-typedef _box * boxptr;
+typedef box * boxptr;


 LOCAL(boxptr)
@@ -279,7 +279,7 @@ find_biggest_color_pop (boxptr boxlist, int numboxes)
  register int i;
  register long maxc = 0;
  boxptr which = NULL;
-
+  
  for (i = 0, boxp = boxlist; i < numboxes; i++, boxp++) {
    if (boxp->colorcount > maxc && boxp->volume > 0) {
      which = boxp;
@@ -299,7 +299,7 @@ find_biggest_volume (boxptr boxlist, int numboxes)
  register int i;
  register INT32 maxv = 0;
  boxptr which = NULL;
-
+  
  for (i = 0, boxp = boxlist; i < numboxes; i++, boxp++) {
    if (boxp->volume > maxv) {
      which = boxp;
@@ -322,75 +322,75 @@ update_box (j_decompress_ptr cinfo, boxptr boxp)
  int c0min,c0max,c1min,c1max,c2min,c2max;
  INT32 dist0,dist1,dist2;
  long ccount;
-
+  
  c0min = boxp->c0min;  c0max = boxp->c0max;
  c1min = boxp->c1min;  c1max = boxp->c1max;
  c2min = boxp->c2min;  c2max = boxp->c2max;
-
+  
  if (c0max > c0min)
    for (c0 = c0min; c0 <= c0max; c0++)
      for (c1 = c1min; c1 <= c1max; c1++) {
-        histp = & histogram[c0][c1][c2min];
-        for (c2 = c2min; c2 <= c2max; c2++)
-          if (*histp++ != 0) {
-            boxp->c0min = c0min = c0;
-            goto have_c0min;
-          }
+	histp = & histogram[c0][c1][c2min];
+	for (c2 = c2min; c2 <= c2max; c2++)
+	  if (*histp++ != 0) {
+	    boxp->c0min = c0min = c0;
+	    goto have_c0min;
+	  }
      }
 have_c0min:
  if (c0max > c0min)
    for (c0 = c0max; c0 >= c0min; c0--)
      for (c1 = c1min; c1 <= c1max; c1++) {
-        histp = & histogram[c0][c1][c2min];
-        for (c2 = c2min; c2 <= c2max; c2++)
-          if (*histp++ != 0) {
-            boxp->c0max = c0max = c0;
-            goto have_c0max;
-          }
+	histp = & histogram[c0][c1][c2min];
+	for (c2 = c2min; c2 <= c2max; c2++)
+	  if (*histp++ != 0) {
+	    boxp->c0max = c0max = c0;
+	    goto have_c0max;
+	  }
      }
 have_c0max:
  if (c1max > c1min)
    for (c1 = c1min; c1 <= c1max; c1++)
      for (c0 = c0min; c0 <= c0max; c0++) {
-        histp = & histogram[c0][c1][c2min];
-        for (c2 = c2min; c2 <= c2max; c2++)
-          if (*histp++ != 0) {
-            boxp->c1min = c1min = c1;
-            goto have_c1min;
-          }
+	histp = & histogram[c0][c1][c2min];
+	for (c2 = c2min; c2 <= c2max; c2++)
+	  if (*histp++ != 0) {
+	    boxp->c1min = c1min = c1;
+	    goto have_c1min;
+	  }
      }
 have_c1min:
  if (c1max > c1min)
    for (c1 = c1max; c1 >= c1min; c1--)
      for (c0 = c0min; c0 <= c0max; c0++) {
-        histp = & histogram[c0][c1][c2min];
-        for (c2 = c2min; c2 <= c2max; c2++)
-          if (*histp++ != 0) {
-            boxp->c1max = c1max = c1;
-            goto have_c1max;
-          }
+	histp = & histogram[c0][c1][c2min];
+	for (c2 = c2min; c2 <= c2max; c2++)
+	  if (*histp++ != 0) {
+	    boxp->c1max = c1max = c1;
+	    goto have_c1max;
+	  }
      }
 have_c1max:
  if (c2max > c2min)
    for (c2 = c2min; c2 <= c2max; c2++)
      for (c0 = c0min; c0 <= c0max; c0++) {
-        histp = & histogram[c0][c1min][c2];
-        for (c1 = c1min; c1 <= c1max; c1++, histp += HIST_C2_ELEMS)
-          if (*histp != 0) {
-            boxp->c2min = c2min = c2;
-            goto have_c2min;
-          }
+	histp = & histogram[c0][c1min][c2];
+	for (c1 = c1min; c1 <= c1max; c1++, histp += HIST_C2_ELEMS)
+	  if (*histp != 0) {
+	    boxp->c2min = c2min = c2;
+	    goto have_c2min;
+	  }
      }
 have_c2min:
  if (c2max > c2min)
    for (c2 = c2max; c2 >= c2min; c2--)
      for (c0 = c0min; c0 <= c0max; c0++) {
-        histp = & histogram[c0][c1min][c2];
-        for (c1 = c1min; c1 <= c1max; c1++, histp += HIST_C2_ELEMS)
-          if (*histp != 0) {
-            boxp->c2max = c2max = c2;
-            goto have_c2max;
-          }
+	histp = & histogram[c0][c1min][c2];
+	for (c1 = c1min; c1 <= c1max; c1++, histp += HIST_C2_ELEMS)
+	  if (*histp != 0) {
+	    boxp->c2max = c2max = c2;
+	    goto have_c2max;
+	  }
      }
 have_c2max:

@@ -406,16 +406,16 @@ update_box (j_decompress_ptr cinfo, boxptr boxp)
  dist1 = ((c1max - c1min) << C1_SHIFT) * C1_SCALE;
  dist2 = ((c2max - c2min) << C2_SHIFT) * C2_SCALE;
  boxp->volume = dist0*dist0 + dist1*dist1 + dist2*dist2;
-
+  
  /* Now scan remaining volume of box and compute population */
  ccount = 0;
  for (c0 = c0min; c0 <= c0max; c0++)
    for (c1 = c1min; c1 <= c1max; c1++) {
      histp = & histogram[c0][c1][c2min];
      for (c2 = c2min; c2 <= c2max; c2++, histp++)
-        if (*histp != 0) {
-          ccount++;
-        }
+	if (*histp != 0) {
+	  ccount++;
+	}
    }
  boxp->colorcount = ccount;
 }
@@ -423,7 +423,7 @@ update_box (j_decompress_ptr cinfo, boxptr boxp)

 LOCAL(int)
 median_cut (j_decompress_ptr cinfo, boxptr boxlist, int numboxes,
-            int desired_colors)
+	    int desired_colors)
 /* Repeatedly select and split the largest box until we have enough boxes */
 {
  int n,lb;
@@ -512,24 +512,24 @@ compute_color (j_decompress_ptr cinfo, boxptr boxp, int icolor)
  long c0total = 0;
  long c1total = 0;
  long c2total = 0;
-
+  
  c0min = boxp->c0min;  c0max = boxp->c0max;
  c1min = boxp->c1min;  c1max = boxp->c1max;
  c2min = boxp->c2min;  c2max = boxp->c2max;
-
+  
  for (c0 = c0min; c0 <= c0max; c0++)
    for (c1 = c1min; c1 <= c1max; c1++) {
      histp = & histogram[c0][c1][c2min];
      for (c2 = c2min; c2 <= c2max; c2++) {
-        if ((count = *histp++) != 0) {
-          total += count;
-          c0total += ((c0 << C0_SHIFT) + ((1<<C0_SHIFT)>>1)) * count;
-          c1total += ((c1 << C1_SHIFT) + ((1<<C1_SHIFT)>>1)) * count;
-          c2total += ((c2 << C2_SHIFT) + ((1<<C2_SHIFT)>>1)) * count;
-        }
+	if ((count = *histp++) != 0) {
+	  total += count;
+	  c0total += ((c0 << C0_SHIFT) + ((1<<C0_SHIFT)>>1)) * count;
+	  c1total += ((c1 << C1_SHIFT) + ((1<<C1_SHIFT)>>1)) * count;
+	  c2total += ((c2 << C2_SHIFT) + ((1<<C2_SHIFT)>>1)) * count;
+	}
      }
    }
-
+  
  cinfo->colormap[0][icolor] = (JSAMPLE) ((c0total + (total>>1)) / total);
  cinfo->colormap[1][icolor] = (JSAMPLE) ((c1total + (total>>1)) / total);
  cinfo->colormap[2][icolor] = (JSAMPLE) ((c2total + (total>>1)) / total);
@@ -546,7 +546,7 @@ select_colors (j_decompress_ptr cinfo, int desired_colors)

  /* Allocate workspace for box list */
  boxlist = (boxptr) (*cinfo->mem->alloc_small)
-    ((j_common_ptr) cinfo, JPOOL_IMAGE, desired_colors * SIZEOF(_box));
+    ((j_common_ptr) cinfo, JPOOL_IMAGE, desired_colors * SIZEOF(box));
  /* Initialize one box containing whole space */
  numboxes = 1;
  boxlist[0].c0min = 0;
@@ -645,7 +645,7 @@ select_colors (j_decompress_ptr cinfo, int desired_colors)

 LOCAL(int)
 find_nearby_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,
-                    JSAMPLE colorlist[])
+		    JSAMPLE colorlist[])
 /* Locate the colormap entries close enough to an update box to be candidates
 * for the nearest entry to some cell(s) in the update box.  The update box
 * is specified by the center coordinates of its first cell.  The number of
@@ -702,11 +702,11 @@ find_nearby_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,
      /* within cell range so no contribution to min_dist */
      min_dist = 0;
      if (x <= centerc0) {
-        tdist = (x - maxc0) * C0_SCALE;
-        max_dist = tdist*tdist;
+	tdist = (x - maxc0) * C0_SCALE;
+	max_dist = tdist*tdist;
      } else {
-        tdist = (x - minc0) * C0_SCALE;
-        max_dist = tdist*tdist;
+	tdist = (x - minc0) * C0_SCALE;
+	max_dist = tdist*tdist;
      }
    }

@@ -724,11 +724,11 @@ find_nearby_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,
    } else {
      /* within cell range so no contribution to min_dist */
      if (x <= centerc1) {
-        tdist = (x - maxc1) * C1_SCALE;
-        max_dist += tdist*tdist;
+	tdist = (x - maxc1) * C1_SCALE;
+	max_dist += tdist*tdist;
      } else {
-        tdist = (x - minc1) * C1_SCALE;
-        max_dist += tdist*tdist;
+	tdist = (x - minc1) * C1_SCALE;
+	max_dist += tdist*tdist;
      }
    }

@@ -746,11 +746,11 @@ find_nearby_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,
    } else {
      /* within cell range so no contribution to min_dist */
      if (x <= centerc2) {
-        tdist = (x - maxc2) * C2_SCALE;
-        max_dist += tdist*tdist;
+	tdist = (x - maxc2) * C2_SCALE;
+	max_dist += tdist*tdist;
      } else {
-        tdist = (x - minc2) * C2_SCALE;
-        max_dist += tdist*tdist;
+	tdist = (x - minc2) * C2_SCALE;
+	max_dist += tdist*tdist;
      }
    }

@@ -774,7 +774,7 @@ find_nearby_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,

 LOCAL(void)
 find_best_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,
-                  int numcolors, JSAMPLE colorlist[], JSAMPLE bestcolor[])
+		  int numcolors, JSAMPLE colorlist[], JSAMPLE bestcolor[])
 /* Find the closest colormap entry for each cell in the update box,
 * given the list of candidate colors prepared by find_nearby_colors.
 * Return the indexes of the closest entries in the bestcolor[] array.
@@ -798,17 +798,17 @@ find_best_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,
  bptr = bestdist;
  for (i = BOX_C0_ELEMS*BOX_C1_ELEMS*BOX_C2_ELEMS-1; i >= 0; i--)
    *bptr++ = 0x7FFFFFFFL;
-
+  
  /* For each color selected by find_nearby_colors,
   * compute its distance to the center of each cell in the box.
   * If that's less than best-so-far, update best distance and color number.
   */
-
+  
  /* Nominal steps between cell centers ("x" in Thomas article) */
 #define STEP_C0  ((1 << C0_SHIFT) * C0_SCALE)
 #define STEP_C1  ((1 << C1_SHIFT) * C1_SCALE)
 #define STEP_C2  ((1 << C2_SHIFT) * C2_SCALE)
-
+  
  for (i = 0; i < numcolors; i++) {
    icolor = GETJSAMPLE(colorlist[i]);
    /* Compute (square of) distance from minc0/c1/c2 to this color */
@@ -830,20 +830,20 @@ find_best_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2,
      dist1 = dist0;
      xx1 = inc1;
      for (ic1 = BOX_C1_ELEMS-1; ic1 >= 0; ic1--) {
-        dist2 = dist1;
-        xx2 = inc2;
-        for (ic2 = BOX_C2_ELEMS-1; ic2 >= 0; ic2--) {
-          if (dist2 < *bptr) {
-            *bptr = dist2;
-            *cptr = (JSAMPLE) icolor;
-          }
-          dist2 += xx2;
-          xx2 += 2 * STEP_C2 * STEP_C2;
-          bptr++;
-          cptr++;
-        }
-        dist1 += xx1;
-        xx1 += 2 * STEP_C1 * STEP_C1;
+	dist2 = dist1;
+	xx2 = inc2;
+	for (ic2 = BOX_C2_ELEMS-1; ic2 >= 0; ic2--) {
+	  if (dist2 < *bptr) {
+	    *bptr = dist2;
+	    *cptr = (JSAMPLE) icolor;
+	  }
+	  dist2 += xx2;
+	  xx2 += 2 * STEP_C2 * STEP_C2;
+	  bptr++;
+	  cptr++;
+	}
+	dist1 += xx1;
+	xx1 += 2 * STEP_C1 * STEP_C1;
      }
      dist0 += xx0;
      xx0 += 2 * STEP_C0 * STEP_C0;
@@ -882,7 +882,7 @@ fill_inverse_cmap (j_decompress_ptr cinfo, int c0, int c1, int c2)
  minc0 = (c0 << BOX_C0_SHIFT) + ((1 << C0_SHIFT) >> 1);
  minc1 = (c1 << BOX_C1_SHIFT) + ((1 << C1_SHIFT) >> 1);
  minc2 = (c2 << BOX_C2_SHIFT) + ((1 << C2_SHIFT) >> 1);
-
+  
  /* Determine which colormap entries are close enough to be candidates
   * for the nearest entry to some cell in the update box.
   */
@@ -890,7 +890,7 @@ fill_inverse_cmap (j_decompress_ptr cinfo, int c0, int c1, int c2)

  /* Determine the actually nearest colors. */
  find_best_colors(cinfo, minc0, minc1, minc2, numcolors, colorlist,
-                   bestcolor);
+		   bestcolor);

  /* Save the best color numbers (plus 1) in the main cache array */
  c0 <<= BOX_C0_LOG;		/* convert ID back to base cell indexes */
@@ -901,7 +901,7 @@ fill_inverse_cmap (j_decompress_ptr cinfo, int c0, int c1, int c2)
    for (ic1 = 0; ic1 < BOX_C1_ELEMS; ic1++) {
      cachep = & histogram[c0+ic0][c1+ic1][c2];
      for (ic2 = 0; ic2 < BOX_C2_ELEMS; ic2++) {
-        *cachep++ = (histcell) (GETJSAMPLE(*cptr++) + 1);
+	*cachep++ = (histcell) (GETJSAMPLE(*cptr++) + 1);
      }
    }
  }
@@ -914,7 +914,7 @@ fill_inverse_cmap (j_decompress_ptr cinfo, int c0, int c1, int c2)

 METHODDEF(void)
 pass2_no_dither (j_decompress_ptr cinfo,
-                 JSAMPARRAY input_buf, JSAMPARRAY output_buf, int num_rows)
+		 JSAMPARRAY input_buf, JSAMPARRAY output_buf, int num_rows)
 /* This version performs no dithering */
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
@@ -938,7 +938,7 @@ pass2_no_dither (j_decompress_ptr cinfo,
      /* If we have not seen this color before, find nearest colormap entry */
      /* and update the cache */
      if (*cachep == 0)
-        fill_inverse_cmap(cinfo, c0,c1,c2);
+	fill_inverse_cmap(cinfo, c0,c1,c2);
      /* Now emit the colormap index for this cell */
      *outptr++ = (JSAMPLE) (*cachep - 1);
    }
@@ -948,7 +948,7 @@ pass2_no_dither (j_decompress_ptr cinfo,

 METHODDEF(void)
 pass2_fs_dither (j_decompress_ptr cinfo,
-                 JSAMPARRAY input_buf, JSAMPARRAY output_buf, int num_rows)
+		 JSAMPARRAY input_buf, JSAMPARRAY output_buf, int num_rows)
 /* This version performs Floyd-Steinberg dithering */
 {
  my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize;
@@ -1029,14 +1029,14 @@ pass2_fs_dither (j_decompress_ptr cinfo,
      /* If we have not seen this color before, find nearest colormap */
      /* entry and update the cache */
      if (*cachep == 0)
-        fill_inverse_cmap(cinfo, cur0>>C0_SHIFT,cur1>>C1_SHIFT,cur2>>C2_SHIFT);
+	fill_inverse_cmap(cinfo, cur0>>C0_SHIFT,cur1>>C1_SHIFT,cur2>>C2_SHIFT);
      /* Now emit the colormap index for this cell */
      { register int pixcode = *cachep - 1;
-        *outptr = (JSAMPLE) pixcode;
-        /* Compute representation error for this pixel */
-        cur0 -= GETJSAMPLE(colormap0[pixcode]);
-        cur1 -= GETJSAMPLE(colormap1[pixcode]);
-        cur2 -= GETJSAMPLE(colormap2[pixcode]);
+	*outptr = (JSAMPLE) pixcode;
+	/* Compute representation error for this pixel */
+	cur0 -= GETJSAMPLE(colormap0[pixcode]);
+	cur1 -= GETJSAMPLE(colormap1[pixcode]);
+	cur2 -= GETJSAMPLE(colormap2[pixcode]);
      }
      /* Compute error fractions to be propagated to adjacent pixels.
       * Add these into the running sums, and simultaneously shift the
@@ -1044,30 +1044,30 @@ pass2_fs_dither (j_decompress_ptr cinfo,
       */
      { register LOCFSERROR bnexterr, delta;

-        bnexterr = cur0;	/* Process component 0 */
-        delta = cur0 * 2;
-        cur0 += delta;		/* form error * 3 */
-        errorptr[0] = (FSERROR) (bpreverr0 + cur0);
-        cur0 += delta;		/* form error * 5 */
-        bpreverr0 = belowerr0 + cur0;
-        belowerr0 = bnexterr;
-        cur0 += delta;		/* form error * 7 */
-        bnexterr = cur1;	/* Process component 1 */
-        delta = cur1 * 2;
-        cur1 += delta;		/* form error * 3 */
-        errorptr[1] = (FSERROR) (bpreverr1 + cur1);
-        cur1 += delta;		/* form error * 5 */
-        bpreverr1 = belowerr1 + cur1;
-        belowerr1 = bnexterr;
-        cur1 += delta;		/* form error * 7 */
-        bnexterr = cur2;	/* Process component 2 */
-        delta = cur2 * 2;
-        cur2 += delta;		/* form error * 3 */
-        errorptr[2] = (FSERROR) (bpreverr2 + cur2);
-        cur2 += delta;		/* form error * 5 */
-        bpreverr2 = belowerr2 + cur2;
-        belowerr2 = bnexterr;
-        cur2 += delta;		/* form error * 7 */
+	bnexterr = cur0;	/* Process component 0 */
+	delta = cur0 * 2;
+	cur0 += delta;		/* form error * 3 */
+	errorptr[0] = (FSERROR) (bpreverr0 + cur0);
+	cur0 += delta;		/* form error * 5 */
+	bpreverr0 = belowerr0 + cur0;
+	belowerr0 = bnexterr;
+	cur0 += delta;		/* form error * 7 */
+	bnexterr = cur1;	/* Process component 1 */
+	delta = cur1 * 2;
+	cur1 += delta;		/* form error * 3 */
+	errorptr[1] = (FSERROR) (bpreverr1 + cur1);
+	cur1 += delta;		/* form error * 5 */
+	bpreverr1 = belowerr1 + cur1;
+	belowerr1 = bnexterr;
+	cur1 += delta;		/* form error * 7 */
+	bnexterr = cur2;	/* Process component 2 */
+	delta = cur2 * 2;
+	cur2 += delta;		/* form error * 3 */
+	errorptr[2] = (FSERROR) (bpreverr2 + cur2);
+	cur2 += delta;		/* form error * 5 */
+	bpreverr2 = belowerr2 + cur2;
+	belowerr2 = bnexterr;
+	cur2 += delta;		/* form error * 7 */
      }
      /* At this point curN contains the 7/16 error value to be propagated
       * to the next pixel on the current line, and all the errors for the
@@ -1198,16 +1198,16 @@ start_pass_2_quant (j_decompress_ptr cinfo, boolean is_pre_scan)

    if (cinfo->dither_mode == JDITHER_FS) {
      size_t arraysize = (size_t) ((cinfo->output_width + 2) *
-                                   (3 * SIZEOF(FSERROR)));
+				   (3 * SIZEOF(FSERROR)));
      /* Allocate Floyd-Steinberg workspace if we didn't already. */
      if (cquantize->fserrors == NULL)
-        cquantize->fserrors = (FSERRPTR) (*cinfo->mem->alloc_large)
-          ((j_common_ptr) cinfo, JPOOL_IMAGE, arraysize);
+	cquantize->fserrors = (FSERRPTR) (*cinfo->mem->alloc_large)
+	  ((j_common_ptr) cinfo, JPOOL_IMAGE, arraysize);
      /* Initialize the propagated errors to zero. */
      FMEMZERO((void FAR *) cquantize->fserrors, arraysize);
      /* Make the error-limit table if we didn't already. */
      if (cquantize->error_limiter == NULL)
-        init_error_limit(cinfo);
+	init_error_limit(cinfo);
      cquantize->on_odd_row = FALSE;
    }

@@ -1216,7 +1216,7 @@ start_pass_2_quant (j_decompress_ptr cinfo, boolean is_pre_scan)
  if (cquantize->needs_zeroed) {
    for (i = 0; i < HIST_C0_ELEMS; i++) {
      FMEMZERO((void FAR *) histogram[i],
-               HIST_C1_ELEMS*HIST_C2_ELEMS * SIZEOF(histcell));
+	       HIST_C1_ELEMS*HIST_C2_ELEMS * SIZEOF(histcell));
    }
    cquantize->needs_zeroed = FALSE;
  }
@@ -1249,7 +1249,7 @@ jinit_2pass_quantizer (j_decompress_ptr cinfo)

  cquantize = (my_cquantize_ptr)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
-                                SIZEOF(my_cquantizer));
+				SIZEOF(my_cquantizer));
  cinfo->cquantize = (struct jpeg_color_quantizer *) cquantize;
  cquantize->pub.start_pass = start_pass_2_quant;
  cquantize->pub.new_color_map = new_color_map_2_quant;
@@ -175,8 +175,8 @@ jzero_far (void FAR * target, size_t bytestozero)

 GLOBAL(void)
 jcopy_sample_rows (JSAMPARRAY input_array, int source_row,
-                   JSAMPARRAY output_array, int dest_row,
-                   int num_rows, JDIMENSION num_cols)
+		   JSAMPARRAY output_array, int dest_row,
+		   int num_rows, JDIMENSION num_cols)
 /* Copy some rows of samples from one place to another.
 * num_rows rows are copied from input_array[source_row++]
 * to output_array[dest_row++]; these areas may overlap for duplication.
@@ -209,7 +209,7 @@ jcopy_sample_rows (JSAMPARRAY input_array, int source_row,

 GLOBAL(void)
 jcopy_block_row (JBLOCKROW input_row, JBLOCKROW output_row,
-                 JDIMENSION num_blocks)
+		 JDIMENSION num_blocks)
 /* Copy a row of coefficient blocks from one place to another. */
 {
 #ifdef FMEMCOPY
@@ -1,7 +1,7 @@
 /*
 * jversion.h
 *
- * Copyright (C) 1991-2013, Thomas G. Lane, Guido Vollbeding.
+ * Copyright (C) 1991-2016, Thomas G. Lane, Guido Vollbeding.
 * This file is part of the Independent JPEG Group's software.
 * For conditions of distribution and use, see the accompanying README file.
 *
@@ -9,6 +9,6 @@
 */


-#define JVERSION	"9  13-Jan-2013"
+#define JVERSION	"9b  17-Jan-2016"

-#define JCOPYRIGHT	"Copyright (C) 2013, Thomas G. Lane, Guido Vollbeding"
+#define JCOPYRIGHT	"Copyright (C) 2016, Thomas G. Lane, Guido Vollbeding"
@@ -593,7 +593,7 @@ Version 1.0.5e [November 30, 1999]
    with trailing compressed parts easier in the future, and added new functions
    png_free_iCCP, png_free_pCAL, png_free_sPLT, png_free_text, png_get_iCCP,
    png_get_spalettes, png_set_iCCP, png_set_spalettes (Eric S. Raymond).
-  NOTE: Applications that write text chunks MUST define png_text->lang
+    NOTE: Applications that write text chunks MUST define png_text->lang
    before calling png_set_text(). It must be set to NULL if you want to
    write tEXt or zTXt chunks.  If you want your application to be able to
    run with older versions of libpng, use
@@ -833,7 +833,7 @@ Version 1.0.7beta11 [May 7, 2000]
  Removed the new PNG_CREATED_READ_STRUCT and PNG_CREATED_WRITE_STRUCT modes
    which are no longer used.
  Eliminated the three new members of png_text when PNG_LEGACY_SUPPORTED is
-    defined or when neither PNG_READ_iTXt_SUPPORTED nor PNG_WRITE_iTXT_SUPPORTED
+    defined or when neither PNG_READ_iTXt_SUPPORTED nor PNG_WRITE_iTXt_SUPPORTED
    is defined.
  Made PNG_NO_READ|WRITE_iTXt the default setting, to avoid memory
    overrun when old applications fill the info_ptr->text structure directly.
@@ -5675,6 +5675,372 @@ Version 1.6.24rc03 [August 2, 2016]
 Version 1.6.24[August 4, 2016]
  No changes.

+Version 1.6.25beta01 [August 12, 2016]
+  Reject oversized iCCP profile immediately.
+  Cleaned up PNG_DEBUG compile of pngtest.c.
+  Conditionally compile png_inflate().
+
+Version 1.6.25beta02 [August 18, 2016]
+  Don't install pngcp; it conflicts with pngcp in the pngtools package.
+  Minor editing of INSTALL, (whitespace, added copyright line)
+
+Version 1.6.25rc01 [August 24, 2016]
+  No changes.
+
+Version 1.6.25rc02 [August 29, 2016]
+  Added MIPS support (Mandar Sahastrabuddhe <Mandar.Sahastrabuddhe@imgtec.com>).
+  Only the UP filter is currently implemented.
+
+Version 1.6.25rc03 [August 29, 2016]
+  Rebased contrib/intel/intel_sse.patch after the MIPS implementation.
+
+Version 1.6.25rc04 [August 30, 2016]
+  Added MIPS support for SUB, AVG, and PAETH filters (Mandar Sahastrabuddhe).
+
+Version 1.6.25rc05 [August 30, 2016]
+  Rebased contrib/intel/intel_sse.patch after the MIPS implementation update..
+
+Version 1.6.25 [September 1, 2016]
+  No changes.
+
+Version 1.6.26beta01 [September 26, 2016]
+  Fixed handling zero length IDAT in pngfix (bug report by Agostino Sarubbo,
+    bugfix by John Bowler).
+  Do not issue a png_error() on read in png_set_pCAL() because png_handle_pCAL
+    has allocated memory that libpng needs to free.
+  Conditionally compile png_set_benign_errors() in pngread.c and pngtest.c
+  Issue a png_benign_error instead of a png_error on ADLER32 mismatch
+    while decoding compressed data chunks.
+  Changed PNG_ZLIB_VERNUM to ZLIB_VERNUM in pngpriv.h, pngstruct.h, and
+    pngrutil.c.
+  If CRC handling of critical chunks has been set to PNG_CRC_QUIET_USE,
+    ignore the ADLER32 checksum in the IDAT chunk as well as the chunk CRCs.
+  Issue png_benign_error() on ADLER32 checksum mismatch instead of png_error().
+  Add tests/badcrc.png and tests/badadler.png to tests/pngtest.
+  Merged pngtest.c with libpng-1.7.0beta84/pngtest.c
+
+Version 1.6.26beta02 [October 1, 2016]
+  Updated the documentation about CRC and ADLER32 handling.
+  Quieted 117 warnings from clang-3.8 in pngtrans.c, pngread.c,
+     pngwrite.c, pngunknown.c, and pngvalid.c.
+  Quieted 58 (out of 144) -Wconversion compiler warnings by changing
+    flag definitions in pngpriv.h from 0xnnnn to 0xnnnnU and trivial changes
+    in png.c, pngread.c, and pngwutil.c.
+
+Version 1.6.26beta03 [October 2, 2016]
+  Removed contrib/libtests/*.orig and *.rej that slipped into the tarballs.
+  Quieted the 86 remaining -Wconversion compiler warnings by
+    revising the png_isaligned() macro and trivial changes in png.c,
+    pngerror.c, pngget.c, pngmem.c, pngset.c, pngrtran.c, pngrutil.c,
+    pngwtran.c, pngwrite.c, and pngwutil.c.
+
+Version 1.6.26beta04 [October 3, 2016]
+  Quieted (bogus?) clang warnings about "absolute value has no effect"
+    when PNG_USE_ABS is defined.
+  Fixed offsets in contrib/intel/intel_sse.patch
+
+Version 1.6.26beta05 [October 6, 2016]
+  Changed integer constant 4294967294 to unsigned 4294967294U in pngconf.h
+    to avoid a signed/unsigned compare in the preprocessor.
+
+Version 1.6.26beta06 [October 7, 2016]
+  Use zlib-1.2.8.1 inflateValidate() instead of inflateReset2() to
+    optionally avoid ADLER32 evaluation.
+
+Version 1.6.26rc01 [October 12, 2016]
+  No changes.
+
+Version 1.6.26 [October 20, 2016]
+  Cosmetic change, "ptr != 0" to "ptr != NULL" in png.c and pngrutil.c
+  Despammed email addresses (replaced "@" with " at ").
+
+Version 1.6.27beta01 [November 2, 2016]
+  Restrict the new ADLER32-skipping to IDAT chunks.  It broke iCCP chunk
+    handling: an erroneous iCCP chunk would throw a png_error and reject the
+    entire PNG image instead of rejecting just the iCCP chunk with a warning,
+    if built with zlib-1.2.8.1.
+
+Version 1.6.27rc01 [December 27, 2016]
+  Control ADLER32 checking with new PNG_IGNORE_ADLER32 option. Fixes
+    an endless loop when handling erroneous ADLER32 checksums; bug
+    introduced in libpng-1.6.26.
+  Removed the use of a macro containing the pre-processor 'defined'
+    operator.  It is unclear whether this is valid; a macro that
+    "generates" 'defined' is not permitted, but the use of the word
+    "generates" within the C90 standard seems to imply more than simple
+    substitution of an expression itself containing a well-formed defined
+    operation.
+  Added ARM support to CMakeLists.txt (Andreas Franek).
+
+Version 1.6.27 [December 29, 2016]
+  Fixed a potential null pointer dereference in png_set_text_2() (bug report
+    and patch by Patrick Keshishian, CVE-2016-10087).
+
+Version 1.6.28rc01 [January 3, 2017]
+  Fixed arm/aarch64 detection in CMakeLists.txt (Gianfranco Costamagna).
+  Added option to Cmake build allowing a custom location of zlib to be
+    specified in a scenario where libpng is being built as a subproject
+    alongside zlib by another project (Sam Serrels).
+  Changed png_ptr->options from a png_byte to png_uint_32, to accomodate
+    up to 16 options.
+
+Version 1.6.28rc02 [January 4, 2017]
+  Added "include(GNUInstallDirs)" to CMakeLists.txt (Gianfranco Costamagna).
+  Moved SSE2 optimization code into the main libpng source directory.
+    Configure libpng with "configure --enable-intel-sse" or compile
+    libpng with "-DPNG_INTEL_SSE" in CPPFLAGS to enable it.
+
+Version 1.6.28rc03 [January 4, 2017]
+  Backed out the SSE optimization and last CMakeLists.txt to allow time for QA.
+
+Version 1.6.28 [January 5, 2017]
+  No changes.
+
+Version 1.6.29beta01 [January 12, 2017]
+  Readded "include(GNUInstallDirs)" to CMakeLists.txt (Gianfranco Costamagna).
+  Moved SSE2 optimization code into the main libpng source directory.
+    Configure libpng with "configure --enable-intel-sse" or compile
+    libpng with "-DPNG_INTEL_SSE" in CPPFLAGS to enable it.
+  Simplified conditional compilation in pngvalid.c, for AIX (Michael Felt).
+
+Version 1.6.29beta02 [February 22, 2017]
+  Avoid conditional directives that break statements in pngrutil.c (Romero
+    Malaquias)
+  The contrib/examples/pngtopng.c recovery code was in the wrong "if"
+    branches; the comments were correct.
+  Added code for PowerPC VSX optimisation (Vadim Barkov).
+
+Version 1.6.29beta03 [March 1, 2017]
+  Avoid potential overflow of shift operations in png_do_expand() (Aaron Boxer).
+  Change test ZLIB_VERNUM >= 0x1281 to ZLIB_VERNUM >= 0x1290 in pngrutil.c
+    because Solaris 11 distributes zlib-1.2.8.f that is older than 1.2.8.1,
+    as suggested in zlib FAQ, item 24.
+  Suppress clang warnings about implicit sign changes in png.c
+
+Version 1.6.29 [March 16, 2017]
+  No changes.
+
+Version 1.6.30beta01 [April 1, 2017]
+  Added missing "$(CPPFLAGS)" to the compile line for c.pic.o in
+    makefile.linux and makefile.solaris-x86 (Cosmin).
+  Revised documentation of png_get_error_ptr() in the libpng manual.
+  Silence clang -Wcomma and const drop warnings (Viktor Szakats).
+  Update Sourceforge URLs in documentation (https instead of http).
+
+Version 1.6.30beta02 [April 22, 2017]
+  Document need to check for integer overflow when allocating a pixel
+    buffer for multiple rows in contrib/gregbook, contrib/pngminus,
+    example.c, and in the manual (suggested by Jaeseung Choi). This
+    is similar to the bug reported against pngquant in CVE-2016-5735.
+  Removed reference to the obsolete PNG_SAFE_LIMITS macro in the documentation.
+
+Version 1.6.30beta03 [May 22, 2017]
+  Check for integer overflow in contrib/visupng and contrib/tools/genpng.
+  Do not double evaluate CMAKE_SYSTEM_PROCESSOR in CMakeLists.txt.
+  Test CMAKE_HOST_WIN32 instead of WIN32 in CMakeLists.txt.
+  Fix some URL in documentation.
+
+Version 1.6.30beta04 [June 7, 2017]
+  Avoid writing an empty IDAT when the last IDAT exactly fills the
+    compression buffer (bug report by Brian Baird). This bug was
+    introduced in libpng-1.6.0.
+
+Version 1.6.30rc01 [June 14, 2017]
+  No changes.
+
+Version 1.6.30rc02 [June 25, 2017]
+  Update copyright year in pnglibconf.h, make ltmain.sh executable.
+  Add a reference to the libpng.download site in README.
+
+Version 1.6.30 [June 28, 2017]
+  No changes.
+
+Version 1.6.31beta01 [July 5, 2017]
+  Guard the definition of _POSIX_SOURCE in pngpriv.h (AIX already defines it;
+    bug report by Michael Felt).
+  Revised pngpriv.h to work around failure to compile arm/filter_neon.S
+    ("typedef" directive is unrecognized by the assembler). The problem
+    was introduced in libpng-1.6.30beta01.
+  Added "Requires: zlib" to libpng.pc.in (Pieter Neerincx).
+  Added special case for FreeBSD in arm/filter_neon.S (Maya Rashish).
+
+Version 1.6.31beta02 [July 8, 2017]
+  Added instructions for disabling hardware optimizations in INSTALL.
+  Added "--enable-hardware-optimizations" configuration flag to enable
+    or disable all hardware optimizations with one flag.
+
+Version 1.6.31beta03 [July 9, 2017]
+  Updated CMakeLists.txt to add INTEL_SSE and MIPS_MSA platforms.
+  Changed "int" to "png_size_t" in intel/filter_sse2.c to prevent
+    possible integer overflow (Bug report by John Bowler).
+  Quieted "declaration after statement" warnings in intel/filter_sse2.c.
+  Added scripts/makefile-linux-opt, which has hardware optimizations enabled.
+
+Version 1.6.31beta04 [July 11, 2017]
+  Removed one of the GCC-7.1.0 'strict-overflow' warnings that result when
+    integers appear on both sides of a compare.  Worked around the others by
+    forcing the strict-overflow setting in the relevant functions to a level
+    where they are not reported (John Bowler).
+  Changed "FALL THROUGH" comments to "FALLTHROUGH" because GCC doesn't like
+    the space.
+  Worked around some C-style casts from (void*) because g++ 5.4.0 objects
+    to them.
+  Increased the buffer size for 'sprint' to pass the gcc 7.1.0 'sprint
+    overflow' check that is on by default with -Wall -Wextra.
+
+Version 1.6.31beta05 [July 13, 2017]
+  Added eXIf chunk support.
+
+Version 1.6.31beta06 [July 17, 2017]
+  Added a minimal eXIf chunk (with Orientation and FocalLengthIn35mmFilm
+    tags) to pngtest.png.
+
+Version 1.6.31beta07 [July 18, 2017]
+  Revised the eXIf chunk in pngtest.png to fix "Bad IFD1 Directory" warning.
+
+Version 1.6.31rc01 [July 19, 2017]
+  No changes.
+
+Version 1.6.31rc02 [July 25, 2017]
+  Fixed typo in example.c (png_free_image should be png_image_free) (Bug
+    report by John Smith)
+
+Version 1.6.31 [July 27, 2017]
+  No changes.
+
+Version 1.6.32beta01 [July 31, 2017]
+  Avoid possible NULL dereference in png_handle_eXIf when benign_errors
+    are allowed. Avoid leaking the input buffer "eXIf_buf".
+  Eliminated png_ptr->num_exif member from pngstruct.h and added num_exif
+    to arguments for png_get_eXIf() and png_set_eXIf().
+  Added calls to png_handle_eXIf(() in pngread.c and png_write_eXIf() in
+    pngwrite.c, and made various other fixes to png_write_eXIf().
+  Changed name of png_get_eXIF and png_set_eXIf() to png_get_eXIf_1() and
+    png_set_eXIf_1(), respectively, to avoid breaking API compatibility
+    with libpng-1.6.31.
+
+Version 1.6.32beta02 [August 1, 2017]
+  Updated contrib/libtests/pngunknown.c with eXIf chunk.
+
+Version 1.6.32beta03 [August 2, 2017]
+  Initialized btoa[] in pngstest.c
+  Stop memory leak when returning from png_handle_eXIf() with an error
+    (Bug report from the OSS-fuzz project).
+
+Version 1.6.32beta04 [August 2, 2017]
+  Replaced local eXIf_buf with info_ptr-eXIf_buf in png_handle_eXIf().
+  Update libpng.3 and libpng-manual.txt about eXIf functions.
+
+Version 1.6.32beta05 [August 2, 2017]
+  Restored png_get_eXIf() and png_set_eXIf() to maintain API compatability.
+
+Version 1.6.32beta06 [August 2, 2017]
+  Removed png_get_eXIf_1() and png_set_eXIf_1().
+
+Version 1.6.32beta07 [August 3, 2017]
+  Check length of all chunks except IDAT against user limit to fix an
+    OSS-fuzz issue (Fixes CVE-2017-12652).
+
+Version 1.6.32beta08 [August 3, 2017]
+  Check length of IDAT against maximum possible IDAT size, accounting
+    for height, rowbytes, interlacing and zlib/deflate overhead.
+  Restored png_get_eXIf_1() and png_set_eXIf_1(), because strlen(eXIf_buf)
+    does not work (the eXIf chunk data can contain zeroes).
+
+Version 1.6.32beta09 [August 3, 2017]
+  Require cmake-2.8.8 in CMakeLists.txt. Revised symlink creation,
+    no longer using deprecated cmake LOCATION feature (Clifford Yapp).
+  Fixed five-byte error in the calculation of IDAT maximum possible size.
+  
+Version 1.6.32beta10 [August 5, 2017]
+  Moved chunk-length check into a png_check_chunk_length() private
+    function (Suggested by Max Stepin).
+  Moved bad pngs from tests to contrib/libtests/crashers
+  Moved testing of bad pngs into a separate tests/pngtest-badpngs script
+  Added the --xfail (expected FAIL) option to pngtest.c. It writes XFAIL
+    in the output but PASS for the libpng test.
+  Require cmake-3.0.2 in CMakeLists.txt (Clifford Yapp).
+  Fix "const" declaration info_ptr argument to png_get_eXIf_1() and the
+    num_exif argument to png_get_eXIf_1() (Github Issue 171).
+
+Version 1.6.32beta11 [August 7, 2017]
+  Added "eXIf" to "chunks_to_ignore[]" in png_set_keep_unknown_chunks().
+  Added huge_IDAT.png and empty_ancillary_chunks.png to testpngs/crashers.
+  Make pngtest --strict, --relax, --xfail options imply -m (multiple).
+  Removed unused chunk_name parameter from png_check_chunk_length().
+  Relocated setting free_me for eXIf data, to stop an OSS-fuzz leak.
+  Initialize profile_header[] in png_handle_iCCP() to fix OSS-fuzz issue.
+  Initialize png_ptr->row_buf[0] to 255 in png_read_row() to fix OSS-fuzz UMR.
+  Attempt to fix a UMR in png_set_text_2() to fix OSS-fuzz issue.
+  Increase minimum zlib stream from 9 to 14 in png_handle_iCCP(), to account
+    for the minimum 'deflate' stream, and relocate the test to a point
+    after the keyword has been read.
+  Check that the eXIf chunk has at least 2 bytes and begins with "II" or "MM".
+
+Version 1.6.32rc01 [August 18, 2017]
+  Added a set of "huge_xxxx_chunk.png" files to contrib/testpngs/crashers,
+    one for each known chunk type, with length = 2GB-1.
+  Check for 0 return from png_get_rowbytes() and added some (size_t) typecasts
+    in contrib/pngminus/*.c to stop some Coverity issues (162705, 162706,
+    and 162707).
+  Renamed chunks in contrib/testpngs/crashers to avoid having files whose
+    names differ only in case; this causes problems with some platforms
+    (github issue #172).
+
+Version 1.6.32rc02 [August 22, 2017]
+  Added contrib/oss-fuzz directory which contains files used by the oss-fuzz
+    project (https://github.com/google/oss-fuzz/tree/master/projects/libpng).
+
+Version 1.6.32 [August 24, 2017]
+  No changes.
+
+Version 1.6.33beta01 [August 28, 2017]
+  Added PNGMINUS_UNUSED macro to contrib/pngminus/p*.c and added missing
+    parenthesis in contrib/pngminus/pnm2png.c (bug report by Christian Hesse).
+  Fixed off-by-one error in png_do_check_palette_indexes() (Bug report
+    by Mick P., Source Forge Issue #269).
+
+Version 1.6.33beta02 [September 3, 2017]
+  Initialize png_handler.row_ptr in contrib/oss-fuzz/libpng_read_fuzzer.cc
+    to fix shortlived oss-fuzz issue 3234.
+  Compute a larger limit on IDAT because some applications write a deflate
+    buffer for each row (Bug report by Andrew Church).
+  Use current date (DATE) instead of release-date (RDATE) in last
+    changed date of contrib/oss-fuzz files.
+  Enabled ARM support in CMakeLists.txt (Bernd Kuhls).
+
+Version 1.6.33beta03 [September 14, 2017]
+  Fixed incorrect typecast of some arguments to png_malloc() and
+    png_calloc() that were png_uint_32 instead of png_alloc_size_t
+    (Bug report by "irwir" in Github libpng issue #175).
+  Use pnglibconf.h.prebuilt when building for ANDROID with cmake (Github
+    issue 162, by rcdailey).
+
+Version 1.6.33rc01 [September 20, 2017]
+  Initialize memory allocated by png_inflate to zero, using memset, to
+    stop an oss-fuzz "use of uninitialized value" detection in png_set_text_2()
+    due to truncated iTXt or zTXt chunk.
+  Initialize memory allocated by png_read_buffer to zero, using memset, to
+    stop an oss-fuzz "use of uninitialized value" detection in
+    png_icc_check_tag_table() due to truncated iCCP chunk.
+  Removed a redundant test (suggested by "irwir" in Github issue #180).
+
+Version 1.6.33rc02 [September 23, 2017]
+  Added an interlaced version of each file in contrib/pngsuite.
+  Relocate new memset() call in pngrutil.c.
+  Removed more redundant tests (suggested by "irwir" in Github issue #180).
+  Add support for loading images with associated alpha in the Simplified
+    API (Samuel Williams).
+
+Version 1.6.33 [September 28, 2017]
+  Revert contrib/oss-fuzz/libpng_read_fuzzer.cc to libpng-1.6.32 state.
+  Initialize png_handler.row_ptr in contrib/oss-fuzz/libpng_read_fuzzer.cc
+  Add end_info structure and png_read_end() to the libpng fuzzer.
+
+Version 1.6.34 [September 29, 2017]
+  Removed contrib/pngsuite/i*.png; some of these were incorrect and caused
+    test failures.
+
 Send comments/corrections/commendations to png-mng-implement at lists.sf.net
 (subscription required; visit
 https://lists.sourceforge.net/lists/listinfo/png-mng-implement
@@ -37,12 +37,21 @@ if(ARM OR AARCH64)
  endif()
 endif()

-if(ENABLE_SSE
+if(";${CPU_BASELINE_FINAL};" MATCHES "SSE2"
    AND (NOT MSVC OR (MSVC_VERSION GREATER 1799))) # MSVS2013+ (issue #7232)
-  list(APPEND lib_srcs contrib/intel/intel_init.c contrib/intel/filter_sse2_intrinsics.c)
+  list(APPEND lib_srcs intel/intel_init.c intel/filter_sse2_intrinsics.c)
  add_definitions(-DPNG_INTEL_SSE)
 endif()

+if(PPC64LE OR PPC64)
+  if(ENABLE_VSX AND NOT PPC64)
+    list(APPEND lib_srcs powerpc/powerpc_init.c powerpc/filter_vsx_intrinsics.c)
+    add_definitions(-DPNG_POWERPC_VSX_OPT=2)
+  else()
+    add_definitions(-DPNG_POWERPC_VSX_OPT=0)    
+  endif()
+endif()
+
 # ----------------------------------------------------------------------------------
 #         Define the library target:
 # ----------------------------------------------------------------------------------
@@ -54,12 +63,6 @@ endif(MSVC)
 add_library(${PNG_LIBRARY} STATIC ${lib_srcs} ${lib_hdrs})
 target_link_libraries(${PNG_LIBRARY} ${ZLIB_LIBRARIES})

-if(UNIX)
-  if(CMAKE_COMPILER_IS_GNUCXX OR CV_ICC)
-    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fPIC")
-  endif()
-endif()
-
 ocv_warnings_disable(CMAKE_C_FLAGS -Wcast-align)

 set_target_properties(${PNG_LIBRARY}
@@ -10,8 +10,8 @@ this sentence.

 This code is released under the libpng license.

-libpng versions 1.0.7, July 1, 2000 through 1.6.24, August 4, 2016 are
-Copyright (c) 2000-2002, 2004, 2006-2016 Glenn Randers-Pehrson, are
+libpng versions 1.0.7, July 1, 2000 through 1.6.34, September 29, 2017 are
+Copyright (c) 2000-2002, 2004, 2006-2017 Glenn Randers-Pehrson, are
 derived from libpng-1.0.6, and are distributed according to the same
 disclaimer and license as libpng-1.0.6 with the following individuals
 added to the list of Contributing Authors:
@@ -22,6 +22,9 @@ added to the list of Contributing Authors:
   Cosmin Truta
   Gilles Vollant
   James Yu
+   Mandar Sahastrabuddhe
+   Google Inc.
+   Vadim Barkov

 and with the following additions to the disclaimer:

@@ -127,4 +130,4 @@ any encryption software.  See the EAR, paragraphs 734.3(b)(3) and

 Glenn Randers-Pehrson
 glennrp at users.sourceforge.net
-August 4, 2016
+September 29, 2017
@@ -1,4 +1,4 @@
-README for libpng version 1.6.24 - August 4, 2016 (shared library 16.0)
+README for libpng version 1.6.34 - September 29, 2017 (shared library 16.0)
 See the note about version numbers near the top of png.h

 See INSTALL for instructions on how to install libpng.
@@ -23,7 +23,7 @@ earlier versions if you are using a shared library.  The type of the
 png_uint_32, which will affect shared-library applications that use
 this function.

-To avoid problems with changes to the internals of png info_struct,
+To avoid problems with changes to the internals of the png info_struct,
 new APIs have been made available in 0.95 to avoid direct application
 access to info_ptr.  These functions are the png_set_<chunk> and
 png_get_<chunk> functions.  These functions should be used when
@@ -88,11 +88,11 @@ zlib should be available at the same place that libpng is, or at zlib.net.

 You may also want a copy of the PNG specification.  It is available
 as an RFC, a W3C Recommendation, and an ISO/IEC Standard.  You can find
-these at http://www.libpng.org/pub/png/documents/
+these at http://www.libpng.org/pub/png/pngdocs.html .

-This code is currently being archived at libpng.sf.net in the
-[DOWNLOAD] area, and at ftp://ftp.simplesystems.org.  If you can't find it
-in any of those places, e-mail me, and I'll help you find it.
+This code is currently being archived at libpng.sourceforge.io in the
+[DOWNLOAD] area, and at http://libpng.download/src .  If you
+can't find it in any of those places, e-mail me, and I'll help you find it.

 I am not a lawyer, but I believe that the Export Control Classification
 Number (ECCN) for libpng is EAR99, which means not subject to export
@@ -179,18 +179,25 @@ Files in this distribution:
      pngwtran.c    =>  Write data transformations
      pngwutil.c    =>  Write utility functions
      arm           =>  Contains optimized code for the ARM platform
+      powerpc       =>  Contains optimized code for the PowerPC platform
      contrib       =>  Contributions
+       arm-neon         =>  Optimized code for ARM-NEON platform
+       powerpc-vsx      =>  Optimized code for POWERPC-VSX platform
       examples         =>  Example programs
       gregbook         =>  source code for PNG reading and writing, from
                            Greg Roelofs' "PNG: The Definitive Guide",
                            O'Reilly, 1999
       libtests         =>  Test programs
+       mips-msa         =>  Optimized code for MIPS-MSA platform
       pngminim         =>  Minimal decoder, encoder, and progressive decoder
                            programs demonstrating use of pngusr.dfa
       pngminus         =>  Simple pnm2png and png2pnm programs
       pngsuite         =>  Test images
+       testpngs
       tools            =>  Various tools
       visupng          =>  Contains a MSVC workspace for VisualPng
+      intel             =>  Optimized code for INTEL-SSE2 platform
+      mips              =>  Optimized code for MIPS platform
      projects      =>  Contains project files and workspaces for
                        building a DLL
       owatcom          =>  Contains a WATCOM project for building libpng
@@ -1,9 +1,9 @@

 /* filter_neon.S - NEON optimised filter functions
 *
- * Copyright (c) 2014 Glenn Randers-Pehrson
+ * Copyright (c) 2014,2017 Glenn Randers-Pehrson
 * Written by Mans Rullgard, 2011.
- * Last changed in libpng 1.6.16 [December 22, 2014]
+ * Last changed in libpng 1.6.31 [July 27, 2017]
 *
 * This code is released under the libpng license.
 * For conditions of distribution and use, see the disclaimer
@@ -16,7 +16,7 @@
 #define PNG_VERSION_INFO_ONLY
 #include "../pngpriv.h"

-#if defined(__linux__) && defined(__ELF__)
+#if (defined(__linux__) || defined(__FreeBSD__)) && defined(__ELF__)
 .section .note.GNU-stack,"",%progbits /* mark stack as non-executable */
 #endif

@@ -1,158 +0,0 @@
-Enabling SSE support
-
-Copyright (c) 2016 Google, Inc.
-Written by Mike Klein, Matt Sarett
-
-This INSTALL file written by Glenn Randers-Pehrson, 2016.
-
-If you have moved intel_init.c and filter_sse2_intrinsics.c to a different
-directory, be sure to update the '#include "../../pngpriv.h"' line in both
-files if necessary to point to the correct relative location of pngpriv.h
-with respect to the new location of those files.
-
-To enable SSE support in libpng, follow the instructions in I, II, or III,
-below:
-
-I. Using patched "configure" scripts:
-
-First, apply intel_sse.patch in your build directory.
-
-   patch -i contrib/intel/intel_sse.patch -p1
-
-Then, if you are not building in a new GIT clone, e.g., in a tar
-distribution, remove any existing pre-built configure scripts:
-
-   ./configure --enable-maintainer-mode
-   make maintainer-clean
-   ./autogen.sh --maintainer --clean
-
-Finally, configure libpng with -DPNG_INTEL_SSE in CPPFLAGS:
-
-   ./autogen.sh --maintainer
-   CPPFLAGS="-DPNG_INTEL_SSE" ./configure [options]
-   make CPPFLAGS="-DPNG_INTEL_SSE" [options]
-   make
-
-II. Using a custom makefile:
-
-If you are using a custom makefile makefile, you will have to update it
-manually to include contrib/intel/*.o in the dependencies, and to define
-PNG_INTEL_SSE.
-
-III. Using manually updated "configure" scripts:
-
-If you prefer, manually edit pngpriv.h, configure.ac, and Makefile.am,
-following the instructions below, then follow the instructions in
-section II of INSTALL in the main libpng directory, then configure libpng
-with -DPNG_INTEL_SSE in CPPFLAGS.
-
-1. Add the following code to configure.ac under HOST SPECIFIC OPTIONS
-directly beneath the section for ARM:
-
-----------------cut----------------
-# INTEL
-# =====
-#
-# INTEL SSE (SIMD) support.
-
-AC_ARG_ENABLE([intel-sse],
-   AS_HELP_STRING([[[--enable-intel-sse]]],
-      [Enable Intel SSE optimizations: =no/off, yes/on:]
-      [no/off: disable the optimizations;]
-      [yes/on: enable the optimizations.]
-      [If not specified: determined by the compiler.]),
-   [case "$enableval" in
-      no|off)
-         # disable the default enabling:
-         AC_DEFINE([PNG_INTEL_SSE_OPT], [0],
-                   [Disable Intel SSE optimizations])
-         # Prevent inclusion of the assembler files below:
-         enable_intel_sse=no;;
-      yes|on)
-         AC_DEFINE([PNG_INTEL_SSE_OPT], [1],
-                   [Enable Intel SSE optimizations]);;
-      *)
-         AC_MSG_ERROR([--enable-intel-sse=${enable_intel_sse}: invalid value])
-   esac])
-
-# Add Intel specific files to all builds where the host_cpu is Intel ('x86*')
-# or where Intel optimizations were explicitly requested (this allows a
-# fallback if a future host CPU does not match 'x86*')
-AM_CONDITIONAL([PNG_INTEL_SSE],
-   [test "$enable_intel_sse" != 'no' &&
-    case "$host_cpu" in
-      i?86|x86_64) :;;
-      *)    test "$enable_intel_sse" != '';;
-    esac])
-----------------cut----------------
-
-2. Add the following code to Makefile.am under HOST SPECIFIC OPTIONS
-directly beneath the "if PNG_ARM_NEON ... endif" statement:
-
-----------------cut----------------
-if PNG_INTEL_SSE
-libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += contrib/intel/intel_init.c\
-    contrib/intel/filter_sse2_intrinsics.c
-endif
-----------------cut----------------
-
-3. Add the following lines to pngpriv.h, following the PNG_ARM_NEON_OPT
-code:
-
-----------------cut----------------
-#ifndef PNG_INTEL_SSE_OPT
-#   ifdef PNG_INTEL_SSE
-      /* Only check for SSE if the build configuration has been modified to
-       * enable SSE optimizations.  This means that these optimizations will
-       * be off by default.  See contrib/intel for more details.
-       */
-#     if defined(__SSE4_1__) || defined(__AVX__) || defined(__SSSE3__) || \
-       defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \
-       (defined(_M_IX86_FP) && _M_IX86_FP >= 2)
-#         define PNG_INTEL_SSE_OPT 1
-#      endif
-#   endif
-#endif
-
-#if PNG_INTEL_SSE_OPT > 0
-#   ifndef PNG_INTEL_SSE_IMPLEMENTATION
-#      if defined(__SSE4_1__) || defined(__AVX__)
-          /* We are not actually using AVX, but checking for AVX is the best
-             way we can detect SSE4.1 and SSSE3 on MSVC.
-          */
-#         define PNG_INTEL_SSE_IMPLEMENTATION 3
-#      elif defined(__SSSE3__)
-#         define PNG_INTEL_SSE_IMPLEMENTATION 2
-#      elif defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \
-       (defined(_M_IX86_FP) && _M_IX86_FP >= 2)
-#         define PNG_INTEL_SSE_IMPLEMENTATION 1
-#      else
-#         define PNG_INTEL_SSE_IMPLEMENTATION 0
-#      endif
-#   endif
-
-#   if PNG_INTEL_SSE_IMPLEMENTATION > 0
-#      define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_sse2
-#   endif
-#endif
-
-----------------cut----------------
-
-4. Add the following lines to pngpriv.h, following the prototype for
-png_read_filter_row_paeth4_neon:
-
-----------------cut----------------
-PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_sse2,(png_row_infop
-    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_sse2,(png_row_infop
-    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_sse2,(png_row_infop
-    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_sse2,(png_row_infop
-    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_sse2,(png_row_infop
-    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_sse2,(png_row_infop
-    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-
-----------------cut----------------
@@ -1,187 +0,0 @@
-diff --git a/configure.ac b/configure.ac
--- a/configure.ac	2016-05-25 18:59:10.000000000 -0400
-+++ b/configure.ac	2016-05-25 19:48:10.631751170 -0400
-@@ -341,16 +341,50 @@ AC_ARG_ENABLE([arm-neon],
- 
- AM_CONDITIONAL([PNG_ARM_NEON],
-    [test "$enable_arm_neon" != 'no' &&
-     case "$host_cpu" in
-       arm*|aarch64*) :;;
-       *)    test "$enable_arm_neon" != '';;
-     esac])
- 
-+# INTEL
-+# =====
-+#
-+# INTEL SSE (SIMD) support.
-+
-+AC_ARG_ENABLE([intel-sse],
-+   AS_HELP_STRING([[[--enable-intel-sse]]],
-+      [Enable Intel SSE optimizations: =no/off, yes/on:]
-+      [no/off: disable the optimizations;]
-+      [yes/on: enable the optimizations.]
-+      [If not specified: determined by the compiler.]),
-+   [case "$enableval" in
-+      no|off)
-+         # disable the default enabling:
-+         AC_DEFINE([PNG_INTEL_SSE_OPT], [0],
-+                   [Disable Intel SSE optimizations])
-+         # Prevent inclusion of the assembler files below:
-+         enable_intel_sse=no;;
-+      yes|on)
-+         AC_DEFINE([PNG_INTEL_SSE_OPT], [1],
-+                   [Enable Intel SSE optimizations]);;
-+      *)
-+         AC_MSG_ERROR([--enable-intel-sse=${enable_intel_sse}: invalid value])
-+   esac])
-+
-+# Add Intel specific files to all builds where the host_cpu is Intel ('x86*')
-+# or where Intel optimizations were explicitly requested (this allows a
-+# fallback if a future host CPU does not match 'x86*')
-+AM_CONDITIONAL([PNG_INTEL_SSE],
-+   [test "$enable_intel_sse" != 'no' &&
-+    case "$host_cpu" in
-+      i?86|x86_64) :;;
-+      *)    test "$enable_intel_sse" != '';;
-+    esac])
- AC_MSG_NOTICE([[Extra options for compiler: $PNG_COPTS]])
- 
- # Config files, substituting as above
- AC_CONFIG_FILES([Makefile libpng.pc:libpng.pc.in])
- AC_CONFIG_FILES([libpng-config:libpng-config.in],
-    [chmod +x libpng-config])
- 
- AC_OUTPUT
-diff --git a/Makefile.am b/Makefile.am
--- a/Makefile.am	2016-05-17 18:15:12.000000000 -0400
-+++ b/Makefile.am	2016-05-25 19:48:10.631751170 -0400
-@@ -92,16 +92,20 @@ libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SO
- 	pngset.c pngtrans.c pngwio.c pngwrite.c pngwtran.c pngwutil.c\
- 	png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h pngusr.dfa
- 
- if PNG_ARM_NEON
- libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += arm/arm_init.c\
- 	arm/filter_neon.S arm/filter_neon_intrinsics.c
- endif
- 
-+if PNG_INTEL_SSE
-+libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += contrib/intel/intel_init.c\
-+    contrib/intel/filter_sse2_intrinsics.c
-+endif
- nodist_libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES = pnglibconf.h
- 
- libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_LDFLAGS = -no-undefined -export-dynamic \
- 	-version-number @PNGLIB_MAJOR@@PNGLIB_MINOR@:@PNGLIB_RELEASE@:0
- 
- if HAVE_LD_VERSION_SCRIPT
- #   Versioned symbols and restricted exports
- if HAVE_SOLARIS_LD
-diff --git a/pngpriv.h b/pngpriv.h
--- a/pngpriv.h	2016-08-01 18:13:38.770128810 -0500
-+++ b/pngpriv.h	2016-08-01 18:50:19.130179017 -0500
-@@ -177,16 +177,52 @@
- #  endif /* !PNG_ARM_NEON_IMPLEMENTATION */
- 
- #  ifndef PNG_ARM_NEON_IMPLEMENTATION
-       /* Use the intrinsics code by default. */
- #     define PNG_ARM_NEON_IMPLEMENTATION 1
- #  endif
- #endif /* PNG_ARM_NEON_OPT > 0 */
- 
-+#ifndef PNG_INTEL_SSE_OPT
-+#   ifdef PNG_INTEL_SSE
-+      /* Only check for SSE if the build configuration has been modified to
-+       * enable SSE optimizations.  This means that these optimizations will
-+       * be off by default.  See contrib/intel for more details.
-+       */
-+#     if defined(__SSE4_1__) || defined(__AVX__) || defined(__SSSE3__) || \
-+       defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \
-+       (defined(_M_IX86_FP) && _M_IX86_FP >= 2)
-+#         define PNG_INTEL_SSE_OPT 1
-+#      endif
-+#   endif
-+#endif
-+
-+#if PNG_INTEL_SSE_OPT > 0
-+#   ifndef PNG_INTEL_SSE_IMPLEMENTATION
-+#      if defined(__SSE4_1__) || defined(__AVX__)
-+          /* We are not actually using AVX, but checking for AVX is the best
-+             way we can detect SSE4.1 and SSSE3 on MSVC.
-+          */
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 3
-+#      elif defined(__SSSE3__)
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 2
-+#      elif defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \
-+       (defined(_M_IX86_FP) && _M_IX86_FP >= 2)
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 1
-+#      else
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 0
-+#      endif
-+#   endif
-+
-+#   if PNG_INTEL_SSE_IMPLEMENTATION > 0
-+#      define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_sse2
-+#   endif
-+#endif
-+
- /* Is this a build of a DLL where compilation of the object modules requires
-  * different preprocessor settings to those required for a simple library?  If
-  * so PNG_BUILD_DLL must be set.
-  *
-  * If libpng is used inside a DLL but that DLL does not export the libpng APIs
-  * PNG_BUILD_DLL must not be set.  To avoid the code below kicking in build a
-  * static library of libpng then link the DLL against that.
-  */
-@@ -1185,16 +1221,31 @@ PNG_INTERNAL_FUNCTION(void,png_read_filt
-     row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
- PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_neon,(png_row_infop
-     row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
- PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_neon,(png_row_infop
-     row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
- PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_neon,(png_row_infop
-     row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
- #endif
-+ 
-+#if PNG_INTEL_SSE_IMPLEMENTATION > 0
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+#endif
- 
- /* Choose the best filter to use and filter the row data */
- PNG_INTERNAL_FUNCTION(void,png_write_find_filter,(png_structrp png_ptr,
-     png_row_infop row_info),PNG_EMPTY);
- 
- #ifdef PNG_SEQUENTIAL_READ_SUPPORTED
- PNG_INTERNAL_FUNCTION(void,png_read_IDAT_data,(png_structrp png_ptr,
-    png_bytep output, png_alloc_size_t avail_out),PNG_EMPTY);
-@@ -1914,16 +1965,20 @@ PNG_INTERNAL_FUNCTION(void, PNG_FILTER_O
-    /* List *all* the possible optimizations here - this branch is required if
-     * the builder of libpng passes the definition of PNG_FILTER_OPTIMIZATIONS in
-     * CFLAGS in place of CPPFLAGS *and* uses symbol prefixing.
-     */
- #  if PNG_ARM_NEON_OPT > 0
- PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_neon,
-    (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
- #  endif
-+#  if PNG_INTEL_SSE_IMPLEMENTATION > 0
-+PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_sse2,
-+   (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
-+#  endif
- #endif
- 
- PNG_INTERNAL_FUNCTION(png_uint_32, png_check_keyword, (png_structrp png_ptr,
-    png_const_charp key, png_bytep new_key), PNG_EMPTY);
- 
- /* Maintainer: Put new private prototypes here ^ */
- 
- #include "pngdebug.h"
@@ -1,19 +1,18 @@

 /* filter_sse2_intrinsics.c - SSE2 optimized filter functions
 *
- * Copyright (c) 2016 Google, Inc.
+ * Copyright (c) 2016-2017 Glenn Randers-Pehrson
 * Written by Mike Klein and Matt Sarett
- * Derived from arm/filter_neon_intrinsics.c, which was
- * Copyright (c) 2014,2016 Glenn Randers-Pehrson
+ * Derived from arm/filter_neon_intrinsics.c
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
+ * Last changed in libpng 1.6.31 [July 27, 2017]
 *
 * This code is released under the libpng license.
 * For conditions of distribution and use, see the disclaimer
 * and license in png.h
 */

-#include "../../pngpriv.h"
+#include "../pngpriv.h"

 #ifdef PNG_READ_SUPPORTED

@@ -41,7 +40,7 @@ static __m128i load3(const void* p) {
   /* We'll load 2 bytes, then 1 byte,
    * then mask them together, and finally load into SSE.
    */
-   const png_uint_16* p01 = p;
+   const png_uint_16* p01 = (png_const_uint_16p)p;
   const png_byte*    p2  = (const png_byte*)(p01+1);

   png_uint_32 v012 = (png_uint_32)(*p01)
@@ -54,12 +53,15 @@ static void store3(void* p, __m128i v) {
    * its bottom two bytes, then its third byte.
    */
   png_uint_32 v012;
+   png_uint_16* p01;
+   png_byte*    p2;
+
   store4(&v012, v);

-   png_uint_16* p01 = p;
-   png_byte*    p2  = (png_byte*)(p01+1);
-   *p01 = v012;
-   *p2  = v012 >> 16;
+   p01 = (png_uint_16p)p;
+   p2  = (png_byte*)(p01+1);
+   *p01 = (png_uint_16)v012;
+   *p2  = (png_byte)(v012 >> 16);
 }

 void png_read_filter_row_sub3_sse2(png_row_infop row_info, png_bytep row,
@@ -69,11 +71,13 @@ void png_read_filter_row_sub3_sse2(png_row_infop row_info, png_bytep row,
    * There is no pixel to the left of the first pixel.  It's encoded directly.
    * That works with our main loop if we just say that left pixel was zero.
    */
-   (void) prev; //unused parameter
-   png_debug(1, "in png_read_filter_row_sub3_sse2");
+   png_size_t rb;
+
   __m128i a, d = _mm_setzero_si128();

-   png_size_t rb = row_info->rowbytes;
+   png_debug(1, "in png_read_filter_row_sub3_sse2");
+
+   rb = row_info->rowbytes;
   while (rb >= 4) {
      a = d; d = load4(row);
      d = _mm_add_epi8(d, a);
@@ -90,6 +94,7 @@ void png_read_filter_row_sub3_sse2(png_row_infop row_info, png_bytep row,
      row += 3;
      rb  -= 3;
   }
+   PNG_UNUSED(prev)
 }

 void png_read_filter_row_sub4_sse2(png_row_infop row_info, png_bytep row,
@@ -99,12 +104,14 @@ void png_read_filter_row_sub4_sse2(png_row_infop row_info, png_bytep row,
    * There is no pixel to the left of the first pixel.  It's encoded directly.
    * That works with our main loop if we just say that left pixel was zero.
    */
-   (void) prev; //unused parameter
-   png_debug(1, "in png_read_filter_row_sub4_sse2");
+   png_size_t rb;
+
   __m128i a, d = _mm_setzero_si128();

-   png_size_t rb = row_info->rowbytes;
-   while (rb > 0) {
+   png_debug(1, "in png_read_filter_row_sub4_sse2");
+
+   rb = row_info->rowbytes+4;
+   while (rb > 4) {
      a = d; d = load4(row);
      d = _mm_add_epi8(d, a);
      store4(row, d);
@@ -112,6 +119,7 @@ void png_read_filter_row_sub4_sse2(png_row_infop row_info, png_bytep row,
      row += 4;
      rb  -= 4;
   }
+   PNG_UNUSED(prev)
 }

 void png_read_filter_row_avg3_sse2(png_row_infop row_info, png_bytep row,
@@ -122,18 +130,23 @@ void png_read_filter_row_avg3_sse2(png_row_infop row_info, png_bytep row,
    * predicted to be half of the pixel above it.  So again, this works
    * perfectly with our loop if we make sure a starts at zero.
    */
-   png_debug(1, "in png_read_filter_row_avg3_sse2");
+
+   png_size_t rb;
+
   const __m128i zero = _mm_setzero_si128();
+
   __m128i    b;
   __m128i a, d = zero;

-   png_size_t rb = row_info->rowbytes;
+   png_debug(1, "in png_read_filter_row_avg3_sse2");
+   rb = row_info->rowbytes;
   while (rb >= 4) {
+      __m128i avg;
             b = load4(prev);
      a = d; d = load4(row );

      /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */
-      __m128i avg = _mm_avg_epu8(a,b);
+      avg = _mm_avg_epu8(a,b);
      /* ...but we can fix it up by subtracting off 1 if it rounded up. */
      avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b),
                                            _mm_set1_epi8(1)));
@@ -145,11 +158,12 @@ void png_read_filter_row_avg3_sse2(png_row_infop row_info, png_bytep row,
      rb   -= 3;
   }
   if (rb > 0) {
+      __m128i avg;
             b = load3(prev);
      a = d; d = load3(row );

      /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */
-      __m128i avg = _mm_avg_epu8(a,b);
+      avg = _mm_avg_epu8(a,b);
      /* ...but we can fix it up by subtracting off 1 if it rounded up. */
      avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b),
                                            _mm_set1_epi8(1)));
@@ -171,18 +185,21 @@ void png_read_filter_row_avg4_sse2(png_row_infop row_info, png_bytep row,
    * predicted to be half of the pixel above it.  So again, this works
    * perfectly with our loop if we make sure a starts at zero.
    */
-   png_debug(1, "in png_read_filter_row_avg4_sse2");
+   png_size_t rb;
   const __m128i zero = _mm_setzero_si128();
   __m128i    b;
   __m128i a, d = zero;

-   png_size_t rb = row_info->rowbytes;
-   while (rb > 0) {
+   png_debug(1, "in png_read_filter_row_avg4_sse2");
+
+   rb = row_info->rowbytes+4;
+   while (rb > 4) {
+      __m128i avg;
             b = load4(prev);
      a = d; d = load4(row );

      /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */
-      __m128i avg = _mm_avg_epu8(a,b);
+      avg = _mm_avg_epu8(a,b);
      /* ...but we can fix it up by subtracting off 1 if it rounded up. */
      avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b),
                                            _mm_set1_epi8(1)));
@@ -240,38 +257,42 @@ void png_read_filter_row_paeth3_sse2(png_row_infop row_info, png_bytep row,
    * Here we zero b and d, which become c and a respectively at the start of
    * the loop.
    */
-   png_debug(1, "in png_read_filter_row_paeth3_sse2");
+   png_size_t rb;
   const __m128i zero = _mm_setzero_si128();
   __m128i c, b = zero,
           a, d = zero;

-   png_size_t rb = row_info->rowbytes;
+   png_debug(1, "in png_read_filter_row_paeth3_sse2");
+
+   rb = row_info->rowbytes;
   while (rb >= 4) {
      /* It's easiest to do this math (particularly, deal with pc) with 16-bit
       * intermediates.
       */
+      __m128i pa,pb,pc,smallest,nearest;
      c = b; b = _mm_unpacklo_epi8(load4(prev), zero);
      a = d; d = _mm_unpacklo_epi8(load4(row ), zero);

      /* (p-a) == (a+b-c - a) == (b-c) */
-      __m128i pa = _mm_sub_epi16(b,c);
+   
+      pa = _mm_sub_epi16(b,c);

      /* (p-b) == (a+b-c - b) == (a-c) */
-      __m128i pb = _mm_sub_epi16(a,c);
+      pb = _mm_sub_epi16(a,c);

      /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */
-      __m128i pc = _mm_add_epi16(pa,pb);
+      pc = _mm_add_epi16(pa,pb);

      pa = abs_i16(pa);  /* |p-a| */
      pb = abs_i16(pb);  /* |p-b| */
      pc = abs_i16(pc);  /* |p-c| */

-      __m128i smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));
+      smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));

      /* Paeth breaks ties favoring a over b over c. */
-      __m128i nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
-                         if_then_else(_mm_cmpeq_epi16(smallest, pb), b,
-                                                                     c));
+      nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
+                 if_then_else(_mm_cmpeq_epi16(smallest, pb), b,
+                                                             c));

      /* Note `_epi8`: we need addition to wrap modulo 255. */
      d = _mm_add_epi8(d, nearest);
@@ -285,26 +306,27 @@ void png_read_filter_row_paeth3_sse2(png_row_infop row_info, png_bytep row,
      /* It's easiest to do this math (particularly, deal with pc) with 16-bit
       * intermediates.
       */
+      __m128i pa,pb,pc,smallest,nearest;
      c = b; b = _mm_unpacklo_epi8(load3(prev), zero);
      a = d; d = _mm_unpacklo_epi8(load3(row ), zero);

      /* (p-a) == (a+b-c - a) == (b-c) */
-      __m128i pa = _mm_sub_epi16(b,c);
+      pa = _mm_sub_epi16(b,c);

      /* (p-b) == (a+b-c - b) == (a-c) */
-      __m128i pb = _mm_sub_epi16(a,c);
+      pb = _mm_sub_epi16(a,c);

      /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */
-      __m128i pc = _mm_add_epi16(pa,pb);
+      pc = _mm_add_epi16(pa,pb);

      pa = abs_i16(pa);  /* |p-a| */
      pb = abs_i16(pb);  /* |p-b| */
      pc = abs_i16(pc);  /* |p-c| */

-      __m128i smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));
+      smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));

      /* Paeth breaks ties favoring a over b over c. */
-      __m128i nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
+      nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
                         if_then_else(_mm_cmpeq_epi16(smallest, pb), b,
                                                                     c));

@@ -334,13 +356,16 @@ void png_read_filter_row_paeth4_sse2(png_row_infop row_info, png_bytep row,
    * Here we zero b and d, which become c and a respectively at the start of
    * the loop.
    */
-   png_debug(1, "in png_read_filter_row_paeth4_sse2");
+   png_size_t rb;
   const __m128i zero = _mm_setzero_si128();
+   __m128i pa,pb,pc,smallest,nearest;
   __m128i c, b = zero,
           a, d = zero;

-   png_size_t rb = row_info->rowbytes;
-   while (rb > 0) {
+   png_debug(1, "in png_read_filter_row_paeth4_sse2");
+
+   rb = row_info->rowbytes+4;
+   while (rb > 4) {
      /* It's easiest to do this math (particularly, deal with pc) with 16-bit
       * intermediates.
       */
@@ -348,22 +373,22 @@ void png_read_filter_row_paeth4_sse2(png_row_infop row_info, png_bytep row,
      a = d; d = _mm_unpacklo_epi8(load4(row ), zero);

      /* (p-a) == (a+b-c - a) == (b-c) */
-      __m128i pa = _mm_sub_epi16(b,c);
+      pa = _mm_sub_epi16(b,c);

      /* (p-b) == (a+b-c - b) == (a-c) */
-      __m128i pb = _mm_sub_epi16(a,c);
+      pb = _mm_sub_epi16(a,c);

      /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */
-      __m128i pc = _mm_add_epi16(pa,pb);
+      pc = _mm_add_epi16(pa,pb);

      pa = abs_i16(pa);  /* |p-a| */
      pb = abs_i16(pb);  /* |p-b| */
      pc = abs_i16(pc);  /* |p-c| */

-      __m128i smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));
+      smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));

      /* Paeth breaks ties favoring a over b over c. */
-      __m128i nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
+      nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
                         if_then_else(_mm_cmpeq_epi16(smallest, pb), b,
                                                                     c));

@@ -1,19 +1,18 @@

 /* intel_init.c - SSE2 optimized filter functions
 *
- * Copyright (c) 2016 Google, Inc.
- * Written by Mike Klein and Matt Sarett
- * Derived from arm/arm_init.c, which was
- * Copyright (c) 2014,2016 Glenn Randers-Pehrson
+ * Copyright (c) 2016-2017 Glenn Randers-Pehrson
+ * Written by Mike Klein and Matt Sarett, Google, Inc.
+ * Derived from arm/arm_init.c
 *
- * Last changed in libpng 1.6.22 [May 26, 2016]
+ * Last changed in libpng 1.6.29 [March 16, 2017]
 *
 * This code is released under the libpng license.
 * For conditions of distribution and use, see the disclaimer
 * and license in png.h
 */

-#include "../../pngpriv.h"
+#include "../pngpriv.h"

 #ifdef PNG_READ_SUPPORTED
 #if PNG_INTEL_SSE_IMPLEMENTATION > 0
@@ -1,67 +1,15 @@
-diff --git a/3rdparty/libpng/CMakeLists.txt b/3rdparty/libpng/CMakeLists.txt
-index fee9c99..eaf2095 100644
--- a/3rdparty/libpng/CMakeLists.txt
-+++ b/3rdparty/libpng/CMakeLists.txt
-@@ -33,6 +33,11 @@ if(ENABLE_NEON)
-   add_definitions(-DPNG_ARM_NEON_OPT=2)
- endif()
- 
-+if(ENABLE_SSE)
-+  list(APPEND lib_srcs contrib/intel/intel_init.c contrib/intel/filter_sse2_intrinsics.c)
-+  add_definitions(-DPNG_INTEL_SSE)
-+endif()
-+
- # ----------------------------------------------------------------------------------
- #         Define the library target:
- # ----------------------------------------------------------------------------------
-diff --git a/3rdparty/libpng/pngpriv.h b/3rdparty/libpng/pngpriv.h
-index fe3355d..c0aa785 100644
 --- a/3rdparty/libpng/pngpriv.h
 +++ b/3rdparty/libpng/pngpriv.h
-@@ -182,6 +182,42 @@
- #  endif
- #endif /* PNG_ARM_NEON_OPT > 0 */
+@@ -214,7 +214,7 @@
+ #   endif
+ #endif
 
-+#ifndef PNG_INTEL_SSE_OPT
-+#   ifdef PNG_INTEL_SSE
-+      /* Only check for SSE if the build configuration has been modified to
-+       * enable SSE optimizations.  This means that these optimizations will
-+       * be off by default.  See contrib/intel for more details.
-+       */
-+#     if defined(__SSE4_1__) || defined(__AVX__) || defined(__SSSE3__) || \
-+       defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \
-+       (defined(_M_IX86_FP) && _M_IX86_FP >= 2)
-+#         define PNG_INTEL_SSE_OPT 1
-+#      endif
-+#   endif
-+#endif
-+
+-#if PNG_INTEL_SSE_OPT > 0
 +#if defined(PNG_INTEL_SSE_OPT) && PNG_INTEL_SSE_OPT > 0
-+#   ifndef PNG_INTEL_SSE_IMPLEMENTATION
-+#      if defined(__SSE4_1__) || defined(__AVX__)
-+          /* We are not actually using AVX, but checking for AVX is the best
-+             way we can detect SSE4.1 and SSSE3 on MSVC.
-+          */
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 3
-+#      elif defined(__SSSE3__)
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 2
-+#      elif defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \
-+       (defined(_M_IX86_FP) && _M_IX86_FP >= 2)
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 1
-+#      else
-+#         define PNG_INTEL_SSE_IMPLEMENTATION 0
-+#      endif
-+#   endif
-+
-+#   if PNG_INTEL_SSE_IMPLEMENTATION > 0
-+#      define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_sse2
-+#   endif
-+#endif
-+
- /* Is this a build of a DLL where compilation of the object modules requires
-  * different preprocessor settings to those required for a simple library?  If
-  * so PNG_BUILD_DLL must be set.
-@@ -457,7 +493,7 @@
+ #   ifndef PNG_INTEL_SSE_IMPLEMENTATION
+ #      if defined(__SSE4_1__) || defined(__AVX__)
+           /* We are not actually using AVX, but checking for AVX is the best
+@@ -547,7 +547,7 @@
 
 /* Memory model/platform independent fns */
 #ifndef PNG_ABORT
@@ -70,36 +18,21 @@ index fe3355d..c0aa785 100644
 #    define PNG_ABORT() ExitProcess(0)
 #  else
 #    define PNG_ABORT() abort()
-@@ -1190,6 +1226,21 @@ PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_neon,(png_row_infop
- PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_neon,(png_row_infop
+@@ -1340,7 +1340,7 @@ PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_vsx,(png_row_infop
     row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
 #endif
-+ 
-+#if defined(PNG_INTEL_SSE_IMPLEMENTATION) && PNG_INTEL_SSE_IMPLEMENTATION > 0
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_sse2,(png_row_infop
-+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
-+#endif
 
- /* Choose the best filter to use and filter the row data */
- PNG_INTERNAL_FUNCTION(void,png_write_find_filter,(png_structrp png_ptr,
-@@ -1919,6 +1970,10 @@ PNG_INTERNAL_FUNCTION(void, PNG_FILTER_OPTIMIZATIONS, (png_structp png_ptr,
- PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_neon,
+-#if PNG_INTEL_SSE_IMPLEMENTATION > 0
+#if defined(PNG_INTEL_SSE_IMPLEMENTATION) && PNG_INTEL_SSE_IMPLEMENTATION > 0
+ PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_sse2,(png_row_infop
+     row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+ PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_sse2,(png_row_infop
+@@ -2099,7 +2099,7 @@ PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_msa,
    (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
- #  endif
-+#  if defined(PNG_INTEL_SSE_IMPLEMENTATION) && PNG_INTEL_SSE_IMPLEMENTATION > 0
-+PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_sse2,
-+   (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
-+#  endif
 #endif
 
- PNG_INTERNAL_FUNCTION(png_uint_32, png_check_keyword, (png_structrp png_ptr,
+-#  if PNG_INTEL_SSE_IMPLEMENTATION > 0
+#  if defined(PNG_INTEL_SSE_IMPLEMENTATION) && PNG_INTEL_SSE_IMPLEMENTATION > 0
+ PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_sse2,
+    (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
+ #  endif
@@ -1,8 +1,8 @@

 /* png.c - location for general purpose libpng functions
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2015 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.33 [September 28, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -14,7 +14,27 @@
 #include "pngpriv.h"

 /* Generate a compiler error if there is an old png.h in the search path. */
-typedef png_libpng_version_1_6_24 Your_png_h_is_not_version_1_6_24;
+typedef png_libpng_version_1_6_34 Your_png_h_is_not_version_1_6_34;
+
+#ifdef __GNUC__
+/* The version tests may need to be added to, but the problem warning has
+ * consistently been fixed in GCC versions which obtain wide-spread release.
+ * The problem is that many versions of GCC rearrange comparison expressions in
+ * the optimizer in such a way that the results of the comparison will change
+ * if signed integer overflow occurs.  Such comparisons are not permitted in
+ * ANSI C90, however GCC isn't clever enough to work out that that do not occur
+ * below in png_ascii_from_fp and png_muldiv, so it produces a warning with
+ * -Wextra.  Unfortunately this is highly dependent on the optimizer and the
+ * machine architecture so the warning comes and goes unpredictably and is
+ * impossible to "fix", even were that a good idea.
+ */
+#if __GNUC__ == 7 && __GNUC_MINOR__ == 1
+#define GCC_STRICT_OVERFLOW 1
+#endif /* GNU 7.1.x */
+#endif /* GNU */
+#ifndef GCC_STRICT_OVERFLOW
+#define GCC_STRICT_OVERFLOW 0
+#endif

 /* Tells libpng that we have already handled the first "num_bytes" bytes
 * of the PNG file signature.  If the PNG data is embedded into another
@@ -458,7 +478,7 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask,

 #ifdef PNG_TEXT_SUPPORTED
   /* Free text item num or (if num == -1) all text items */
-   if (info_ptr->text != 0 &&
+   if (info_ptr->text != NULL &&
       ((mask & PNG_FREE_TEXT) & info_ptr->free_me) != 0)
   {
      if (num != -1)
@@ -477,6 +497,7 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask,
         png_free(png_ptr, info_ptr->text);
         info_ptr->text = NULL;
         info_ptr->num_text = 0;
+         info_ptr->max_text = 0;
      }
   }
 #endif
@@ -541,7 +562,7 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask,

 #ifdef PNG_sPLT_SUPPORTED
   /* Free a given sPLT entry, or (if num == -1) all sPLT entries */
-   if (info_ptr->splt_palettes != 0 &&
+   if (info_ptr->splt_palettes != NULL &&
       ((mask & PNG_FREE_SPLT) & info_ptr->free_me) != 0)
   {
      if (num != -1)
@@ -571,7 +592,7 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask,
 #endif

 #ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED
-   if (info_ptr->unknown_chunks != 0 &&
+   if (info_ptr->unknown_chunks != NULL &&
       ((mask & PNG_FREE_UNKN) & info_ptr->free_me) != 0)
   {
      if (num != -1)
@@ -594,6 +615,26 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask,
   }
 #endif

+#ifdef PNG_eXIf_SUPPORTED
+   /* Free any eXIf entry */
+   if (((mask & PNG_FREE_EXIF) & info_ptr->free_me) != 0)
+   {
+# ifdef PNG_READ_eXIf_SUPPORTED
+      if (info_ptr->eXIf_buf)
+      {
+         png_free(png_ptr, info_ptr->eXIf_buf);
+         info_ptr->eXIf_buf = NULL;
+      }
+# endif
+      if (info_ptr->exif)
+      {
+         png_free(png_ptr, info_ptr->exif);
+         info_ptr->exif = NULL;
+      }
+      info_ptr->valid &= ~PNG_INFO_eXIf;
+   }
+#endif
+
 #ifdef PNG_hIST_SUPPORTED
   /* Free any hIST entry */
   if (((mask & PNG_FREE_HIST) & info_ptr->free_me) != 0)
@@ -617,7 +658,7 @@ png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask,
   /* Free any image bits attached to the info structure */
   if (((mask & PNG_FREE_ROWS) & info_ptr->free_me) != 0)
   {
-      if (info_ptr->row_pointers != 0)
+      if (info_ptr->row_pointers != NULL)
      {
         png_uint_32 row;
         for (row = 0; row < info_ptr->height; row++)
@@ -684,7 +725,7 @@ png_init_io(png_structrp png_ptr, png_FILE_p fp)
 void PNGAPI
 png_save_int_32(png_bytep buf, png_int_32 i)
 {
-   png_save_uint_32(buf, i);
+   png_save_uint_32(buf, (png_uint_32)i);
 }
 #  endif

@@ -775,15 +816,15 @@ png_get_copyright(png_const_structrp png_ptr)
 #else
 #  ifdef __STDC__
   return PNG_STRING_NEWLINE \
-      "libpng version 1.6.24 - August 4, 2016" PNG_STRING_NEWLINE \
-      "Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson" \
+      "libpng version 1.6.34 - September 29, 2017" PNG_STRING_NEWLINE \
+      "Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson" \
      PNG_STRING_NEWLINE \
      "Copyright (c) 1996-1997 Andreas Dilger" PNG_STRING_NEWLINE \
      "Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc." \
      PNG_STRING_NEWLINE;
 #  else
-   return "libpng version 1.6.24 - August 4, 2016\
-      Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson\
+   return "libpng version 1.6.34 - September 29, 2017\
+      Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson\
      Copyright (c) 1996-1997 Andreas Dilger\
      Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc.";
 #  endif
@@ -1872,12 +1913,12 @@ png_colorspace_set_sRGB(png_const_structrp png_ptr, png_colorspacerp colorspace,
    */
   if (intent < 0 || intent >= PNG_sRGB_INTENT_LAST)
      return png_icc_profile_error(png_ptr, colorspace, "sRGB",
-          (unsigned)intent, "invalid sRGB rendering intent");
+          (png_alloc_size_t)intent, "invalid sRGB rendering intent");

   if ((colorspace->flags & PNG_COLORSPACE_HAVE_INTENT) != 0 &&
       colorspace->rendering_intent != intent)
      return png_icc_profile_error(png_ptr, colorspace, "sRGB",
-         (unsigned)intent, "inconsistent rendering intents");
+         (png_alloc_size_t)intent, "inconsistent rendering intents");

   if ((colorspace->flags & PNG_COLORSPACE_FROM_sRGB) != 0)
   {
@@ -1931,16 +1972,49 @@ png_colorspace_set_sRGB(png_const_structrp png_ptr, png_colorspacerp colorspace,
 static const png_byte D50_nCIEXYZ[12] =
   { 0x00, 0x00, 0xf6, 0xd6, 0x00, 0x01, 0x00, 0x00, 0x00, 0x00, 0xd3, 0x2d };

-int /* PRIVATE */
-png_icc_check_length(png_const_structrp png_ptr, png_colorspacerp colorspace,
+static int /* bool */
+icc_check_length(png_const_structrp png_ptr, png_colorspacerp colorspace,
    png_const_charp name, png_uint_32 profile_length)
 {
   if (profile_length < 132)
      return png_icc_profile_error(png_ptr, colorspace, name, profile_length,
          "too short");
+   return 1;
+}
+
+#ifdef PNG_READ_iCCP_SUPPORTED
+int /* PRIVATE */
+png_icc_check_length(png_const_structrp png_ptr, png_colorspacerp colorspace,
+    png_const_charp name, png_uint_32 profile_length)
+{
+   if (!icc_check_length(png_ptr, colorspace, name, profile_length))
+      return 0;
+
+   /* This needs to be here because the 'normal' check is in
+    * png_decompress_chunk, yet this happens after the attempt to
+    * png_malloc_base the required data.  We only need this on read; on write
+    * the caller supplies the profile buffer so libpng doesn't allocate it.  See
+    * the call to icc_check_length below (the write case).
+    */
+#  ifdef PNG_SET_USER_LIMITS_SUPPORTED
+      else if (png_ptr->user_chunk_malloc_max > 0 &&
+               png_ptr->user_chunk_malloc_max < profile_length)
+         return png_icc_profile_error(png_ptr, colorspace, name, profile_length,
+             "exceeds application limits");
+#  elif PNG_USER_CHUNK_MALLOC_MAX > 0
+      else if (PNG_USER_CHUNK_MALLOC_MAX < profile_length)
+         return png_icc_profile_error(png_ptr, colorspace, name, profile_length,
+             "exceeds libpng limits");
+#  else /* !SET_USER_LIMITS */
+      /* This will get compiled out on all 32-bit and better systems. */
+      else if (PNG_SIZE_MAX < profile_length)
+         return png_icc_profile_error(png_ptr, colorspace, name, profile_length,
+             "exceeds system limits");
+#  endif /* !SET_USER_LIMITS */

   return 1;
 }
+#endif /* READ_iCCP */

 int /* PRIVATE */
 png_icc_check_header(png_const_structrp png_ptr, png_colorspacerp colorspace,
@@ -2149,15 +2223,6 @@ png_icc_check_tag_table(png_const_structrp png_ptr, png_colorspacerp colorspace,
       * being in range.  All defined tag types have an 8 byte header - a 4 byte
       * type signature then 0.
       */
-      if ((tag_start & 3) != 0)
-      {
-         /* CNHP730S.icc shipped with Microsoft Windows 64 violates this, it is
-          * only a warning here because libpng does not care about the
-          * alignment.
-          */
-         (void)png_icc_profile_error(png_ptr, NULL, name, tag_id,
-             "ICC profile tag start not a multiple of 4");
-      }

      /* This is a hard error; potentially it can cause read outside the
       * profile.
@@ -2165,6 +2230,16 @@ png_icc_check_tag_table(png_const_structrp png_ptr, png_colorspacerp colorspace,
      if (tag_start > profile_length || tag_length > profile_length - tag_start)
         return png_icc_profile_error(png_ptr, colorspace, name, tag_id,
             "ICC profile tag outside profile");
+
+      if ((tag_start & 3) != 0)
+      {
+         /* CNHP730S.icc shipped with Microsoft Windows 64 violates this; it is
+          * only a warning here because libpng does not care about the
+          * alignment.
+          */
+         (void)png_icc_profile_error(png_ptr, NULL, name, tag_id,
+             "ICC profile tag start not a multiple of 4");
+      }
   }

   return 1; /* success, maybe with warnings */
@@ -2377,7 +2452,7 @@ png_colorspace_set_ICC(png_const_structrp png_ptr, png_colorspacerp colorspace,
   if ((colorspace->flags & PNG_COLORSPACE_INVALID) != 0)
      return 0;

-   if (png_icc_check_length(png_ptr, colorspace, name, profile_length) != 0 &&
+   if (icc_check_length(png_ptr, colorspace, name, profile_length) != 0 &&
       png_icc_check_header(png_ptr, colorspace, name, profile_length, profile,
           color_type) != 0 &&
       png_icc_check_tag_table(png_ptr, colorspace, name, profile_length,
@@ -2495,7 +2570,7 @@ png_check_IHDR(png_const_structrp png_ptr,
      error = 1;
   }

-   if (png_gt(((width + 7) & (~7)),
+   if (png_gt(((width + 7) & (~7U)),
       ((PNG_SIZE_MAX
           - 48        /* big_row_buf hack */
           - 1)        /* filter byte */
@@ -2797,7 +2872,7 @@ png_pow10(int power)
   if (power < 0)
   {
      if (power < DBL_MIN_10_EXP) return 0;
-      recip = 1, power = -power;
+      recip = 1; power = -power;
   }

   if (power > 0)
@@ -2822,6 +2897,14 @@ png_pow10(int power)
 /* Function to format a floating point value in ASCII with a given
 * precision.
 */
+#if GCC_STRICT_OVERFLOW
+#pragma GCC diagnostic push
+/* The problem arises below with exp_b10, which can never overflow because it
+ * comes, originally, from frexp and is therefore limited to a range which is
+ * typically +/-710 (log2(DBL_MAX)/log2(DBL_MIN)).
+ */
+#pragma GCC diagnostic warning "-Wstrict-overflow=2"
+#endif /* GCC_STRICT_OVERFLOW */
 void /* PRIVATE */
 png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
    double fp, unsigned int precision)
@@ -2875,7 +2958,9 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
            double test = png_pow10(exp_b10+1);

            if (test <= DBL_MAX)
-               ++exp_b10, base = test;
+            {
+               ++exp_b10; base = test;
+            }

            else
               break;
@@ -2889,7 +2974,10 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
          * test on DBL_MAX above.
          */
         fp /= base;
-         while (fp >= 1) fp /= 10, ++exp_b10;
+         while (fp >= 1)
+         {
+            fp /= 10; ++exp_b10;
+         }

         /* Because of the code above fp may, at this point, be
          * less than .1, this is ok because the code below can
@@ -2906,7 +2994,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
             */
            if (exp_b10 < 0 && exp_b10 > -3) /* PLUS 3 TOTAL 4 */
            {
-               czero = -exp_b10; /* PLUS 2 digits: TOTAL 3 */
+               czero = 0U-exp_b10; /* PLUS 2 digits: TOTAL 3 */
               exp_b10 = 0;      /* Dot added below before first output. */
            }
            else
@@ -2940,7 +3028,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
                     /* Rounding up to 10, handle that here. */
                     if (czero > 0)
                     {
-                        --czero, d = 1;
+                        --czero; d = 1;
                        if (cdigits == 0) --clead;
                     }
                     else
@@ -2954,7 +3042,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,

                           else if (ch == 46)
                           {
-                              ch = *--ascii, ++size;
+                              ch = *--ascii; ++size;
                              /* Advance exp_b10 to '1', so that the
                               * decimal point happens after the
                               * previous digit.
@@ -2981,7 +3069,9 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
                              int ch = *--ascii;

                              if (ch == 46)
-                                 ++size, exp_b10 = 1;
+                              {
+                                 ++size; exp_b10 = 1;
+                              }

                              /* Else lost a leading zero, so 'exp_b10' is
                               * still ok at (-1)
@@ -3017,21 +3107,26 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
                      */
                     if (exp_b10 != (-1))
                     {
-                        if (exp_b10 == 0) *ascii++ = 46, --size;
+                        if (exp_b10 == 0)
+                        {
+                           *ascii++ = 46; --size;
+                        }
                        /* PLUS 1: TOTAL 4 */
                        --exp_b10;
                     }
-                     *ascii++ = 48, --czero;
+                     *ascii++ = 48; --czero;
                  }

                  if (exp_b10 != (-1))
                  {
                     if (exp_b10 == 0)
-                        *ascii++ = 46, --size; /* counted above */
+                     {
+                        *ascii++ = 46; --size; /* counted above */
+                     }

                     --exp_b10;
                  }
-                  *ascii++ = (char)(48 + (int)d), ++cdigits;
+                  *ascii++ = (char)(48 + (int)d); ++cdigits;
               }
            }
            while (cdigits+czero < precision+clead && fp > DBL_MIN);
@@ -3040,7 +3135,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,

            /* Check for an exponent, if we don't need one we are
             * done and just need to terminate the string.  At
-             * this point exp_b10==(-1) is effectively if flag - it got
+             * this point exp_b10==(-1) is effectively a flag - it got
             * to '-1' because of the decrement after outputting
             * the decimal point above (the exponent required is
             * *not* -1!)
@@ -3054,7 +3149,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
                * zeros were *not* output, so this doesn't increase
                * the output count.
                */
-               while (--exp_b10 >= 0) *ascii++ = 48;
+               while (exp_b10-- > 0) *ascii++ = 48;

               *ascii = 0;

@@ -3072,7 +3167,7 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
             */
            size -= cdigits;

-            *ascii++ = 69, --size;    /* 'E': PLUS 1 TOTAL 2+precision */
+            *ascii++ = 69; --size;    /* 'E': PLUS 1 TOTAL 2+precision */

            /* The following use of an unsigned temporary avoids ambiguities in
             * the signed arithmetic on exp_b10 and permits GCC at least to do
@@ -3083,12 +3178,12 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,

               if (exp_b10 < 0)
               {
-                  *ascii++ = 45, --size; /* '-': PLUS 1 TOTAL 3+precision */
-                  uexp_b10 = -exp_b10;
+                  *ascii++ = 45; --size; /* '-': PLUS 1 TOTAL 3+precision */
+                  uexp_b10 = 0U-exp_b10;
               }

               else
-                  uexp_b10 = exp_b10;
+                  uexp_b10 = 0U+exp_b10;

               cdigits = 0;

@@ -3131,6 +3226,9 @@ png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, png_size_t size,
   /* Here on buffer too small. */
   png_error(png_ptr, "ASCII conversion buffer too small");
 }
+#if GCC_STRICT_OVERFLOW
+#pragma GCC diagnostic pop
+#endif /* GCC_STRICT_OVERFLOW */

 #  endif /* FLOATING_POINT */

@@ -3150,9 +3248,11 @@ png_ascii_from_fixed(png_const_structrp png_ptr, png_charp ascii,

      /* Avoid overflow here on the minimum integer. */
      if (fp < 0)
-         *ascii++ = 45, num = -fp;
+      {
+         *ascii++ = 45; num = (png_uint_32)(-fp);
+      }
      else
-         num = fp;
+         num = (png_uint_32)fp;

      if (num <= 0x80000000) /* else overflowed */
      {
@@ -3188,7 +3288,10 @@ png_ascii_from_fixed(png_const_structrp png_ptr, png_charp ascii,
                * then ndigits digits to first:
                */
               i = 5;
-               while (ndigits < i) *ascii++ = 48, --i;
+               while (ndigits < i)
+               {
+                  *ascii++ = 48; --i;
+               }
               while (ndigits >= first) *ascii++ = digits[--ndigits];
               /* Don't output the trailing zeros! */
            }
@@ -3239,6 +3342,15 @@ png_fixed(png_const_structrp png_ptr, double fp, png_const_charp text)
 * the nearest .00001).  Overflow and divide by zero are signalled in
 * the result, a boolean - true on success, false on overflow.
 */
+#if GCC_STRICT_OVERFLOW /* from above */
+/* It is not obvious which comparison below gets optimized in such a way that
+ * signed overflow would change the result; looking through the code does not
+ * reveal any tests which have the form GCC complains about, so presumably the
+ * optimizer is moving an add or subtract into the 'if' somewhere.
+ */
+#pragma GCC diagnostic push
+#pragma GCC diagnostic warning "-Wstrict-overflow=2"
+#endif /* GCC_STRICT_OVERFLOW */
 int
 png_muldiv(png_fixed_point_p res, png_fixed_point a, png_int_32 times,
    png_int_32 divisor)
@@ -3353,6 +3465,9 @@ png_muldiv(png_fixed_point_p res, png_fixed_point a, png_int_32 times,

   return 0;
 }
+#if GCC_STRICT_OVERFLOW
+#pragma GCC diagnostic pop
+#endif /* GCC_STRICT_OVERFLOW */
 #endif /* READ_GAMMA || INCH_CONVERSIONS */

 #if defined(PNG_READ_GAMMA_SUPPORTED) || defined(PNG_INCH_CONVERSIONS_SUPPORTED)
@@ -3646,7 +3761,7 @@ png_log16bit(png_uint_32 x)
 * of getting this accuracy in practice.
 *
 * To deal with this the following exp() function works out the exponent of the
- * frational part of the logarithm by using an accurate 32-bit value from the
+ * fractional part of the logarithm by using an accurate 32-bit value from the
 * top four fractional bits then multiplying in the remaining bits.
 */
 static const png_uint_32
@@ -4225,13 +4340,13 @@ png_set_option(png_structrp png_ptr, int option, int onoff)
   if (png_ptr != NULL && option >= 0 && option < PNG_OPTION_NEXT &&
      (option & 1) == 0)
   {
-      int mask = 3 << option;
-      int setting = (2 + (onoff != 0)) << option;
-      int current = png_ptr->options;
+      png_uint_32 mask = 3U << option;
+      png_uint_32 setting = (2U + (onoff != 0)) << option;
+      png_uint_32 current = png_ptr->options;

-      png_ptr->options = (png_byte)(((current & ~mask) | setting) & 0xff);
+      png_ptr->options = (png_uint_32)(((current & ~mask) | setting) & 0xff);

-      return (current & mask) >> option;
+      return (int)(current & mask) >> option;
   }

   return PNG_OPTION_INVALID;
@@ -4243,7 +4358,7 @@ png_set_option(png_structrp png_ptr, int option, int onoff)
   defined(PNG_SIMPLIFIED_WRITE_SUPPORTED)
 /* sRGB conversion tables; these are machine generated with the code in
 * contrib/tools/makesRGB.c.  The actual sRGB transfer curve defined in the
- * specification (see the article at http://en.wikipedia.org/wiki/SRGB)
+ * specification (see the article at https://en.wikipedia.org/wiki/SRGB)
 * is used, not the gamma=1/2.2 approximation use elsewhere in libpng.
 * The sRGB to linear table is exact (to the nearest 16-bit linear fraction).
 * The inverse (linear to sRGB) table has accuracies as follows:
@@ -1,9 +1,9 @@

 /* png.h - header file for PNG reference library
 *
- * libpng version 1.6.24, August 4, 2016
+ * libpng version 1.6.34, September 29, 2017
 *
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -12,7 +12,7 @@
 * Authors and maintainers:
 *   libpng versions 0.71, May 1995, through 0.88, January 1996: Guy Schalnat
 *   libpng versions 0.89, June 1996, through 0.96, May 1997: Andreas Dilger
- *   libpng versions 0.97, January 1998, through 1.6.24, August 4, 2016:
+ *   libpng versions 0.97, January 1998, through 1.6.34, September 29, 2017:
 *     Glenn Randers-Pehrson.
 *   See also "Contributing Authors", below.
 */
@@ -25,12 +25,8 @@
 *
 * This code is released under the libpng license.
 *
- * Some files in the "contrib" directory and some configure-generated
- * files that are distributed with libpng have other copyright owners and
- * are released under other open source licenses.
- *
- * libpng versions 1.0.7, July 1, 2000 through 1.6.24, August 4, 2016 are
- * Copyright (c) 2000-2002, 2004, 2006-2016 Glenn Randers-Pehrson, are
+ * libpng versions 1.0.7, July 1, 2000 through 1.6.34, September 29, 2017 are
+ * Copyright (c) 2000-2002, 2004, 2006-2017 Glenn Randers-Pehrson, are
 * derived from libpng-1.0.6, and are distributed according to the same
 * disclaimer and license as libpng-1.0.6 with the following individuals
 * added to the list of Contributing Authors:
@@ -41,6 +37,9 @@
 *    Cosmin Truta
 *    Gilles Vollant
 *    James Yu
+ *    Mandar Sahastrabuddhe
+ *    Google Inc.
+ *    Vadim Barkov
 *
 * and with the following additions to the disclaimer:
 *
@@ -51,10 +50,10 @@
 *    risk of satisfactory quality, performance, accuracy, and effort is with
 *    the user.
 *
- * Some files in the "contrib" directory have other copyright owners and
+ * Some files in the "contrib" directory and some configure-generated
+ * files that are distributed with libpng have other copyright owners and
 * are released under other open source licenses.
 *
- *
 * libpng versions 0.97, January 1998, through 1.0.6, March 20, 2000, are
 * Copyright (c) 1998-2000 Glenn Randers-Pehrson, are derived from
 * libpng-0.96, and are distributed according to the same disclaimer and
@@ -65,9 +64,6 @@
 *    Glenn Randers-Pehrson
 *    Willem van Schaik
 *
- * Some files in the "scripts" directory have different copyright owners
- * but are also released under this license.
- *
 * libpng versions 0.89, June 1996, through 0.96, May 1997, are
 * Copyright (c) 1996-1997 Andreas Dilger, are derived from libpng-0.88,
 * and are distributed according to the same disclaimer and license as
@@ -213,11 +209,11 @@
 *    ...
 *    1.0.19                  10    10019  10.so.0.19[.0]
 *    ...
- *    1.2.56                  13    10256  12.so.0.56[.0]
+ *    1.2.59                  13    10257  12.so.0.59[.0]
 *    ...
- *    1.5.27                  15    10527  15.so.15.27[.0]
+ *    1.5.30                  15    10527  15.so.15.30[.0]
 *    ...
- *    1.6.24                  16    10624  16.so.16.24[.0]
+ *    1.6.34                  16    10633  16.so.16.34[.0]
 *
 *    Henceforth the source version will match the shared-library major
 *    and minor numbers; the shared-library major version number will be
@@ -238,20 +234,20 @@
 *
 * See libpng.txt or libpng.3 for more information.  The PNG specification
 * is available as a W3C Recommendation and as an ISO Specification,
- * <http://www.w3.org/TR/2003/REC-PNG-20031110/
+ * <https://www.w3.org/TR/2003/REC-PNG-20031110/
 */

 /*
 * Y2K compliance in libpng:
 * =========================
 *
- *    August 4, 2016
+ *    September 29, 2017
 *
 *    Since the PNG Development group is an ad-hoc body, we can't make
 *    an official declaration.
 *
 *    This is your unofficial assurance that libpng from version 0.71 and
- *    upward through 1.6.24 are Y2K compliant.  It is my belief that
+ *    upward through 1.6.34 are Y2K compliant.  It is my belief that
 *    earlier versions were also Y2K compliant.
 *
 *    Libpng only has two year fields.  One is a 2-byte unsigned integer
@@ -313,8 +309,8 @@
 */

 /* Version information for png.h - this should match the version in png.c */
-#define PNG_LIBPNG_VER_STRING "1.6.24"
-#define PNG_HEADER_VERSION_STRING " libpng version 1.6.24 - August 4, 2016\n"
+#define PNG_LIBPNG_VER_STRING "1.6.34"
+#define PNG_HEADER_VERSION_STRING " libpng version 1.6.34 - September 29, 2017\n"

 #define PNG_LIBPNG_VER_SONUM   16
 #define PNG_LIBPNG_VER_DLLNUM  16
@@ -322,7 +318,7 @@
 /* These should match the first 3 components of PNG_LIBPNG_VER_STRING: */
 #define PNG_LIBPNG_VER_MAJOR   1
 #define PNG_LIBPNG_VER_MINOR   6
-#define PNG_LIBPNG_VER_RELEASE 24
+#define PNG_LIBPNG_VER_RELEASE 34

 /* This should match the numeric part of the final component of
 * PNG_LIBPNG_VER_STRING, omitting any leading zero:
@@ -353,7 +349,7 @@
 * version 1.0.0 was mis-numbered 100 instead of 10000).  From
 * version 1.0.1 it's    xxyyzz, where x=major, y=minor, z=release
 */
-#define PNG_LIBPNG_VER 10624 /* 1.6.24 */
+#define PNG_LIBPNG_VER 10634 /* 1.6.34 */

 /* Library configuration: these options cannot be changed after
 * the library has been built.
@@ -463,7 +459,7 @@ extern "C" {
 /* This triggers a compiler error in png.c, if png.c and png.h
 * do not agree upon the version number.
 */
-typedef char* png_libpng_version_1_6_24;
+typedef char* png_libpng_version_1_6_34;

 /* Basic control structions.  Read libpng-manual.txt or libpng.3 for more info.
 *
@@ -780,6 +776,7 @@ typedef png_unknown_chunk * * png_unknown_chunkpp;
 #define PNG_INFO_sPLT 0x2000U  /* ESR, 1.0.6 */
 #define PNG_INFO_sCAL 0x4000U  /* ESR, 1.0.6 */
 #define PNG_INFO_IDAT 0x8000U  /* ESR, 1.0.6 */
+#define PNG_INFO_eXIf 0x10000U /* GR-P, 1.6.31 */

 /* This is used for the transformation routines, as some of them
 * change these values for the row.  It also should enable using
@@ -1792,7 +1789,8 @@ PNG_EXPORT(99, void, png_data_freer, (png_const_structrp png_ptr,
 #define PNG_FREE_PLTE 0x1000U
 #define PNG_FREE_TRNS 0x2000U
 #define PNG_FREE_TEXT 0x4000U
-#define PNG_FREE_ALL  0x7fffU
+#define PNG_FREE_EXIF 0x8000U /* Added at libpng-1.6.31 */
+#define PNG_FREE_ALL  0xffffU
 #define PNG_FREE_MUL  0x4220U /* PNG_FREE_SPLT|PNG_FREE_TEXT|PNG_FREE_UNKN */

 #ifdef PNG_USER_MEM_SUPPORTED
@@ -2011,6 +2009,18 @@ PNG_FIXED_EXPORT(233, void, png_set_cHRM_XYZ_fixed, (png_const_structrp png_ptr,
    png_fixed_point int_blue_Z))
 #endif

+#ifdef PNG_eXIf_SUPPORTED
+PNG_EXPORT(246, png_uint_32, png_get_eXIf, (png_const_structrp png_ptr,
+    png_inforp info_ptr, png_bytep *exif));
+PNG_EXPORT(247, void, png_set_eXIf, (png_const_structrp png_ptr,
+    png_inforp info_ptr, const png_bytep exif));
+
+PNG_EXPORT(248, png_uint_32, png_get_eXIf_1, (png_const_structrp png_ptr,
+    png_const_inforp info_ptr, png_uint_32 *num_exif, png_bytep *exif));
+PNG_EXPORT(249, void, png_set_eXIf_1, (png_const_structrp png_ptr,
+    png_inforp info_ptr, const png_uint_32 num_exif, const png_bytep exif));
+#endif
+
 #ifdef PNG_gAMA_SUPPORTED
 PNG_FP_EXPORT(137, png_uint_32, png_get_gAMA, (png_const_structrp png_ptr,
    png_const_inforp info_ptr, double *file_gamma))
@@ -2029,9 +2039,6 @@ PNG_FIXED_EXPORT(140, void, png_set_gAMA_fixed, (png_const_structrp png_ptr,
 #ifdef PNG_hIST_SUPPORTED
 PNG_EXPORT(141, png_uint_32, png_get_hIST, (png_const_structrp png_ptr,
    png_inforp info_ptr, png_uint_16p *hist));
-#endif
-
-#ifdef PNG_hIST_SUPPORTED
 PNG_EXPORT(142, void, png_set_hIST, (png_const_structrp png_ptr,
    png_inforp info_ptr, png_const_uint_16p hist));
 #endif
@@ -2757,7 +2764,7 @@ typedef struct
 *
 * When the simplified API needs to convert between sRGB and linear colorspaces,
 * the actual sRGB transfer curve defined in the sRGB specification (see the
- * article at http://en.wikipedia.org/wiki/SRGB) is used, not the gamma=1/2.2
+ * article at https://en.wikipedia.org/wiki/SRGB) is used, not the gamma=1/2.2
 * approximation used elsewhere in libpng.
 *
 * When an alpha channel is present it is expected to denote pixel coverage
@@ -2812,6 +2819,8 @@ typedef struct
 #  define PNG_FORMAT_FLAG_AFIRST 0x20U /* alpha channel comes first */
 #endif

+#define PNG_FORMAT_FLAG_ASSOCIATED_ALPHA 0x40U /* alpha channel is associated */
+
 /* Commonly used formats have predefined macros.
 *
 * First the single byte (sRGB) formats:
@@ -3226,7 +3235,14 @@ PNG_EXPORT(245, int, png_image_write_to_memory, (png_imagep image, void *memory,
 #endif
 #define PNG_MAXIMUM_INFLATE_WINDOW 2 /* SOFTWARE: force maximum window */
 #define PNG_SKIP_sRGB_CHECK_PROFILE 4 /* SOFTWARE: Check ICC profile for sRGB */
-#define PNG_OPTION_NEXT  6 /* Next option - numbers must be even */
+#ifdef PNG_MIPS_MSA_API_SUPPORTED
+#  define PNG_MIPS_MSA   6 /* HARDWARE: MIPS Msa SIMD instructions supported */
+#endif
+#define PNG_IGNORE_ADLER32 8
+#ifdef PNG_POWERPC_VSX_API_SUPPORTED
+#  define PNG_POWERPC_VSX   10 /* HARDWARE: PowerPC VSX SIMD instructions supported */
+#endif
+#define PNG_OPTION_NEXT  12 /* Next option - numbers must be even */

 /* Return values: NOTE: there are four values and 'off' is *not* zero */
 #define PNG_OPTION_UNSET   0 /* Unset - defaults to off */
@@ -3250,7 +3266,7 @@ PNG_EXPORT(244, int, png_set_option, (png_structrp png_ptr, int option,
 * one to use is one more than this.)
 */
 #ifdef PNG_EXPORT_LAST_ORDINAL
-  PNG_EXPORT_LAST_ORDINAL(245);
+  PNG_EXPORT_LAST_ORDINAL(249);
 #endif

 #ifdef __cplusplus
@@ -1,7 +1,7 @@

 /* pngconf.h - machine configurable file for libpng
 *
- * libpng version 1.6.24, August 4, 2016
+ * libpng version 1.6.34, September 29, 2017
 *
 * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
@@ -507,9 +507,9 @@
 #  error "libpng requires a signed 32-bit (or more) type"
 #endif

-#if UINT_MAX > 4294967294
+#if UINT_MAX > 4294967294U
   typedef unsigned int png_uint_32;
-#elif ULONG_MAX > 4294967294
+#elif ULONG_MAX > 4294967294U
   typedef unsigned long int png_uint_32;
 #else
 #  error "libpng requires an unsigned 32-bit (or more) type"
@@ -1,8 +1,8 @@

 /* pngerror.c - stub functions for i/o and memory allocation
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.31 [July 27, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -163,7 +163,7 @@ png_format_number(png_const_charp start, png_charp end, int format,
         case PNG_NUMBER_FORMAT_02u:
            /* Expects at least 2 digits. */
            mincount = 2;
-            /* FALL THROUGH */
+            /* FALLTHROUGH */

         case PNG_NUMBER_FORMAT_u:
            *--end = digits[number % 10];
@@ -173,7 +173,7 @@ png_format_number(png_const_charp start, png_charp end, int format,
         case PNG_NUMBER_FORMAT_02x:
            /* This format expects at least two digits */
            mincount = 2;
-            /* FALL THROUGH */
+            /* FALLTHROUGH */

         case PNG_NUMBER_FORMAT_x:
            *--end = digits[number & 0xf];
@@ -573,7 +573,7 @@ png_fixed_error,(png_const_structrp png_ptr, png_const_charp name),PNG_NORETURN)
 {
 #  define fixed_message "fixed point overflow in "
 #  define fixed_message_ln ((sizeof fixed_message)-1)
-   int  iin;
+   unsigned int  iin;
   char msg[fixed_message_ln+PNG_MAX_ERROR_TEXT];
   memcpy(msg, fixed_message, fixed_message_ln);
   iin = 0;
@@ -1,8 +1,8 @@

 /* pngget.c - retrieval of values from info struct
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.32 [August 24, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -338,7 +338,7 @@ ppi_from_ppm(png_uint_32 ppm)
   png_fixed_point result;
   if (ppm <= PNG_UINT_31_MAX && png_muldiv(&result, (png_int_32)ppm, 127,
       5000) != 0)
-      return result;
+      return (png_uint_32)result;

   /* Overflow. */
   return 0;
@@ -773,6 +773,35 @@ png_get_sPLT(png_const_structrp png_ptr, png_inforp info_ptr,
 }
 #endif

+#ifdef PNG_eXIf_SUPPORTED
+png_uint_32 PNGAPI
+png_get_eXIf(png_const_structrp png_ptr, png_inforp info_ptr,
+    png_bytep *exif)
+{
+  png_warning(png_ptr, "png_get_eXIf does not work; use png_get_eXIf_1");
+  PNG_UNUSED(info_ptr)
+  PNG_UNUSED(exif)
+  return 0;
+}
+
+png_uint_32 PNGAPI
+png_get_eXIf_1(png_const_structrp png_ptr, png_const_inforp info_ptr,
+    png_uint_32 *num_exif, png_bytep *exif)
+{
+   png_debug1(1, "in %s retrieval function", "eXIf");
+
+   if (png_ptr != NULL && info_ptr != NULL &&
+       (info_ptr->valid & PNG_INFO_eXIf) != 0 && exif != NULL)
+   {
+      *num_exif = info_ptr->num_exif;
+      *exif = info_ptr->exif;
+      return (PNG_INFO_eXIf);
+   }
+
+   return (0);
+}
+#endif
+
 #ifdef PNG_hIST_SUPPORTED
 png_uint_32 PNGAPI
 png_get_hIST(png_const_structrp png_ptr, png_inforp info_ptr,
@@ -185,6 +185,14 @@ defined(PNG_READ_BACKGROUND_SUPPORTED)
   png_byte phys_unit_type; /* resolution type (see PNG_RESOLUTION_ below) */
 #endif

+#ifdef PNG_eXIf_SUPPORTED
+   int num_exif;  /* Added at libpng-1.6.31 */
+   png_bytep exif;
+# ifdef PNG_READ_eXIf_SUPPORTED
+   png_bytep eXIf_buf;  /* Added at libpng-1.6.32 */
+# endif
+#endif
+
 #ifdef PNG_hIST_SUPPORTED
   /* The hIST chunk contains the relative frequency or importance of the
    * various palette entries, so that a viewer can intelligently select a
@@ -1,10 +1,10 @@
-/* libpng 1.6.24 STANDARD API DEFINITION */
+/* libpng 1.6.34 STANDARD API DEFINITION */

 /* pnglibconf.h - library build configuration */

-/* Libpng version 1.6.24 - August 4, 2016 */
+/* Libpng version 1.6.34 - September 29, 2017 */

-/* Copyright (c) 1998-2015 Glenn Randers-Pehrson */
+/* Copyright (c) 1998-2017 Glenn Randers-Pehrson */

 /* This code is released under the libpng license. */
 /* For conditions of distribution and use, see the disclaimer */
@@ -20,6 +20,8 @@
 #define PNG_ALIGNED_MEMORY_SUPPORTED
 /*#undef PNG_ARM_NEON_API_SUPPORTED*/
 /*#undef PNG_ARM_NEON_CHECK_SUPPORTED*/
+/*#undef PNG_POWERPC_VSX_API_SUPPORTED*/
+/*#undef PNG_POWERPC_VSX_CHECK_SUPPORTED*/
 #define PNG_BENIGN_ERRORS_SUPPORTED
 #define PNG_BENIGN_READ_ERRORS_SUPPORTED
 /*#undef PNG_BENIGN_WRITE_ERRORS_SUPPORTED*/
@@ -82,6 +84,7 @@
 #define PNG_READ_USER_TRANSFORM_SUPPORTED
 #define PNG_READ_bKGD_SUPPORTED
 #define PNG_READ_cHRM_SUPPORTED
+#define PNG_READ_eXIf_SUPPORTED
 #define PNG_READ_gAMA_SUPPORTED
 #define PNG_READ_hIST_SUPPORTED
 #define PNG_READ_iCCP_SUPPORTED
@@ -151,6 +154,7 @@
 #define PNG_WRITE_WEIGHTED_FILTER_SUPPORTED
 #define PNG_WRITE_bKGD_SUPPORTED
 #define PNG_WRITE_cHRM_SUPPORTED
+#define PNG_WRITE_eXIf_SUPPORTED
 #define PNG_WRITE_gAMA_SUPPORTED
 #define PNG_WRITE_hIST_SUPPORTED
 #define PNG_WRITE_iCCP_SUPPORTED
@@ -168,6 +172,7 @@
 #define PNG_WRITE_zTXt_SUPPORTED
 #define PNG_bKGD_SUPPORTED
 #define PNG_cHRM_SUPPORTED
+#define PNG_eXIf_SUPPORTED
 #define PNG_gAMA_SUPPORTED
 #define PNG_hIST_SUPPORTED
 #define PNG_iCCP_SUPPORTED
@@ -1,7 +1,7 @@

 /* pngmem.c - stub functions for memory allocation
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
+ * Last changed in libpng 1.6.26 [October 20, 2016]
 * Copyright (c) 1998-2002,2004,2006-2014,2016 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
@@ -109,7 +109,7 @@ static png_voidp
 png_malloc_array_checked(png_const_structrp png_ptr, int nelements,
    size_t element_size)
 {
-   png_alloc_size_t req = nelements; /* known to be > 0 */
+   png_alloc_size_t req = (png_alloc_size_t)nelements; /* known to be > 0 */

   if (req <= PNG_SIZE_MAX/element_size)
      return png_malloc_base(png_ptr, req * element_size);
@@ -1,8 +1,8 @@

 /* pngpread.c - read a png file in push mode
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.32 [August 24, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -189,6 +189,7 @@ png_push_read_chunk(png_structrp png_ptr, png_inforp info_ptr)
      png_crc_read(png_ptr, chunk_tag, 4);
      png_ptr->chunk_name = PNG_CHUNK_FROM_STRING(chunk_tag);
      png_check_chunk_name(png_ptr, png_ptr->chunk_name);
+      png_check_chunk_length(png_ptr, png_ptr->push_length);
      png_ptr->mode |= PNG_HAVE_CHUNK_HEADER;
   }

@@ -684,7 +685,12 @@ png_process_IDAT_data(png_structrp png_ptr, png_bytep buffer,
            png_warning(png_ptr, "Truncated compressed data in IDAT");

         else
-            png_error(png_ptr, "Decompression error in IDAT");
+         {
+            if (ret == Z_DATA_ERROR)
+               png_benign_error(png_ptr, "IDAT: ADLER32 checksum mismatch");
+            else
+               png_error(png_ptr, "Decompression error in IDAT");
+         }

         /* Skip the check on unprocessed input */
         return;
@@ -1,8 +1,8 @@

 /* pngpriv.h - private declarations for use inside libpng
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.32 [August 24, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -35,7 +35,9 @@
 * Windows/Visual Studio) there is no effect; the OS specific tests below are
 * still required (as of 2011-05-02.)
 */
-#define _POSIX_SOURCE 1 /* Just the POSIX 1003.1 and C89 APIs */
+#ifndef _POSIX_SOURCE
+# define _POSIX_SOURCE 1 /* Just the POSIX 1003.1 and C89 APIs */
+#endif

 #ifndef PNG_VERSION_INFO_ONLY
 /* Standard library headers not required by png.h: */
@@ -182,6 +184,22 @@
 #  endif
 #endif /* PNG_ARM_NEON_OPT > 0 */

+#ifndef PNG_MIPS_MSA_OPT
+#  if defined(__mips_msa) && (__mips_isa_rev >= 5) && defined(PNG_ALIGNED_MEMORY_SUPPORTED)
+#     define PNG_MIPS_MSA_OPT 2
+#  else
+#     define PNG_MIPS_MSA_OPT 0
+#  endif
+#endif
+
+#ifndef PNG_POWERPC_VSX_OPT
+#  if defined(__PPC64__) && defined(__ALTIVEC__) && defined(__VSX__)
+#     define PNG_POWERPC_VSX_OPT 2
+#  else
+#     define PNG_POWERPC_VSX_OPT 0
+#  endif
+#endif
+
 #ifndef PNG_INTEL_SSE_OPT
 #   ifdef PNG_INTEL_SSE
      /* Only check for SSE if the build configuration has been modified to
@@ -218,6 +236,32 @@
 #   endif
 #endif

+#if PNG_MIPS_MSA_OPT > 0
+#  define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_msa
+#  ifndef PNG_MIPS_MSA_IMPLEMENTATION
+#     if defined(__mips_msa)
+#        if defined(__clang__)
+#        elif defined(__GNUC__)
+#           if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 7)
+#              define PNG_MIPS_MSA_IMPLEMENTATION 2
+#           endif /* no GNUC support */
+#        endif /* __GNUC__ */
+#     else /* !defined __mips_msa */
+#        define PNG_MIPS_MSA_IMPLEMENTATION 2
+#     endif /* __mips_msa */
+#  endif /* !PNG_MIPS_MSA_IMPLEMENTATION */
+
+#  ifndef PNG_MIPS_MSA_IMPLEMENTATION
+#     define PNG_MIPS_MSA_IMPLEMENTATION 1
+#  endif
+#endif /* PNG_MIPS_MSA_OPT > 0 */
+
+#if PNG_POWERPC_VSX_OPT > 0
+#  define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_vsx
+#  define PNG_POWERPC_VSX_IMPLEMENTATION 1
+#endif
+
+
 /* Is this a build of a DLL where compilation of the object modules requires
 * different preprocessor settings to those required for a simple library?  If
 * so PNG_BUILD_DLL must be set.
@@ -410,25 +454,6 @@
 #  define png_fixed_error(s1,s2) png_err(s1)
 #endif

-/* C allows up-casts from (void*) to any pointer and (const void*) to any
- * pointer to a const object.  C++ regards this as a type error and requires an
- * explicit, static, cast and provides the static_cast<> rune to ensure that
- * const is not cast away.
- */
-#ifdef __cplusplus
-#  define png_voidcast(type, value) static_cast<type>(value)
-#  define png_constcast(type, value) const_cast<type>(value)
-#  define png_aligncast(type, value) \
-   static_cast<type>(static_cast<void*>(value))
-#  define png_aligncastconst(type, value) \
-   static_cast<type>(static_cast<const void*>(value))
-#else
-#  define png_voidcast(type, value) (value)
-#  define png_constcast(type, value) ((type)(value))
-#  define png_aligncast(type, value) ((void*)(value))
-#  define png_aligncastconst(type, value) ((const void*)(value))
-#endif /* __cplusplus */
-
 /* Some fixed point APIs are still required even if not exported because
 * they get used by the corresponding floating point APIs.  This magic
 * deals with this:
@@ -443,6 +468,35 @@
 /* Other defines specific to compilers can go here.  Try to keep
 * them inside an appropriate ifdef/endif pair for portability.
 */
+
+/* C allows up-casts from (void*) to any pointer and (const void*) to any
+ * pointer to a const object.  C++ regards this as a type error and requires an
+ * explicit, static, cast and provides the static_cast<> rune to ensure that
+ * const is not cast away.
+ */
+#ifdef __cplusplus
+#  define png_voidcast(type, value) static_cast<type>(value)
+#  define png_constcast(type, value) const_cast<type>(value)
+#  define png_aligncast(type, value) \
+   static_cast<type>(static_cast<void*>(value))
+#  define png_aligncastconst(type, value) \
+   static_cast<type>(static_cast<const void*>(value))
+#else
+#  define png_voidcast(type, value) (value)
+#  ifdef _WIN64
+#     ifdef __GNUC__
+         typedef unsigned long long png_ptruint;
+#     else
+         typedef unsigned __int64 png_ptruint;
+#     endif
+#  else
+      typedef unsigned long png_ptruint;
+#  endif
+#  define png_constcast(type, value) ((type)(png_ptruint)(const void*)(value))
+#  define png_aligncast(type, value) ((void*)(value))
+#  define png_aligncastconst(type, value) ((const void*)(value))
+#endif /* __cplusplus */
+
 #if defined(PNG_FLOATING_POINT_SUPPORTED) ||\
    defined(PNG_FLOATING_ARITHMETIC_SUPPORTED)
   /* png.c requires the following ANSI-C constants if the conversion of
@@ -540,7 +594,8 @@
 /* This implicitly assumes alignment is always to a power of 2. */
 #ifdef png_alignof
 #  define png_isaligned(ptr, type)\
-   ((((const char*)ptr-(const char*)0) & (png_alignof(type)-1)) == 0)
+   (((type)((const char*)ptr-(const char*)0) & \
+   (type)(png_alignof(type)-1)) == 0)
 #else
 #  define png_isaligned(ptr, type) 0
 #endif
@@ -557,92 +612,92 @@
 * are defined in png.h because they need to be visible to applications
 * that call png_set_unknown_chunk().
 */
-/* #define PNG_HAVE_IHDR            0x01 (defined in png.h) */
-/* #define PNG_HAVE_PLTE            0x02 (defined in png.h) */
-#define PNG_HAVE_IDAT               0x04
-/* #define PNG_AFTER_IDAT           0x08 (defined in png.h) */
-#define PNG_HAVE_IEND               0x10
-                   /*               0x20 (unused) */
-                   /*               0x40 (unused) */
-                   /*               0x80 (unused) */
-#define PNG_HAVE_CHUNK_HEADER      0x100
-#define PNG_WROTE_tIME             0x200
-#define PNG_WROTE_INFO_BEFORE_PLTE 0x400
-#define PNG_BACKGROUND_IS_GRAY     0x800
-#define PNG_HAVE_PNG_SIGNATURE    0x1000
-#define PNG_HAVE_CHUNK_AFTER_IDAT 0x2000 /* Have another chunk after IDAT */
-                   /*             0x4000 (unused) */
-#define PNG_IS_READ_STRUCT        0x8000 /* Else is a write struct */
+/* #define PNG_HAVE_IHDR            0x01U (defined in png.h) */
+/* #define PNG_HAVE_PLTE            0x02U (defined in png.h) */
+#define PNG_HAVE_IDAT               0x04U
+/* #define PNG_AFTER_IDAT           0x08U (defined in png.h) */
+#define PNG_HAVE_IEND               0x10U
+                   /*               0x20U (unused) */
+                   /*               0x40U (unused) */
+                   /*               0x80U (unused) */
+#define PNG_HAVE_CHUNK_HEADER      0x100U
+#define PNG_WROTE_tIME             0x200U
+#define PNG_WROTE_INFO_BEFORE_PLTE 0x400U
+#define PNG_BACKGROUND_IS_GRAY     0x800U
+#define PNG_HAVE_PNG_SIGNATURE    0x1000U
+#define PNG_HAVE_CHUNK_AFTER_IDAT 0x2000U /* Have another chunk after IDAT */
+                   /*             0x4000U (unused) */
+#define PNG_IS_READ_STRUCT        0x8000U /* Else is a write struct */

 /* Flags for the transformations the PNG library does on the image data */
-#define PNG_BGR                 0x0001
-#define PNG_INTERLACE           0x0002
-#define PNG_PACK                0x0004
-#define PNG_SHIFT               0x0008
-#define PNG_SWAP_BYTES          0x0010
-#define PNG_INVERT_MONO         0x0020
-#define PNG_QUANTIZE            0x0040
-#define PNG_COMPOSE             0x0080     /* Was PNG_BACKGROUND */
-#define PNG_BACKGROUND_EXPAND   0x0100
-#define PNG_EXPAND_16           0x0200     /* Added to libpng 1.5.2 */
-#define PNG_16_TO_8             0x0400     /* Becomes 'chop' in 1.5.4 */
-#define PNG_RGBA                0x0800
-#define PNG_EXPAND              0x1000
-#define PNG_GAMMA               0x2000
-#define PNG_GRAY_TO_RGB         0x4000
-#define PNG_FILLER              0x8000
-#define PNG_PACKSWAP           0x10000
-#define PNG_SWAP_ALPHA         0x20000
-#define PNG_STRIP_ALPHA        0x40000
-#define PNG_INVERT_ALPHA       0x80000
-#define PNG_USER_TRANSFORM    0x100000
-#define PNG_RGB_TO_GRAY_ERR   0x200000
-#define PNG_RGB_TO_GRAY_WARN  0x400000
-#define PNG_RGB_TO_GRAY       0x600000 /* two bits, RGB_TO_GRAY_ERR|WARN */
-#define PNG_ENCODE_ALPHA      0x800000 /* Added to libpng-1.5.4 */
-#define PNG_ADD_ALPHA        0x1000000 /* Added to libpng-1.2.7 */
-#define PNG_EXPAND_tRNS      0x2000000 /* Added to libpng-1.2.9 */
-#define PNG_SCALE_16_TO_8    0x4000000 /* Added to libpng-1.5.4 */
-                       /*    0x8000000 unused */
-                       /*   0x10000000 unused */
-                       /*   0x20000000 unused */
-                       /*   0x40000000 unused */
+#define PNG_BGR                 0x0001U
+#define PNG_INTERLACE           0x0002U
+#define PNG_PACK                0x0004U
+#define PNG_SHIFT               0x0008U
+#define PNG_SWAP_BYTES          0x0010U
+#define PNG_INVERT_MONO         0x0020U
+#define PNG_QUANTIZE            0x0040U
+#define PNG_COMPOSE             0x0080U    /* Was PNG_BACKGROUND */
+#define PNG_BACKGROUND_EXPAND   0x0100U
+#define PNG_EXPAND_16           0x0200U    /* Added to libpng 1.5.2 */
+#define PNG_16_TO_8             0x0400U    /* Becomes 'chop' in 1.5.4 */
+#define PNG_RGBA                0x0800U
+#define PNG_EXPAND              0x1000U
+#define PNG_GAMMA               0x2000U
+#define PNG_GRAY_TO_RGB         0x4000U
+#define PNG_FILLER              0x8000U
+#define PNG_PACKSWAP           0x10000U
+#define PNG_SWAP_ALPHA         0x20000U
+#define PNG_STRIP_ALPHA        0x40000U
+#define PNG_INVERT_ALPHA       0x80000U
+#define PNG_USER_TRANSFORM    0x100000U
+#define PNG_RGB_TO_GRAY_ERR   0x200000U
+#define PNG_RGB_TO_GRAY_WARN  0x400000U
+#define PNG_RGB_TO_GRAY       0x600000U /* two bits, RGB_TO_GRAY_ERR|WARN */
+#define PNG_ENCODE_ALPHA      0x800000U /* Added to libpng-1.5.4 */
+#define PNG_ADD_ALPHA        0x1000000U /* Added to libpng-1.2.7 */
+#define PNG_EXPAND_tRNS      0x2000000U /* Added to libpng-1.2.9 */
+#define PNG_SCALE_16_TO_8    0x4000000U /* Added to libpng-1.5.4 */
+                       /*    0x8000000U unused */
+                       /*   0x10000000U unused */
+                       /*   0x20000000U unused */
+                       /*   0x40000000U unused */
 /* Flags for png_create_struct */
-#define PNG_STRUCT_PNG   0x0001
-#define PNG_STRUCT_INFO  0x0002
+#define PNG_STRUCT_PNG   0x0001U
+#define PNG_STRUCT_INFO  0x0002U

 /* Flags for the png_ptr->flags rather than declaring a byte for each one */
-#define PNG_FLAG_ZLIB_CUSTOM_STRATEGY     0x0001
-#define PNG_FLAG_ZSTREAM_INITIALIZED      0x0002 /* Added to libpng-1.6.0 */
-                                  /*      0x0004    unused */
-#define PNG_FLAG_ZSTREAM_ENDED            0x0008 /* Added to libpng-1.6.0 */
-                                  /*      0x0010    unused */
-                                  /*      0x0020    unused */
-#define PNG_FLAG_ROW_INIT                 0x0040
-#define PNG_FLAG_FILLER_AFTER             0x0080
-#define PNG_FLAG_CRC_ANCILLARY_USE        0x0100
-#define PNG_FLAG_CRC_ANCILLARY_NOWARN     0x0200
-#define PNG_FLAG_CRC_CRITICAL_USE         0x0400
-#define PNG_FLAG_CRC_CRITICAL_IGNORE      0x0800
-#define PNG_FLAG_ASSUME_sRGB              0x1000 /* Added to libpng-1.5.4 */
-#define PNG_FLAG_OPTIMIZE_ALPHA           0x2000 /* Added to libpng-1.5.4 */
-#define PNG_FLAG_DETECT_UNINITIALIZED     0x4000 /* Added to libpng-1.5.4 */
-/* #define PNG_FLAG_KEEP_UNKNOWN_CHUNKS      0x8000 */
-/* #define PNG_FLAG_KEEP_UNSAFE_CHUNKS      0x10000 */
-#define PNG_FLAG_LIBRARY_MISMATCH        0x20000
-#define PNG_FLAG_STRIP_ERROR_NUMBERS     0x40000
-#define PNG_FLAG_STRIP_ERROR_TEXT        0x80000
-#define PNG_FLAG_BENIGN_ERRORS_WARN     0x100000 /* Added to libpng-1.4.0 */
-#define PNG_FLAG_APP_WARNINGS_WARN      0x200000 /* Added to libpng-1.6.0 */
-#define PNG_FLAG_APP_ERRORS_WARN        0x400000 /* Added to libpng-1.6.0 */
-                                  /*    0x800000    unused */
-                                  /*   0x1000000    unused */
-                                  /*   0x2000000    unused */
-                                  /*   0x4000000    unused */
-                                  /*   0x8000000    unused */
-                                  /*  0x10000000    unused */
-                                  /*  0x20000000    unused */
-                                  /*  0x40000000    unused */
+#define PNG_FLAG_ZLIB_CUSTOM_STRATEGY     0x0001U
+#define PNG_FLAG_ZSTREAM_INITIALIZED      0x0002U /* Added to libpng-1.6.0 */
+                                  /*      0x0004U    unused */
+#define PNG_FLAG_ZSTREAM_ENDED            0x0008U /* Added to libpng-1.6.0 */
+                                  /*      0x0010U    unused */
+                                  /*      0x0020U    unused */
+#define PNG_FLAG_ROW_INIT                 0x0040U
+#define PNG_FLAG_FILLER_AFTER             0x0080U
+#define PNG_FLAG_CRC_ANCILLARY_USE        0x0100U
+#define PNG_FLAG_CRC_ANCILLARY_NOWARN     0x0200U
+#define PNG_FLAG_CRC_CRITICAL_USE         0x0400U
+#define PNG_FLAG_CRC_CRITICAL_IGNORE      0x0800U
+#define PNG_FLAG_ASSUME_sRGB              0x1000U /* Added to libpng-1.5.4 */
+#define PNG_FLAG_OPTIMIZE_ALPHA           0x2000U /* Added to libpng-1.5.4 */
+#define PNG_FLAG_DETECT_UNINITIALIZED     0x4000U /* Added to libpng-1.5.4 */
+/* #define PNG_FLAG_KEEP_UNKNOWN_CHUNKS      0x8000U */
+/* #define PNG_FLAG_KEEP_UNSAFE_CHUNKS      0x10000U */
+#define PNG_FLAG_LIBRARY_MISMATCH        0x20000U
+#define PNG_FLAG_STRIP_ERROR_NUMBERS     0x40000U
+#define PNG_FLAG_STRIP_ERROR_TEXT        0x80000U
+#define PNG_FLAG_BENIGN_ERRORS_WARN     0x100000U /* Added to libpng-1.4.0 */
+#define PNG_FLAG_APP_WARNINGS_WARN      0x200000U /* Added to libpng-1.6.0 */
+#define PNG_FLAG_APP_ERRORS_WARN        0x400000U /* Added to libpng-1.6.0 */
+                                  /*    0x800000U    unused */
+                                  /*   0x1000000U    unused */
+                                  /*   0x2000000U    unused */
+                                  /*   0x4000000U    unused */
+                                  /*   0x8000000U    unused */
+                                  /*  0x10000000U    unused */
+                                  /*  0x20000000U    unused */
+                                  /*  0x40000000U    unused */

 #define PNG_FLAG_CRC_ANCILLARY_MASK (PNG_FLAG_CRC_ANCILLARY_USE | \
                                     PNG_FLAG_CRC_ANCILLARY_NOWARN)
@@ -676,6 +731,24 @@
    ((png_size_t)(width) * (((png_size_t)(pixel_bits)) >> 3)) : \
    (( ((png_size_t)(width) * ((png_size_t)(pixel_bits))) + 7) >> 3) )

+/* This returns the number of trailing bits in the last byte of a row, 0 if the
+ * last byte is completely full of pixels.  It is, in principle, (pixel_bits x
+ * width) % 8, but that would overflow for large 'width'.  The second macro is
+ * the same except that it returns the number of unused bits in the last byte;
+ * (8-TRAILBITS), but 0 when TRAILBITS is 0.
+ *
+ * NOTE: these macros are intended to be self-evidently correct and never
+ * overflow on the assumption that pixel_bits is in the range 0..255.  The
+ * arguments are evaluated only once and they can be signed (e.g. as a result of
+ * the integral promotions).  The result of the expression always has type
+ * (png_uint_32), however the compiler always knows it is in the range 0..7.
+ */
+#define PNG_TRAILBITS(pixel_bits, width) \
+    (((pixel_bits) * ((width) % (png_uint_32)8)) % 8)
+
+#define PNG_PADBITS(pixel_bits, width) \
+    ((8 - PNG_TRAILBITS(pixel_bits, width)) % 8)
+
 /* PNG_OUT_OF_RANGE returns true if value is outside the range
 * ideal-delta..ideal+delta.  Each argument is evaluated twice.
 * "ideal" and "delta" should be constants, normally simple
@@ -769,6 +842,7 @@
 #define png_PLTE PNG_U32( 80,  76,  84,  69)
 #define png_bKGD PNG_U32( 98,  75,  71,  68)
 #define png_cHRM PNG_U32( 99,  72,  82,  77)
+#define png_eXIf PNG_U32(101,  88,  73, 102) /* registered July 2017 */
 #define png_fRAc PNG_U32(102,  82,  65,  99) /* registered, not defined */
 #define png_gAMA PNG_U32(103,  65,  77,  65)
 #define png_gIFg PNG_U32(103,  73,  70, 103)
@@ -1069,6 +1143,11 @@ PNG_INTERNAL_FUNCTION(void,png_write_sRGB,(png_structrp png_ptr,
    int intent),PNG_EMPTY);
 #endif

+#ifdef PNG_WRITE_eXIf_SUPPORTED
+PNG_INTERNAL_FUNCTION(void,png_write_eXIf,(png_structrp png_ptr,
+    png_bytep exif, int num_exif),PNG_EMPTY);
+#endif
+
 #ifdef PNG_WRITE_iCCP_SUPPORTED
 PNG_INTERNAL_FUNCTION(void,png_write_iCCP,(png_structrp png_ptr,
   png_const_charp name, png_const_bytep profile), PNG_EMPTY);
@@ -1226,7 +1305,41 @@ PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_neon,(png_row_infop
 PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_neon,(png_row_infop
    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
 #endif
- 
+
+#if PNG_MIPS_MSA_OPT > 0
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_msa,(png_row_infop row_info,
+    png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_msa,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_msa,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_msa,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_msa,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_msa,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_msa,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+#endif
+
+#if PNG_POWERPC_VSX_OPT > 0
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_vsx,(png_row_infop row_info,
+    png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_vsx,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_vsx,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_vsx,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_vsx,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_vsx,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_vsx,(png_row_infop
+    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
+#endif
+
 #if defined(PNG_INTEL_SSE_IMPLEMENTATION) && PNG_INTEL_SSE_IMPLEMENTATION > 0
 PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_sse2,(png_row_infop
    row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY);
@@ -1268,7 +1381,7 @@ PNG_INTERNAL_FUNCTION(void,png_read_finish_row,(png_structrp png_ptr),
 /* Initialize the row buffers, etc. */
 PNG_INTERNAL_FUNCTION(void,png_read_start_row,(png_structrp png_ptr),PNG_EMPTY);

-#if PNG_ZLIB_VERNUM >= 0x1240
+#if ZLIB_VERNUM >= 0x1240
 PNG_INTERNAL_FUNCTION(int,png_zlib_inflate,(png_structrp png_ptr, int flush),
      PNG_EMPTY);
 #  define PNG_INFLATE(pp, flush) png_zlib_inflate(pp, flush)
@@ -1334,6 +1447,11 @@ PNG_INTERNAL_FUNCTION(void,png_handle_cHRM,(png_structrp png_ptr,
    png_inforp info_ptr, png_uint_32 length),PNG_EMPTY);
 #endif

+#ifdef PNG_READ_eXIf_SUPPORTED
+PNG_INTERNAL_FUNCTION(void,png_handle_eXIf,(png_structrp png_ptr,
+    png_inforp info_ptr, png_uint_32 length),PNG_EMPTY);
+#endif
+
 #ifdef PNG_READ_gAMA_SUPPORTED
 PNG_INTERNAL_FUNCTION(void,png_handle_gAMA,(png_structrp png_ptr,
    png_inforp info_ptr, png_uint_32 length),PNG_EMPTY);
@@ -1409,8 +1527,11 @@ PNG_INTERNAL_FUNCTION(void,png_handle_zTXt,(png_structrp png_ptr,
    png_inforp info_ptr, png_uint_32 length),PNG_EMPTY);
 #endif

-PNG_INTERNAL_FUNCTION(void,png_check_chunk_name,(png_structrp png_ptr,
-    png_uint_32 chunk_name),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_check_chunk_name,(png_const_structrp png_ptr,
+    const png_uint_32 chunk_name),PNG_EMPTY);
+
+PNG_INTERNAL_FUNCTION(void,png_check_chunk_length,(png_const_structrp png_ptr,
+    const png_uint_32 chunk_length),PNG_EMPTY);

 PNG_INTERNAL_FUNCTION(void,png_handle_unknown,(png_structrp png_ptr,
    png_inforp info_ptr, png_uint_32 length, int keep),PNG_EMPTY);
@@ -1545,9 +1666,11 @@ PNG_INTERNAL_FUNCTION(int,png_colorspace_set_ICC,(png_const_structrp png_ptr,
   /* The 'name' is used for information only */

 /* Routines for checking parts of an ICC profile. */
+#ifdef PNG_READ_iCCP_SUPPORTED
 PNG_INTERNAL_FUNCTION(int,png_icc_check_length,(png_const_structrp png_ptr,
   png_colorspacerp colorspace, png_const_charp name,
   png_uint_32 profile_length), PNG_EMPTY);
+#endif /* READ_iCCP */
 PNG_INTERNAL_FUNCTION(int,png_icc_check_header,(png_const_structrp png_ptr,
   png_colorspacerp colorspace, png_const_charp name,
   png_uint_32 profile_length,
@@ -1969,7 +2092,13 @@ PNG_INTERNAL_FUNCTION(void, PNG_FILTER_OPTIMIZATIONS, (png_structp png_ptr,
 #  if PNG_ARM_NEON_OPT > 0
 PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_neon,
   (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
-#  endif
+#endif
+
+#if PNG_MIPS_MSA_OPT > 0
+PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_msa,
+   (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
+#endif
+
 #  if defined(PNG_INTEL_SSE_IMPLEMENTATION) && PNG_INTEL_SSE_IMPLEMENTATION > 0
 PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_sse2,
   (png_structp png_ptr, unsigned int bpp), PNG_EMPTY);
@@ -1,8 +1,8 @@

 /* pngread.c - read a PNG file
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.33 [September 28, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -175,6 +175,11 @@ png_read_info(png_structrp png_ptr, png_inforp info_ptr)
         png_handle_cHRM(png_ptr, info_ptr, length);
 #endif

+#ifdef PNG_READ_eXIf_SUPPORTED
+      else if (chunk_name == png_eXIf)
+         png_handle_eXIf(png_ptr, info_ptr, length);
+#endif
+
 #ifdef PNG_READ_gAMA_SUPPORTED
      else if (chunk_name == png_gAMA)
         png_handle_gAMA(png_ptr, info_ptr, length);
@@ -359,9 +364,9 @@ png_do_read_intrapixel(png_row_infop row_info, png_bytep row)

         for (i = 0, rp = row; i < row_width; i++, rp += bytes_per_pixel)
         {
-            png_uint_32 s0   = (*(rp    ) << 8) | *(rp + 1);
-            png_uint_32 s1   = (*(rp + 2) << 8) | *(rp + 3);
-            png_uint_32 s2   = (*(rp + 4) << 8) | *(rp + 5);
+            png_uint_32 s0   = (png_uint_32)(*(rp    ) << 8) | *(rp + 1);
+            png_uint_32 s1   = (png_uint_32)(*(rp + 2) << 8) | *(rp + 3);
+            png_uint_32 s2   = (png_uint_32)(*(rp + 4) << 8) | *(rp + 5);
            png_uint_32 red  = (s0 + s1 + 65536) & 0xffff;
            png_uint_32 blue = (s2 + s1 + 65536) & 0xffff;
            *(rp    ) = (png_byte)((red >> 8) & 0xff);
@@ -534,6 +539,7 @@ png_read_row(png_structrp png_ptr, png_bytep row, png_bytep dsp_row)
      png_error(png_ptr, "Invalid attempt to read row data");

   /* Fill the row with IDAT data: */
+   png_ptr->row_buf[0]=255; /* to force error if no data was found */
   png_read_IDAT_data(png_ptr, png_ptr->row_buf, row_info.rowbytes + 1);

   if (png_ptr->row_buf[0] > PNG_FILTER_VALUE_NONE)
@@ -842,6 +848,11 @@ png_read_end(png_structrp png_ptr, png_inforp info_ptr)
         png_handle_cHRM(png_ptr, info_ptr, length);
 #endif

+#ifdef PNG_READ_eXIf_SUPPORTED
+      else if (chunk_name == png_eXIf)
+         png_handle_eXIf(png_ptr, info_ptr, length);
+#endif
+
 #ifdef PNG_READ_gAMA_SUPPORTED
      else if (chunk_name == png_gAMA)
         png_handle_gAMA(png_ptr, info_ptr, length);
@@ -1393,7 +1404,9 @@ png_image_read_header(png_voidp argument)
   png_structrp png_ptr = image->opaque->png_ptr;
   png_inforp info_ptr = image->opaque->info_ptr;

+#ifdef PNG_BENIGN_ERRORS_SUPPORTED
   png_set_benign_errors(png_ptr, 1/*warn*/);
+#endif
   png_read_info(png_ptr, info_ptr);

   /* Do this the fast way; just read directly out of png_struct. */
@@ -1431,7 +1444,7 @@ png_image_read_header(png_voidp argument)
            break;

         case PNG_COLOR_TYPE_PALETTE:
-            cmap_entries = png_ptr->num_palette;
+            cmap_entries = (png_uint_32)png_ptr->num_palette;
            break;

         default:
@@ -1881,7 +1894,7 @@ png_create_colormap_entry(png_image_read_control *display,
         {
            case 4:
               entry[afirst ? 0 : 3] = (png_uint_16)alpha;
-               /* FALL THROUGH */
+               /* FALLTHROUGH */

            case 3:
               if (alpha < 65535)
@@ -1903,7 +1916,7 @@ png_create_colormap_entry(png_image_read_control *display,

            case 2:
               entry[1 ^ afirst] = (png_uint_16)alpha;
-               /* FALL THROUGH */
+               /* FALLTHROUGH */

            case 1:
               if (alpha < 65535)
@@ -1932,6 +1945,7 @@ png_create_colormap_entry(png_image_read_control *display,
         {
            case 4:
               entry[afirst ? 0 : 3] = (png_byte)alpha;
+               /* FALLTHROUGH */
            case 3:
               entry[afirst + (2 ^ bgr)] = (png_byte)blue;
               entry[afirst + 1] = (png_byte)green;
@@ -1940,6 +1954,7 @@ png_create_colormap_entry(png_image_read_control *display,

            case 2:
               entry[1 ^ afirst] = (png_byte)alpha;
+               /* FALLTHROUGH */
            case 1:
               entry[afirst] = (png_byte)green;
               break;
@@ -1966,7 +1981,7 @@ make_gray_file_colormap(png_image_read_control *display)
   for (i=0; i<256; ++i)
      png_create_colormap_entry(display, i, i, i, i, 255, P_FILE);

-   return i;
+   return (int)i;
 }

 static int
@@ -1977,7 +1992,7 @@ make_gray_colormap(png_image_read_control *display)
   for (i=0; i<256; ++i)
      png_create_colormap_entry(display, i, i, i, i, 255, P_sRGB);

-   return i;
+   return (int)i;
 }
 #define PNG_GRAY_COLORMAP_ENTRIES 256

@@ -2031,7 +2046,7 @@ make_ga_colormap(png_image_read_control *display)
             P_sRGB);
   }

-   return i;
+   return (int)i;
 }

 #define PNG_GA_COLORMAP_ENTRIES 256
@@ -2056,7 +2071,7 @@ make_rgb_colormap(png_image_read_control *display)
      }
   }

-   return i;
+   return (int)i;
 }

 #define PNG_RGB_COLORMAP_ENTRIES 216
@@ -2249,7 +2264,7 @@ png_image_read_colormap(png_voidp argument)
            if (PNG_GRAY_COLORMAP_ENTRIES > image->colormap_entries)
               png_error(png_ptr, "gray[16] color-map: too few entries");

-            cmap_entries = make_gray_colormap(display);
+            cmap_entries = (unsigned int)make_gray_colormap(display);

            if (png_ptr->num_trans > 0)
            {
@@ -2347,7 +2362,7 @@ png_image_read_colormap(png_voidp argument)
            if (PNG_GA_COLORMAP_ENTRIES > image->colormap_entries)
               png_error(png_ptr, "gray+alpha color-map: too few entries");

-            cmap_entries = make_ga_colormap(display);
+            cmap_entries = (unsigned int)make_ga_colormap(display);

            background_index = PNG_CMAP_GA_BACKGROUND;
            output_processing = PNG_CMAP_GA;
@@ -2381,7 +2396,7 @@ png_image_read_colormap(png_voidp argument)
               if (PNG_GRAY_COLORMAP_ENTRIES > image->colormap_entries)
                  png_error(png_ptr, "gray-alpha color-map: too few entries");

-               cmap_entries = make_gray_colormap(display);
+               cmap_entries = (unsigned int)make_gray_colormap(display);

               if (output_encoding == P_LINEAR)
               {
@@ -2520,7 +2535,7 @@ png_image_read_colormap(png_voidp argument)
               if (PNG_GA_COLORMAP_ENTRIES > image->colormap_entries)
                  png_error(png_ptr, "rgb[ga] color-map: too few entries");

-               cmap_entries = make_ga_colormap(display);
+               cmap_entries = (unsigned int)make_ga_colormap(display);
               background_index = PNG_CMAP_GA_BACKGROUND;
               output_processing = PNG_CMAP_GA;
            }
@@ -2546,12 +2561,12 @@ png_image_read_colormap(png_voidp argument)
                  png_ptr->num_trans > 0) &&
                  png_gamma_not_sRGB(png_ptr->colorspace.gamma) != 0)
               {
-                  cmap_entries = make_gray_file_colormap(display);
+                  cmap_entries = (unsigned int)make_gray_file_colormap(display);
                  data_encoding = P_FILE;
               }

               else
-                  cmap_entries = make_gray_colormap(display);
+                  cmap_entries = (unsigned int)make_gray_colormap(display);

               /* But if the input has alpha or transparency it must be removed
                */
@@ -2639,7 +2654,7 @@ png_image_read_colormap(png_voidp argument)
                  if (PNG_RGB_COLORMAP_ENTRIES+1+27 > image->colormap_entries)
                     png_error(png_ptr, "rgb+alpha color-map: too few entries");

-                  cmap_entries = make_rgb_colormap(display);
+                  cmap_entries = (unsigned int)make_rgb_colormap(display);

                  /* Add a transparent entry. */
                  png_create_colormap_entry(display, cmap_entries, 255, 255,
@@ -2688,7 +2703,7 @@ png_image_read_colormap(png_voidp argument)
                  if (PNG_RGB_COLORMAP_ENTRIES+1+27 > image->colormap_entries)
                     png_error(png_ptr, "rgb-alpha color-map: too few entries");

-                  cmap_entries = make_rgb_colormap(display);
+                  cmap_entries = (unsigned int)make_rgb_colormap(display);

                  png_create_colormap_entry(display, cmap_entries, back_r,
                      back_g, back_b, 0/*unused*/, output_encoding);
@@ -2773,7 +2788,7 @@ png_image_read_colormap(png_voidp argument)
               if (PNG_RGB_COLORMAP_ENTRIES > image->colormap_entries)
                  png_error(png_ptr, "rgb color-map: too few entries");

-               cmap_entries = make_rgb_colormap(display);
+               cmap_entries = (unsigned int)make_rgb_colormap(display);
               output_processing = PNG_CMAP_RGB;
            }
         }
@@ -2797,11 +2812,11 @@ png_image_read_colormap(png_voidp argument)

            output_processing = PNG_CMAP_NONE;
            data_encoding = P_FILE; /* Don't change from color-map indices */
-            cmap_entries = png_ptr->num_palette;
+            cmap_entries = (unsigned int)png_ptr->num_palette;
            if (cmap_entries > 256)
               cmap_entries = 256;

-            if (cmap_entries > image->colormap_entries)
+            if (cmap_entries > (unsigned int)image->colormap_entries)
               png_error(png_ptr, "palette color-map: too few entries");

            for (i=0; i < cmap_entries; ++i)
@@ -2859,7 +2874,7 @@ png_image_read_colormap(png_voidp argument)
      case P_sRGB:
         /* Change to 8-bit sRGB */
         png_set_alpha_mode_fixed(png_ptr, PNG_ALPHA_PNG, PNG_GAMMA_sRGB);
-         /* FALL THROUGH */
+         /* FALLTHROUGH */

      case P_FILE:
         if (png_ptr->bit_depth > 8)
@@ -2913,7 +2928,7 @@ png_image_read_colormap(png_voidp argument)
         png_error(png_ptr, "bad background index (internal error)");
   }

-   display->colormap_processing = output_processing;
+   display->colormap_processing = (int)output_processing;

   return 1/*ok*/;
 }
@@ -3177,8 +3192,7 @@ png_image_read_colormapped(png_voidp argument)
            image->colormap_entries == 244 /* 216 + 1 + 27 */)
            break;

-         /* goto bad_output; */
-         /* FALL THROUGH */
+         goto bad_output;

      default:
      bad_output:
@@ -3222,14 +3236,14 @@ png_image_read_colormapped(png_voidp argument)

   else
   {
-      png_alloc_size_t row_bytes = display->row_bytes;
+      png_alloc_size_t row_bytes = (png_alloc_size_t)display->row_bytes;

      while (--passes >= 0)
      {
         png_uint_32      y = image->height;
         png_bytep        row = png_voidcast(png_bytep, display->first_row);

-         while (y-- > 0)
+         for (; y > 0; --y)
         {
            png_read_row(png_ptr, row, NULL);
            row += row_bytes;
@@ -3557,8 +3571,9 @@ png_image_read_background(png_voidp argument)
             * stride which was multiplied by 2 (below) to get row_bytes.
             */
            ptrdiff_t    step_row = display->row_bytes / 2;
-            int preserve_alpha = (image->format & PNG_FORMAT_FLAG_ALPHA) != 0;
-            unsigned int outchannels = 1+preserve_alpha;
+            unsigned int preserve_alpha = (image->format &
+                PNG_FORMAT_FLAG_ALPHA) != 0;
+            unsigned int outchannels = 1U+preserve_alpha;
            int swap_alpha = 0;

 #           ifdef PNG_SIMPLIFIED_READ_AFIRST_SUPPORTED
@@ -3744,7 +3759,13 @@ png_image_read_direct(png_voidp argument)
         mode = PNG_ALPHA_PNG;
         output_gamma = PNG_DEFAULT_sRGB;
      }
-
+      
+      if ((change & PNG_FORMAT_FLAG_ASSOCIATED_ALPHA) != 0)
+      {
+         mode = PNG_ALPHA_OPTIMIZED;
+         change &= ~PNG_FORMAT_FLAG_ASSOCIATED_ALPHA;
+      }
+      
      /* If 'do_local_background' is set check for the presence of gamma
       * correction; this is part of the work-round for the libpng bug
       * described above.
@@ -3970,6 +3991,10 @@ png_image_read_direct(png_voidp argument)
      else if (do_local_compose != 0) /* internal error */
         png_error(png_ptr, "png_image_read: alpha channel lost");

+      if ((format & PNG_FORMAT_FLAG_ASSOCIATED_ALPHA) != 0) {
+         info_format |= PNG_FORMAT_FLAG_ASSOCIATED_ALPHA;
+      }
+
      if (info_ptr->bit_depth == 16)
         info_format |= PNG_FORMAT_FLAG_LINEAR;

@@ -4055,14 +4080,14 @@ png_image_read_direct(png_voidp argument)

   else
   {
-      png_alloc_size_t row_bytes = display->row_bytes;
+      png_alloc_size_t row_bytes = (png_alloc_size_t)display->row_bytes;

      while (--passes >= 0)
      {
         png_uint_32      y = image->height;
         png_bytep        row = png_voidcast(png_bytep, display->first_row);

-         while (y-- > 0)
+         for (; y > 0; --y)
         {
            png_read_row(png_ptr, row, NULL);
            row += row_bytes;
@@ -4091,7 +4116,7 @@ png_image_finish_read(png_imagep image, png_const_colorp background,
       * bits; this is just to verify that the 'row_stride' argument can be
       * represented.
       */
-      if (image->width <= 0x7FFFFFFFU/channels) /* no overflow */
+      if (image->width <= 0x7fffffffU/channels) /* no overflow */
      {
         png_uint_32 check;
         const png_uint_32 png_row_stride = image->width * channels;
@@ -4100,10 +4125,10 @@ png_image_finish_read(png_imagep image, png_const_colorp background,
            row_stride = (png_int_32)/*SAFE*/png_row_stride;

         if (row_stride < 0)
-            check = -row_stride;
+            check = (png_uint_32)(-row_stride);

         else
-            check = row_stride;
+            check = (png_uint_32)row_stride;

         /* This verifies 'check', the absolute value of the actual stride
          * passed in and detects overflow in the application calculation (i.e.
@@ -4128,7 +4153,7 @@ png_image_finish_read(png_imagep image, png_const_colorp background,
             * accomodated on 64-bit systems.
             */
            if (image->height <=
-                0xFFFFFFFFU/PNG_IMAGE_PIXEL_COMPONENT_SIZE(image->format)/check)
+                0xffffffffU/PNG_IMAGE_PIXEL_COMPONENT_SIZE(image->format)/check)
            {
               if ((image->format & PNG_FORMAT_FLAG_COLORMAP) == 0 ||
                  (image->colormap_entries > 0 && colormap != NULL))
@@ -1,8 +1,8 @@

 /* pngrtran.c - transforms the data in a row for PNG readers
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.33 [September 28, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -49,6 +49,7 @@ png_set_crc_action(png_structrp png_ptr, int crit_action, int ancil_action)
      case PNG_CRC_WARN_DISCARD:    /* Not a valid action for critical data */
         png_warning(png_ptr,
             "Can't discard critical data on CRC error");
+         /* FALLTHROUGH */
      case PNG_CRC_ERROR_QUIT:                                /* Error/quit */

      case PNG_CRC_DEFAULT:
@@ -429,7 +430,7 @@ png_set_quantize(png_structrp png_ptr, png_colorp palette,
      int i;

      png_ptr->quantize_index = (png_bytep)png_malloc(png_ptr,
-          (png_uint_32)(num_palette * (sizeof (png_byte))));
+          (png_alloc_size_t)((png_uint_32)num_palette * (sizeof (png_byte))));
      for (i = 0; i < num_palette; i++)
         png_ptr->quantize_index[i] = (png_byte)i;
   }
@@ -446,7 +447,7 @@ png_set_quantize(png_structrp png_ptr, png_colorp palette,

         /* Initialize an array to sort colors */
         png_ptr->quantize_sort = (png_bytep)png_malloc(png_ptr,
-             (png_uint_32)(num_palette * (sizeof (png_byte))));
+             (png_alloc_size_t)((png_uint_32)num_palette * (sizeof (png_byte))));

         /* Initialize the quantize_sort array */
         for (i = 0; i < num_palette; i++)
@@ -580,9 +581,11 @@ png_set_quantize(png_structrp png_ptr, png_colorp palette,

         /* Initialize palette index arrays */
         png_ptr->index_to_palette = (png_bytep)png_malloc(png_ptr,
-             (png_uint_32)(num_palette * (sizeof (png_byte))));
+             (png_alloc_size_t)((png_uint_32)num_palette *
+             (sizeof (png_byte))));
         png_ptr->palette_to_index = (png_bytep)png_malloc(png_ptr,
-             (png_uint_32)(num_palette * (sizeof (png_byte))));
+             (png_alloc_size_t)((png_uint_32)num_palette *
+             (sizeof (png_byte))));

         /* Initialize the sort array */
         for (i = 0; i < num_palette; i++)
@@ -591,7 +594,7 @@ png_set_quantize(png_structrp png_ptr, png_colorp palette,
            png_ptr->palette_to_index[i] = (png_byte)i;
         }

-         hash = (png_dsortpp)png_calloc(png_ptr, (png_uint_32)(769 *
+         hash = (png_dsortpp)png_calloc(png_ptr, (png_alloc_size_t)(769 *
             (sizeof (png_dsortp))));

         num_new_palette = num_palette;
@@ -622,7 +625,7 @@ png_set_quantize(png_structrp png_ptr, png_colorp palette,
                  {

                     t = (png_dsortp)png_malloc_warn(png_ptr,
-                         (png_uint_32)(sizeof (png_dsort)));
+                         (png_alloc_size_t)(sizeof (png_dsort)));

                     if (t == NULL)
                         break;
@@ -747,9 +750,9 @@ png_set_quantize(png_structrp png_ptr, png_colorp palette,
      png_size_t num_entries = ((png_size_t)1 << total_bits);

      png_ptr->palette_lookup = (png_bytep)png_calloc(png_ptr,
-          (png_uint_32)(num_entries * (sizeof (png_byte))));
+          (png_alloc_size_t)(num_entries * (sizeof (png_byte))));

-      distance = (png_bytep)png_malloc(png_ptr, (png_uint_32)(num_entries *
+      distance = (png_bytep)png_malloc(png_ptr, (png_alloc_size_t)(num_entries *
          (sizeof (png_byte))));

      memset(distance, 0xff, num_entries * (sizeof (png_byte)));
@@ -1253,7 +1256,7 @@ png_init_rgb_transformations(png_structrp png_ptr)
            default:

            case 8:
-               /* FALL THROUGH (Already 8 bits) */
+               /* FALLTHROUGH */ /*  (Already 8 bits) */

            case 16:
               /* Already a full 16 bits */
@@ -2150,7 +2153,7 @@ png_do_unpack(png_row_infop row_info, png_bytep row)
         {
            png_bytep sp = row + (png_size_t)((row_width - 1) >> 3);
            png_bytep dp = row + (png_size_t)row_width - 1;
-            png_uint_32 shift = 7 - (int)((row_width + 7) & 0x07);
+            png_uint_32 shift = 7U - ((row_width + 7U) & 0x07);
            for (i = 0; i < row_width; i++)
            {
               *dp = (png_byte)((*sp >> shift) & 0x01);
@@ -2174,7 +2177,7 @@ png_do_unpack(png_row_infop row_info, png_bytep row)

            png_bytep sp = row + (png_size_t)((row_width - 1) >> 2);
            png_bytep dp = row + (png_size_t)row_width - 1;
-            png_uint_32 shift = (int)((3 - ((row_width + 3) & 0x03)) << 1);
+            png_uint_32 shift = ((3U - ((row_width + 3U) & 0x03)) << 1);
            for (i = 0; i < row_width; i++)
            {
               *dp = (png_byte)((*sp >> shift) & 0x03);
@@ -2197,7 +2200,7 @@ png_do_unpack(png_row_infop row_info, png_bytep row)
         {
            png_bytep sp = row + (png_size_t)((row_width - 1) >> 1);
            png_bytep dp = row + (png_size_t)row_width - 1;
-            png_uint_32 shift = (int)((1 - ((row_width + 1) & 0x01)) << 2);
+            png_uint_32 shift = ((1U - ((row_width + 1U) & 0x01)) << 2);
            for (i = 0; i < row_width; i++)
            {
               *dp = (png_byte)((*sp >> shift) & 0x0f);
@@ -2934,7 +2937,7 @@ png_do_gray_to_rgb(png_row_infop row_info, png_bytep row)
 * using the equation given in Poynton's ColorFAQ of 1998-01-04 at
 * <http://www.inforamp.net/~poynton/>  (THIS LINK IS DEAD June 2008 but
 * versions dated 1998 through November 2002 have been archived at
- * http://web.archive.org/web/20000816232553/http://www.inforamp.net/
+ * https://web.archive.org/web/20000816232553/www.inforamp.net/
 * ~poynton/notes/colour_and_gamma/ColorFAQ.txt )
 * Charles Poynton poynton at poynton.com
 *
@@ -3223,7 +3226,8 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr)
                        == png_ptr->trans_color.gray)
                     {
                        unsigned int tmp = *sp & (0x7f7f >> (7 - shift));
-                        tmp |= png_ptr->background.gray << shift;
+                        tmp |=
+                            (unsigned int)(png_ptr->background.gray << shift);
                        *sp = (png_byte)(tmp & 0xff);
                     }

@@ -3252,7 +3256,8 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr)
                            == png_ptr->trans_color.gray)
                        {
                           unsigned int tmp = *sp & (0x3f3f >> (6 - shift));
-                           tmp |= png_ptr->background.gray << shift;
+                           tmp |=
+                              (unsigned int)png_ptr->background.gray << shift;
                           *sp = (png_byte)(tmp & 0xff);
                        }

@@ -3262,7 +3267,7 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr)
                           unsigned int g = (gamma_table [p | (p << 2) |
                               (p << 4) | (p << 6)] >> 6) & 0x03;
                           unsigned int tmp = *sp & (0x3f3f >> (6 - shift));
-                           tmp |= g << shift;
+                           tmp |= (unsigned int)(g << shift);
                           *sp = (png_byte)(tmp & 0xff);
                        }

@@ -3288,7 +3293,8 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr)
                            == png_ptr->trans_color.gray)
                        {
                           unsigned int tmp = *sp & (0x3f3f >> (6 - shift));
-                           tmp |= png_ptr->background.gray << shift;
+                           tmp |=
+                               (unsigned int)png_ptr->background.gray << shift;
                           *sp = (png_byte)(tmp & 0xff);
                        }

@@ -3318,7 +3324,8 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr)
                            == png_ptr->trans_color.gray)
                        {
                           unsigned int tmp = *sp & (0x0f0f >> (4 - shift));
-                           tmp |= png_ptr->background.gray << shift;
+                           tmp |=
+                              (unsigned int)(png_ptr->background.gray << shift);
                           *sp = (png_byte)(tmp & 0xff);
                        }

@@ -3328,7 +3335,7 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr)
                           unsigned int g = (gamma_table[p | (p << 4)] >> 4) &
                              0x0f;
                           unsigned int tmp = *sp & (0x0f0f >> (4 - shift));
-                           tmp |= g << shift;
+                           tmp |= (unsigned int)(g << shift);
                           *sp = (png_byte)(tmp & 0xff);
                        }

@@ -3354,7 +3361,8 @@ png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr)
                            == png_ptr->trans_color.gray)
                        {
                           unsigned int tmp = *sp & (0x0f0f >> (4 - shift));
-                           tmp |= png_ptr->background.gray << shift;
+                           tmp |=
+                              (unsigned int)(png_ptr->background.gray << shift);
                           *sp = (png_byte)(tmp & 0xff);
                        }

@@ -4297,7 +4305,7 @@ png_do_expand_palette(png_row_infop row_info, png_bytep row,
            if (num_trans > 0)
            {
               sp = row + (png_size_t)row_width - 1;
-               dp = row + (png_size_t)(row_width << 2) - 1;
+               dp = row + ((png_size_t)row_width << 2) - 1;

               for (i = 0; i < row_width; i++)
               {
@@ -4458,7 +4466,7 @@ png_do_expand(png_row_infop row_info, png_bytep row,
            {
               gray = gray & 0xff;
               sp = row + (png_size_t)row_width - 1;
-               dp = row + (png_size_t)(row_width << 1) - 1;
+               dp = row + ((png_size_t)row_width << 1) - 1;

               for (i = 0; i < row_width; i++)
               {
@@ -4514,7 +4522,7 @@ png_do_expand(png_row_infop row_info, png_bytep row,
            png_byte green = (png_byte)(trans_color->green & 0xff);
            png_byte blue = (png_byte)(trans_color->blue & 0xff);
            sp = row + (png_size_t)row_info->rowbytes - 1;
-            dp = row + (png_size_t)(row_width << 2) - 1;
+            dp = row + ((png_size_t)row_width << 2) - 1;
            for (i = 0; i < row_width; i++)
            {
               if (*(sp - 2) == red && *(sp - 1) == green && *(sp) == blue)
@@ -4537,7 +4545,7 @@ png_do_expand(png_row_infop row_info, png_bytep row,
            png_byte green_low = (png_byte)(trans_color->green & 0xff);
            png_byte blue_low = (png_byte)(trans_color->blue & 0xff);
            sp = row + row_info->rowbytes - 1;
-            dp = row + (png_size_t)(row_width << 3) - 1;
+            dp = row + ((png_size_t)row_width << 3) - 1;
            for (i = 0; i < row_width; i++)
            {
               if (*(sp - 5) == red_high &&
@@ -4596,7 +4604,9 @@ png_do_expand_16(png_row_infop row_info, png_bytep row)
      png_byte *sp = row + row_info->rowbytes; /* source, last byte + 1 */
      png_byte *dp = sp + row_info->rowbytes;  /* destination, end + 1 */
      while (dp > sp)
-         dp[-2] = dp[-1] = *--sp, dp -= 2;
+      {
+         dp[-2] = dp[-1] = *--sp; dp -= 2;
+      }

      row_info->rowbytes *= 2;
      row_info->bit_depth = 16;
@@ -1,8 +1,8 @@

 /* pngrutil.c - utilities to read a PNG file
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.33 [September 28, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -86,7 +86,7 @@ png_get_int_32)(png_const_bytep buf)
 {
   png_uint_32 uval = png_get_uint_32(buf);
   if ((uval & 0x80000000) == 0) /* non-negative */
-      return uval;
+      return (png_int_32)uval;

   uval = (uval ^ 0xffffffff) + 1;  /* 2's complement: -x = ~x+1 */
   if ((uval & 0x80000000) == 0) /* no overflow */
@@ -181,6 +181,9 @@ png_read_chunk_header(png_structrp png_ptr)
   /* Check to see if chunk name is valid. */
   png_check_chunk_name(png_ptr, png_ptr->chunk_name);

+   /* Check for too-large chunk length */
+   png_check_chunk_length(png_ptr, length);
+
 #ifdef PNG_IO_STATE_SUPPORTED
   png_ptr->io_state = PNG_IO_READING | PNG_IO_CHUNK_DATA;
 #endif
@@ -311,6 +314,7 @@ png_read_buffer(png_structrp png_ptr, png_alloc_size_t new_size, int warn)

      if (buffer != NULL)
      {
+         memset(buffer, 0, new_size); /* just in case */
         png_ptr->read_buffer = buffer;
         png_ptr->read_buffer_size = new_size;
      }
@@ -370,11 +374,10 @@ png_inflate_claim(png_structrp png_ptr, png_uint_32 owner)
    */
   {
      int ret; /* zlib return code */
-#if PNG_ZLIB_VERNUM >= 0x1240
+#if ZLIB_VERNUM >= 0x1240
+      int window_bits = 0;

 # if defined(PNG_SET_OPTION_SUPPORTED) && defined(PNG_MAXIMUM_INFLATE_WINDOW)
-      int window_bits;
-
      if (((png_ptr->options >> PNG_MAXIMUM_INFLATE_WINDOW) & 3) ==
          PNG_OPTION_ON)
      {
@@ -384,13 +387,11 @@ png_inflate_claim(png_structrp png_ptr, png_uint_32 owner)

      else
      {
-         window_bits = 0;
         png_ptr->zstream_start = 1;
      }
-# else
-#   define window_bits 0
 # endif
-#endif
+
+#endif /* ZLIB_VERNUM >= 0x1240 */

      /* Set this for safety, just in case the previous owner left pointers to
       * memory allocations.
@@ -402,25 +403,32 @@ png_inflate_claim(png_structrp png_ptr, png_uint_32 owner)

      if ((png_ptr->flags & PNG_FLAG_ZSTREAM_INITIALIZED) != 0)
      {
-#if PNG_ZLIB_VERNUM < 0x1240
-         ret = inflateReset(&png_ptr->zstream);
-#else
+#if ZLIB_VERNUM >= 0x1240
         ret = inflateReset2(&png_ptr->zstream, window_bits);
+#else
+         ret = inflateReset(&png_ptr->zstream);
 #endif
      }

      else
      {
-#if PNG_ZLIB_VERNUM < 0x1240
-         ret = inflateInit(&png_ptr->zstream);
-#else
+#if ZLIB_VERNUM >= 0x1240
         ret = inflateInit2(&png_ptr->zstream, window_bits);
+#else
+         ret = inflateInit(&png_ptr->zstream);
 #endif

         if (ret == Z_OK)
            png_ptr->flags |= PNG_FLAG_ZSTREAM_INITIALIZED;
      }

+#if ZLIB_VERNUM >= 0x1290 && \
+   defined(PNG_SET_OPTION_SUPPORTED) && defined(PNG_IGNORE_ADLER32)
+      if (((png_ptr->options >> PNG_IGNORE_ADLER32) & 3) == PNG_OPTION_ON)
+         /* Turn off validation of the ADLER32 checksum in IDAT chunks */
+         ret = inflateValidate(&png_ptr->zstream, 0);
+#endif
+
      if (ret == Z_OK)
         png_ptr->zowner = owner;

@@ -435,7 +443,7 @@ png_inflate_claim(png_structrp png_ptr, png_uint_32 owner)
 #endif
 }

-#if PNG_ZLIB_VERNUM >= 0x1240
+#if ZLIB_VERNUM >= 0x1240
 /* Handle the start of the inflate stream if we called inflateInit2(strm,0);
 * in this case some zlib versions skip validation of the CINFO field and, in
 * certain circumstances, libpng may end up displaying an invalid image, in
@@ -461,6 +469,7 @@ png_zlib_inflate(png_structrp png_ptr, int flush)
 #endif /* Zlib >= 1.2.4 */

 #ifdef PNG_READ_COMPRESSED_TEXT_SUPPORTED
+#if defined(PNG_READ_zTXt_SUPPORTED) || defined (PNG_READ_iTXt_SUPPORTED)
 /* png_inflate now returns zlib error codes including Z_OK and Z_STREAM_END to
 * allow the caller to do multiple calls if required.  If the 'finish' flag is
 * set Z_FINISH will be passed to the final inflate() call and Z_STREAM_END must
@@ -590,7 +599,6 @@ png_inflate(png_structrp png_ptr, png_uint_32 owner, int finish,
   }
 }

-#if defined(PNG_READ_zTXt_SUPPORTED) || defined (PNG_READ_iTXt_SUPPORTED)
 /*
 * Decompress trailing data in a chunk.  The assumption is that read_buffer
 * points at an allocated area holding the contents of a chunk with a
@@ -666,6 +674,8 @@ png_decompress_chunk(png_structrp png_ptr,

               if (text != NULL)
               {
+                  memset(text, 0, buffer_size);
+
                  ret = png_inflate(png_ptr, png_ptr->chunk_name, 1/*finish*/,
                      png_ptr->read_buffer + prefix_size, &lzsize,
                      text + prefix_size, newlength);
@@ -713,7 +723,7 @@ png_decompress_chunk(png_structrp png_ptr,
                   * the extra space may otherwise be used as a Trojan Horse.
                   */
                  if (ret == Z_STREAM_END &&
-                     chunklength - prefix_size != lzsize)
+                      chunklength - prefix_size != lzsize)
                     png_chunk_benign_error(png_ptr, "extra compressed data");
               }

@@ -729,9 +739,7 @@ png_decompress_chunk(png_structrp png_ptr,
            {
               /* inflateReset failed, store the error message */
               png_zstream_error(png_ptr, ret);
-
-               if (ret == Z_STREAM_END)
-                  ret = PNG_UNEXPECTED_ZLIB_RETURN;
+               ret = PNG_UNEXPECTED_ZLIB_RETURN;
            }
         }

@@ -823,7 +831,7 @@ png_inflate_read(png_structrp png_ptr, png_bytep read_buffer, uInt read_size,
      return Z_STREAM_ERROR;
   }
 }
-#endif
+#endif /* READ_iCCP */

 /* Read and check the IDHR chunk */

@@ -1011,7 +1019,7 @@ png_handle_PLTE(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
   if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE)
 #endif
   {
-      png_crc_finish(png_ptr, (int) length - num * 3);
+      png_crc_finish(png_ptr, (png_uint_32) (length - (unsigned int)num * 3));
   }

 #ifndef PNG_READ_OPT_PLTE_SUPPORTED
@@ -1373,11 +1381,13 @@ png_handle_iCCP(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
    * chunk is just ignored, so does not invalidate the color space.  An
    * alternative is to set the 'invalid' flags at the start of this routine
    * and only clear them in they were not set before and all the tests pass.
-    * The minimum 'deflate' stream is assumed to be just the 2 byte header and
-    * 4 byte checksum.  The keyword must be at least one character and there is
-    * a terminator (0) byte and the compression method.
    */
-   if (length < 9)
+
+   /* The keyword must be at least one character and there is a
+    * terminator (0) byte and the compression method byte, and the
+    * 'zlib' datastream is at least 11 bytes.
+    */
+   if (length < 14)
   {
      png_crc_finish(png_ptr, length);
      png_chunk_benign_error(png_ptr, "too short");
@@ -1409,6 +1419,16 @@ png_handle_iCCP(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
      png_crc_read(png_ptr, (png_bytep)keyword, read_length);
      length -= read_length;

+      /* The minimum 'zlib' stream is assumed to be just the 2 byte header,
+       * 5 bytes minimum 'deflate' stream, and the 4 byte checksum.
+       */
+      if (length < 11)
+      {
+         png_crc_finish(png_ptr, length);
+         png_chunk_benign_error(png_ptr, "too short");
+         return;
+      }
+
      keyword_length = 0;
      while (keyword_length < 80 && keyword_length < read_length &&
         keyword[keyword_length] != 0)
@@ -1427,7 +1447,7 @@ png_handle_iCCP(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)

            if (png_inflate_claim(png_ptr, png_iCCP) == Z_OK)
            {
-               Byte profile_header[132];
+               Byte profile_header[132]={0};
               Byte local_buffer[PNG_INFLATE_BUF_SIZE];
               png_alloc_size_t size = (sizeof profile_header);

@@ -1457,7 +1477,7 @@ png_handle_iCCP(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
                        /* Now read the tag table; a variable size buffer is
                         * needed at this point, allocate one for the whole
                         * profile.  The header check has already validated
-                         * that none of these stuff will overflow.
+                         * that none of this stuff will overflow.
                         */
                        const png_uint_32 tag_count = png_get_uint_32(
                            profile_header+128);
@@ -1564,19 +1584,11 @@ png_handle_iCCP(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
                                       return;
                                    }
                                 }
-
-                                 else if (size > 0)
-                                    errmsg = "truncated";
-
-#ifndef __COVERITY__
-                                 else
+                                 if (errmsg == NULL)
                                    errmsg = png_ptr->zstream.msg;
-#endif
                              }
-
                              /* else png_icc_check_tag_table output an error */
                           }
-
                           else /* profile truncated */
                              errmsg = png_ptr->zstream.msg;
                        }
@@ -1717,13 +1729,13 @@ png_handle_sPLT(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
   data_length = length - (png_uint_32)(entry_start - buffer);

   /* Integrity-check the data length */
-   if ((data_length % entry_size) != 0)
+   if ((data_length % (unsigned int)entry_size) != 0)
   {
      png_warning(png_ptr, "sPLT chunk has bad length");
      return;
   }

-   dl = (png_int_32)(data_length / entry_size);
+   dl = (png_uint_32)(data_length / (unsigned int)entry_size);
   max_dl = PNG_SIZE_MAX / (sizeof (png_sPLT_entry));

   if (dl > max_dl)
@@ -1732,10 +1744,10 @@ png_handle_sPLT(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
      return;
   }

-   new_palette.nentries = (png_int_32)(data_length / entry_size);
+   new_palette.nentries = (png_int_32)(data_length / (unsigned int)entry_size);

-   new_palette.entries = (png_sPLT_entryp)png_malloc_warn(
-       png_ptr, new_palette.nentries * (sizeof (png_sPLT_entry)));
+   new_palette.entries = (png_sPLT_entryp)png_malloc_warn(png_ptr,
+       (png_alloc_size_t) new_palette.nentries * (sizeof (png_sPLT_entry)));

   if (new_palette.entries == NULL)
   {
@@ -2005,6 +2017,69 @@ png_handle_bKGD(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
 }
 #endif

+#ifdef PNG_READ_eXIf_SUPPORTED
+void /* PRIVATE */
+png_handle_eXIf(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
+{
+   unsigned int i;
+
+   png_debug(1, "in png_handle_eXIf");
+
+   if ((png_ptr->mode & PNG_HAVE_IHDR) == 0)
+      png_chunk_error(png_ptr, "missing IHDR");
+
+   if (length < 2)
+   {
+      png_crc_finish(png_ptr, length);
+      png_chunk_benign_error(png_ptr, "too short");
+      return;
+   }
+
+   else if (info_ptr == NULL || (info_ptr->valid & PNG_INFO_eXIf) != 0)
+   {
+      png_crc_finish(png_ptr, length);
+      png_chunk_benign_error(png_ptr, "duplicate");
+      return;
+   }
+
+   info_ptr->free_me |= PNG_FREE_EXIF;
+
+   info_ptr->eXIf_buf = png_voidcast(png_bytep,
+             png_malloc_warn(png_ptr, length));
+
+   if (info_ptr->eXIf_buf == NULL)
+   {
+      png_crc_finish(png_ptr, length);
+      png_chunk_benign_error(png_ptr, "out of memory");
+      return;
+   }
+
+   for (i = 0; i < length; i++)
+   {
+      png_byte buf[1];
+      png_crc_read(png_ptr, buf, 1);
+      info_ptr->eXIf_buf[i] = buf[0];
+      if (i == 1 && buf[0] != 'M' && buf[0] != 'I'
+                 && info_ptr->eXIf_buf[0] != buf[0])
+      {
+         png_crc_finish(png_ptr, length);
+         png_chunk_benign_error(png_ptr, "incorrect byte-order specifier");
+         png_free(png_ptr, info_ptr->eXIf_buf);
+         info_ptr->eXIf_buf = NULL;
+         return;
+      }
+   }
+
+   if (png_crc_finish(png_ptr, 0) != 0)
+      return;
+
+   png_set_eXIf_1(png_ptr, info_ptr, length, info_ptr->eXIf_buf);
+
+   png_free(png_ptr, info_ptr->eXIf_buf);
+   info_ptr->eXIf_buf = NULL;
+}
+#endif
+
 #ifdef PNG_READ_hIST_SUPPORTED
 void /* PRIVATE */
 png_handle_hIST(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
@@ -2533,6 +2608,9 @@ png_handle_zTXt(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
   if ((png_ptr->mode & PNG_HAVE_IDAT) != 0)
      png_ptr->mode |= PNG_AFTER_IDAT;

+   /* Note, "length" is sufficient here; we won't be adding
+    * a null terminator later.
+    */
   buffer = png_read_buffer(png_ptr, length, 2/*silent*/);

   if (buffer == NULL)
@@ -2579,23 +2657,28 @@ png_handle_zTXt(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length)
      {
         png_text text;

-         /* It worked; png_ptr->read_buffer now looks like a tEXt chunk except
-          * for the extra compression type byte and the fact that it isn't
-          * necessarily '\0' terminated.
-          */
-         buffer = png_ptr->read_buffer;
-         buffer[uncompressed_length+(keyword_length+2)] = 0;
+         if (png_ptr->read_buffer == NULL)
+           errmsg="Read failure in png_handle_zTXt";
+         else
+         {
+            /* It worked; png_ptr->read_buffer now looks like a tEXt chunk
+             * except for the extra compression type byte and the fact that
+             * it isn't necessarily '\0' terminated.
+             */
+            buffer = png_ptr->read_buffer;
+            buffer[uncompressed_length+(keyword_length+2)] = 0;

-         text.compression = PNG_TEXT_COMPRESSION_zTXt;
-         text.key = (png_charp)buffer;
-         text.text = (png_charp)(buffer + keyword_length+2);
-         text.text_length = uncompressed_length;
-         text.itxt_length = 0;
-         text.lang = NULL;
-         text.lang_key = NULL;
+            text.compression = PNG_TEXT_COMPRESSION_zTXt;
+            text.key = (png_charp)buffer;
+            text.text = (png_charp)(buffer + keyword_length+2);
+            text.text_length = uncompressed_length;
+            text.itxt_length = 0;
+            text.lang = NULL;
+            text.lang_key = NULL;

-         if (png_set_text_2(png_ptr, info_ptr, &text, 1) != 0)
-            errmsg = "insufficient memory";
+            if (png_set_text_2(png_ptr, info_ptr, &text, 1) != 0)
+               errmsg = "insufficient memory";
+         }
      }

      else
@@ -2971,7 +3054,7 @@ png_handle_unknown(png_structrp png_ptr, png_inforp info_ptr,
         case 2:
            png_ptr->user_chunk_cache_max = 1;
            png_chunk_benign_error(png_ptr, "no space in chunk cache");
-            /* FALL THROUGH */
+            /* FALLTHROUGH */
         case 1:
            /* NOTE: prior to 1.6.0 this case resulted in an unknown critical
             * chunk being skipped, now there will be a hard error below.
@@ -2980,7 +3063,7 @@ png_handle_unknown(png_structrp png_ptr, png_inforp info_ptr,

         default: /* not at limit */
            --(png_ptr->user_chunk_cache_max);
-            /* FALL THROUGH */
+            /* FALLTHROUGH */
         case 0: /* no limit */
 #  endif /* USER_LIMITS */
            /* Here when the limit isn't reached or when limits are compiled
@@ -3031,20 +3114,58 @@ png_handle_unknown(png_structrp png_ptr, png_inforp info_ptr,
 */

 void /* PRIVATE */
-png_check_chunk_name(png_structrp png_ptr, png_uint_32 chunk_name)
+png_check_chunk_name(png_const_structrp png_ptr, const png_uint_32 chunk_name)
 {
   int i;
+   png_uint_32 cn=chunk_name;

   png_debug(1, "in png_check_chunk_name");

   for (i=1; i<=4; ++i)
   {
-      int c = chunk_name & 0xff;
+      int c = cn & 0xff;

      if (c < 65 || c > 122 || (c > 90 && c < 97))
         png_chunk_error(png_ptr, "invalid chunk type");

-      chunk_name >>= 8;
+      cn >>= 8;
+   }
+}
+
+void /* PRIVATE */
+png_check_chunk_length(png_const_structrp png_ptr, const png_uint_32 length)
+{
+   png_alloc_size_t limit = PNG_UINT_31_MAX;
+
+# ifdef PNG_SET_USER_LIMITS_SUPPORTED
+   if (png_ptr->user_chunk_malloc_max > 0 &&
+       png_ptr->user_chunk_malloc_max < limit)
+      limit = png_ptr->user_chunk_malloc_max;
+# elif PNG_USER_CHUNK_MALLOC_MAX > 0
+   if (PNG_USER_CHUNK_MALLOC_MAX < limit)
+      limit = PNG_USER_CHUNK_MALLOC_MAX;
+# endif
+   if (png_ptr->chunk_name == png_IDAT)
+   {
+      png_alloc_size_t idat_limit = PNG_UINT_31_MAX;
+      size_t row_factor =
+         (png_ptr->width * png_ptr->channels * (png_ptr->bit_depth > 8? 2: 1)
+          + 1 + (png_ptr->interlaced? 6: 0));
+      if (png_ptr->height > PNG_UINT_32_MAX/row_factor)
+         idat_limit=PNG_UINT_31_MAX;
+      else
+         idat_limit = png_ptr->height * row_factor;
+      row_factor = row_factor > 32566? 32566 : row_factor;
+      idat_limit += 6 + 5*(idat_limit/row_factor+1); /* zlib+deflate overhead */
+      idat_limit=idat_limit < PNG_UINT_31_MAX? idat_limit : PNG_UINT_31_MAX;
+      limit = limit < idat_limit? idat_limit : limit;
+   }
+
+   if (length > limit)
+   {
+      png_debug2(0," length = %lu, limit = %lu",
+         (unsigned long)length,(unsigned long)limit);
+      png_chunk_error(png_ptr, "chunk data is too large");
   }
 }

@@ -3099,7 +3220,7 @@ png_combine_row(png_const_structrp png_ptr, png_bytep dp, int display)
 #     ifdef PNG_READ_PACKSWAP_SUPPORTED
      if ((png_ptr->transformations & PNG_PACKSWAP) != 0)
         /* little-endian byte */
-         end_mask = 0xff << end_mask;
+         end_mask = (unsigned int)(0xff << end_mask);

      else /* big-endian byte */
 #     endif
@@ -3373,7 +3494,7 @@ png_combine_row(png_const_structrp png_ptr, png_bytep dp, int display)
                */
               do
               {
-                  dp[0] = sp[0], dp[1] = sp[1];
+                  dp[0] = sp[0]; dp[1] = sp[1];

                  if (row_width <= bytes_to_jump)
                     return;
@@ -3394,7 +3515,7 @@ png_combine_row(png_const_structrp png_ptr, png_bytep dp, int display)
                */
               for (;;)
               {
-                  dp[0] = sp[0], dp[1] = sp[1], dp[2] = sp[2];
+                  dp[0] = sp[0]; dp[1] = sp[1]; dp[2] = sp[2];

                  if (row_width <= bytes_to_jump)
                     return;
@@ -3545,7 +3666,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
 {
   /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */
   /* Offset to next interlace block */
-   static PNG_CONST int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1};
+   static PNG_CONST unsigned int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1};

   png_debug(1, "in png_do_read_interlace");
   if (row != NULL && row_info != NULL)
@@ -3560,9 +3681,10 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
         {
            png_bytep sp = row + (png_size_t)((row_info->width - 1) >> 3);
            png_bytep dp = row + (png_size_t)((final_width - 1) >> 3);
-            int sshift, dshift;
-            int s_start, s_end, s_inc;
-            int jstop = png_pass_inc[pass];
+            unsigned int sshift, dshift;
+            unsigned int s_start, s_end;
+            int s_inc;
+            int jstop = (int)png_pass_inc[pass];
            png_byte v;
            png_uint_32 i;
            int j;
@@ -3570,8 +3692,8 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
 #ifdef PNG_READ_PACKSWAP_SUPPORTED
            if ((transformations & PNG_PACKSWAP) != 0)
            {
-                sshift = (int)((row_info->width + 7) & 0x07);
-                dshift = (int)((final_width + 7) & 0x07);
+                sshift = ((row_info->width + 7) & 0x07);
+                dshift = ((final_width + 7) & 0x07);
                s_start = 7;
                s_end = 0;
                s_inc = -1;
@@ -3580,8 +3702,8 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
            else
 #endif
            {
-                sshift = 7 - (int)((row_info->width + 7) & 0x07);
-                dshift = 7 - (int)((final_width + 7) & 0x07);
+                sshift = 7 - ((row_info->width + 7) & 0x07);
+                dshift = 7 - ((final_width + 7) & 0x07);
                s_start = 0;
                s_end = 7;
                s_inc = 1;
@@ -3593,7 +3715,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
               for (j = 0; j < jstop; j++)
               {
                  unsigned int tmp = *dp & (0x7f7f >> (7 - dshift));
-                  tmp |= v << dshift;
+                  tmp |= (unsigned int)(v << dshift);
                  *dp = (png_byte)(tmp & 0xff);

                  if (dshift == s_end)
@@ -3603,7 +3725,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
                  }

                  else
-                     dshift += s_inc;
+                     dshift = (unsigned int)((int)dshift + s_inc);
               }

               if (sshift == s_end)
@@ -3613,7 +3735,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
               }

               else
-                  sshift += s_inc;
+                  sshift = (unsigned int)((int)sshift + s_inc);
            }
            break;
         }
@@ -3622,16 +3744,17 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
         {
            png_bytep sp = row + (png_uint_32)((row_info->width - 1) >> 2);
            png_bytep dp = row + (png_uint_32)((final_width - 1) >> 2);
-            int sshift, dshift;
-            int s_start, s_end, s_inc;
-            int jstop = png_pass_inc[pass];
+            unsigned int sshift, dshift;
+            unsigned int s_start, s_end;
+            int s_inc;
+            int jstop = (int)png_pass_inc[pass];
            png_uint_32 i;

 #ifdef PNG_READ_PACKSWAP_SUPPORTED
            if ((transformations & PNG_PACKSWAP) != 0)
            {
-               sshift = (int)(((row_info->width + 3) & 0x03) << 1);
-               dshift = (int)(((final_width + 3) & 0x03) << 1);
+               sshift = (((row_info->width + 3) & 0x03) << 1);
+               dshift = (((final_width + 3) & 0x03) << 1);
               s_start = 6;
               s_end = 0;
               s_inc = -2;
@@ -3640,8 +3763,8 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
            else
 #endif
            {
-               sshift = (int)((3 - ((row_info->width + 3) & 0x03)) << 1);
-               dshift = (int)((3 - ((final_width + 3) & 0x03)) << 1);
+               sshift = ((3 - ((row_info->width + 3) & 0x03)) << 1);
+               dshift = ((3 - ((final_width + 3) & 0x03)) << 1);
               s_start = 0;
               s_end = 6;
               s_inc = 2;
@@ -3656,7 +3779,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
               for (j = 0; j < jstop; j++)
               {
                  unsigned int tmp = *dp & (0x3f3f >> (6 - dshift));
-                  tmp |= v << dshift;
+                  tmp |= (unsigned int)(v << dshift);
                  *dp = (png_byte)(tmp & 0xff);

                  if (dshift == s_end)
@@ -3666,7 +3789,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
                  }

                  else
-                     dshift += s_inc;
+                     dshift = (unsigned int)((int)dshift + s_inc);
               }

               if (sshift == s_end)
@@ -3676,7 +3799,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
               }

               else
-                  sshift += s_inc;
+                  sshift = (unsigned int)((int)sshift + s_inc);
            }
            break;
         }
@@ -3685,16 +3808,17 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
         {
            png_bytep sp = row + (png_size_t)((row_info->width - 1) >> 1);
            png_bytep dp = row + (png_size_t)((final_width - 1) >> 1);
-            int sshift, dshift;
-            int s_start, s_end, s_inc;
+            unsigned int sshift, dshift;
+            unsigned int s_start, s_end;
+            int s_inc;
            png_uint_32 i;
-            int jstop = png_pass_inc[pass];
+            int jstop = (int)png_pass_inc[pass];

 #ifdef PNG_READ_PACKSWAP_SUPPORTED
            if ((transformations & PNG_PACKSWAP) != 0)
            {
-               sshift = (int)(((row_info->width + 1) & 0x01) << 2);
-               dshift = (int)(((final_width + 1) & 0x01) << 2);
+               sshift = (((row_info->width + 1) & 0x01) << 2);
+               dshift = (((final_width + 1) & 0x01) << 2);
               s_start = 4;
               s_end = 0;
               s_inc = -4;
@@ -3703,8 +3827,8 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
            else
 #endif
            {
-               sshift = (int)((1 - ((row_info->width + 1) & 0x01)) << 2);
-               dshift = (int)((1 - ((final_width + 1) & 0x01)) << 2);
+               sshift = ((1 - ((row_info->width + 1) & 0x01)) << 2);
+               dshift = ((1 - ((final_width + 1) & 0x01)) << 2);
               s_start = 0;
               s_end = 4;
               s_inc = 4;
@@ -3718,7 +3842,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
               for (j = 0; j < jstop; j++)
               {
                  unsigned int tmp = *dp & (0xf0f >> (4 - dshift));
-                  tmp |= v << dshift;
+                  tmp |= (unsigned int)(v << dshift);
                  *dp = (png_byte)(tmp & 0xff);

                  if (dshift == s_end)
@@ -3728,7 +3852,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
                  }

                  else
-                     dshift += s_inc;
+                     dshift = (unsigned int)((int)dshift + s_inc);
               }

               if (sshift == s_end)
@@ -3738,7 +3862,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,
               }

               else
-                  sshift += s_inc;
+                  sshift = (unsigned int)((int)sshift + s_inc);
            }
            break;
         }
@@ -3752,7 +3876,7 @@ png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass,

            png_bytep dp = row + (png_size_t)(final_width - 1) * pixel_bytes;

-            int jstop = png_pass_inc[pass];
+            int jstop = (int)png_pass_inc[pass];
            png_uint_32 i;

            for (i = 0; i < row_info->width; i++)
@@ -3880,7 +4004,10 @@ png_read_filter_row_paeth_1byte_pixel(png_row_infop row_info, png_bytep row,
      /* Find the best predictor, the least of pa, pb, pc favoring the earlier
       * ones in the case of a tie.
       */
-      if (pb < pa) pa = pb, a = b;
+      if (pb < pa)
+      {
+         pa = pb; a = b;
+      }
      if (pc < pa) a = c;

      /* Calculate the current pixel in a, and move the previous row pixel to c
@@ -3896,7 +4023,7 @@ static void
 png_read_filter_row_paeth_multibyte_pixel(png_row_infop row_info, png_bytep row,
    png_const_bytep prev_row)
 {
-   int bpp = (row_info->pixel_depth + 7) >> 3;
+   unsigned int bpp = (row_info->pixel_depth + 7) >> 3;
   png_bytep rp_end = row + bpp;

   /* Process the first pixel in the row completely (this is the same as 'up'
@@ -3909,7 +4036,7 @@ png_read_filter_row_paeth_multibyte_pixel(png_row_infop row_info, png_bytep row,
   }

   /* Remainder */
-   rp_end += row_info->rowbytes - bpp;
+   rp_end = rp_end + (row_info->rowbytes - bpp);

   while (row < rp_end)
   {
@@ -3932,7 +4059,10 @@ png_read_filter_row_paeth_multibyte_pixel(png_row_infop row_info, png_bytep row,
      pc = (p + pc) < 0 ? -(p + pc) : p + pc;
 #endif

-      if (pb < pa) pa = pb, a = b;
+      if (pb < pa)
+      {
+         pa = pb; a = b;
+      }
      if (pc < pa) a = c;

      a += *row;
@@ -4254,7 +4384,7 @@ png_read_start_row(png_structrp png_ptr)
   /* Offset to next interlace block in the y direction */
   static PNG_CONST png_byte png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2};

-   int max_pixel_depth;
+   unsigned int max_pixel_depth;
   png_size_t row_bytes;

   png_debug(1, "in png_read_start_row");
@@ -4283,7 +4413,7 @@ png_read_start_row(png_structrp png_ptr)
      png_ptr->iwidth = png_ptr->width;
   }

-   max_pixel_depth = png_ptr->pixel_depth;
+   max_pixel_depth = (unsigned int)png_ptr->pixel_depth;

   /* WARNING: * png_read_transform_info (pngrtran.c) performs a simpler set of
    * calculations to calculate the final pixel depth, then
@@ -4418,7 +4548,7 @@ png_read_start_row(png_structrp png_ptr)
 defined(PNG_USER_TRANSFORM_PTR_SUPPORTED)
   if ((png_ptr->transformations & PNG_USER_TRANSFORM) != 0)
   {
-      int user_pixel_depth = png_ptr->user_transform_depth *
+      unsigned int user_pixel_depth = png_ptr->user_transform_depth *
         png_ptr->user_transform_channels;

      if (user_pixel_depth > max_pixel_depth)
@@ -4440,7 +4570,7 @@ defined(PNG_USER_TRANSFORM_PTR_SUPPORTED)
    * for safety's sake
    */
   row_bytes = PNG_ROWBYTES(max_pixel_depth, row_bytes) +
-       1 + ((max_pixel_depth + 7) >> 3);
+       1 + ((max_pixel_depth + 7) >> 3U);

 #ifdef PNG_MAX_MALLOC_64K
   if (row_bytes > (png_uint_32)65536L)
@@ -4509,7 +4639,7 @@ defined(PNG_USER_TRANSFORM_PTR_SUPPORTED)
    * does not, so free the read buffer now regardless; the sequential reader
    * reallocates it on demand.
    */
-   if (png_ptr->read_buffer != 0)
+   if (png_ptr->read_buffer != NULL)
   {
      png_bytep buffer = png_ptr->read_buffer;

@@ -1,8 +1,8 @@

 /* pngset.c - storage of image information into info struct
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.32 [August 24, 2017]
+ * Copyright (c) 1998-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -134,6 +134,53 @@ png_set_cHRM_XYZ(png_const_structrp png_ptr, png_inforp info_ptr, double red_X,

 #endif /* cHRM */

+#ifdef PNG_eXIf_SUPPORTED
+void PNGAPI
+png_set_eXIf(png_const_structrp png_ptr, png_inforp info_ptr,
+    const png_bytep eXIf_buf)
+{
+  png_warning(png_ptr, "png_set_eXIf does not work; use png_set_eXIf_1");
+  PNG_UNUSED(info_ptr)
+  PNG_UNUSED(eXIf_buf)
+}
+
+void PNGAPI
+png_set_eXIf_1(png_const_structrp png_ptr, png_inforp info_ptr,
+    const png_uint_32 num_exif, const png_bytep eXIf_buf)
+{
+   int i;
+
+   png_debug1(1, "in %s storage function", "eXIf");
+
+   if (png_ptr == NULL || info_ptr == NULL)
+      return;
+
+   if (info_ptr->exif)
+   {
+      png_free(png_ptr, info_ptr->exif);
+      info_ptr->exif = NULL;
+   }
+
+   info_ptr->num_exif = num_exif;
+
+   info_ptr->exif = png_voidcast(png_bytep, png_malloc_warn(png_ptr,
+       info_ptr->num_exif));
+
+   if (info_ptr->exif == NULL)
+   {
+      png_warning(png_ptr, "Insufficient memory for eXIf chunk data");
+      return;
+   }
+
+   info_ptr->free_me |= PNG_FREE_EXIF;
+
+   for (i = 0; i < (int) info_ptr->num_exif; i++)
+      info_ptr->exif[i] = eXIf_buf[i];
+
+   info_ptr->valid |= PNG_INFO_eXIf;
+}
+#endif /* eXIf */
+
 #ifdef PNG_gAMA_SUPPORTED
 void PNGFAPI
 png_set_gAMA_fixed(png_const_structrp png_ptr, png_inforp info_ptr,
@@ -283,17 +330,29 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr,

   /* Check that the type matches the specification. */
   if (type < 0 || type > 3)
-      png_error(png_ptr, "Invalid pCAL equation type");
+   {
+      png_chunk_report(png_ptr, "Invalid pCAL equation type",
+            PNG_CHUNK_WRITE_ERROR);
+      return;
+   }

   if (nparams < 0 || nparams > 255)
-      png_error(png_ptr, "Invalid pCAL parameter count");
+   {
+      png_chunk_report(png_ptr, "Invalid pCAL parameter count",
+            PNG_CHUNK_WRITE_ERROR);
+      return;
+   }

   /* Validate params[nparams] */
   for (i=0; i<nparams; ++i)
   {
      if (params[i] == NULL ||
          !png_check_fp_string(params[i], strlen(params[i])))
-         png_error(png_ptr, "Invalid format for pCAL parameter");
+      {
+         png_chunk_report(png_ptr, "Invalid format for pCAL parameter",
+               PNG_CHUNK_WRITE_ERROR);
+         return;
+      }
   }

   info_ptr->pcal_purpose = png_voidcast(png_charp,
@@ -301,8 +360,8 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr,

   if (info_ptr->pcal_purpose == NULL)
   {
-      png_warning(png_ptr, "Insufficient memory for pCAL purpose");
-
+      png_chunk_report(png_ptr, "Insufficient memory for pCAL purpose",
+            PNG_CHUNK_WRITE_ERROR);
      return;
   }

@@ -331,7 +390,7 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr,
   memcpy(info_ptr->pcal_units, units, length);

   info_ptr->pcal_params = png_voidcast(png_charpp, png_malloc_warn(png_ptr,
-       (png_size_t)((nparams + 1) * (sizeof (png_charp)))));
+       (png_size_t)(((unsigned int)nparams + 1) * (sizeof (png_charp)))));

   if (info_ptr->pcal_params == NULL)
   {
@@ -340,7 +399,8 @@ png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr,
      return;
   }

-   memset(info_ptr->pcal_params, 0, (nparams + 1) * (sizeof (png_charp)));
+   memset(info_ptr->pcal_params, 0, ((unsigned int)nparams + 1) *
+       (sizeof (png_charp)));

   for (i = 0; i < nparams; i++)
   {
@@ -563,7 +623,8 @@ png_set_PLTE(png_structrp png_ptr, png_inforp info_ptr,
       PNG_MAX_PALETTE_LENGTH * (sizeof (png_color))));

   if (num_palette > 0)
-      memcpy(png_ptr->palette, palette, num_palette * (sizeof (png_color)));
+      memcpy(png_ptr->palette, palette, (unsigned int)num_palette *
+          (sizeof (png_color)));
   info_ptr->palette = png_ptr->palette;
   info_ptr->num_palette = png_ptr->num_palette = (png_uint_16)num_palette;

@@ -1080,7 +1141,7 @@ png_set_sPLT(png_const_structrp png_ptr,
       * checked it when doing the allocation.
       */
      memcpy(np->entries, entries->entries,
-          entries->nentries * sizeof (png_sPLT_entry));
+          (unsigned int)entries->nentries * sizeof (png_sPLT_entry));

      /* Note that 'continue' skips the advance of the out pointer and out
       * count, so an invalid entry is not added.
@@ -1088,8 +1149,9 @@ png_set_sPLT(png_const_structrp png_ptr,
      info_ptr->valid |= PNG_INFO_sPLT;
      ++(info_ptr->splt_palettes_num);
      ++np;
+      ++entries;
   }
-   while (++entries, --nentries);
+   while (--nentries);

   if (nentries > 0)
      png_chunk_report(png_ptr, "sPLT out of memory", PNG_CHUNK_WRITE_ERROR);
@@ -1249,7 +1311,7 @@ png_set_unknown_chunk_location(png_const_structrp png_ptr, png_inforp info_ptr,
      {
         png_app_error(png_ptr, "invalid unknown chunk location");
         /* Fake out the pre 1.6.0 behavior: */
-         if ((location & PNG_HAVE_IDAT) != 0) /* undocumented! */
+         if (((unsigned int)location & PNG_HAVE_IDAT) != 0) /* undocumented! */
            location = PNG_AFTER_IDAT;

         else
@@ -1340,6 +1402,7 @@ png_set_keep_unknown_chunks(png_structrp png_ptr, int keep,
      static PNG_CONST png_byte chunks_to_ignore[] = {
         98,  75,  71,  68, '\0',  /* bKGD */
         99,  72,  82,  77, '\0',  /* cHRM */
+        101,  88,  73, 102, '\0',  /* eXIf */
        103,  65,  77,  65, '\0',  /* gAMA */
        104,  73,  83,  84, '\0',  /* hIST */
        105,  67,  67,  80, '\0',  /* iCCP */
@@ -1373,7 +1436,7 @@ png_set_keep_unknown_chunks(png_structrp png_ptr, int keep,
         return;
      }

-      num_chunks = num_chunks_in;
+      num_chunks = (unsigned int)num_chunks_in;
   }

   old_num_chunks = png_ptr->num_chunk_list;
@@ -1563,7 +1626,7 @@ void PNGAPI
 png_set_invalid(png_const_structrp png_ptr, png_inforp info_ptr, int mask)
 {
   if (png_ptr != NULL && info_ptr != NULL)
-      info_ptr->valid &= ~mask;
+      info_ptr->valid &= (unsigned int)(~mask);
 }


@@ -1682,14 +1745,16 @@ png_check_keyword(png_structrp png_ptr, png_const_charp key, png_bytep new_key)
      png_byte ch = (png_byte)*key++;

      if ((ch > 32 && ch <= 126) || (ch >= 161 /*&& ch <= 255*/))
-         *new_key++ = ch, ++key_len, space = 0;
+      {
+         *new_key++ = ch; ++key_len; space = 0;
+      }

      else if (space == 0)
      {
         /* A space or an invalid character when one wasn't seen immediately
          * before; output just a space.
          */
-         *new_key++ = 32, ++key_len, space = 1;
+         *new_key++ = 32; ++key_len; space = 1;

         /* If the character was not a space then it is invalid. */
         if (ch != 32)
@@ -1702,7 +1767,7 @@ png_check_keyword(png_structrp png_ptr, png_const_charp key, png_bytep new_key)

   if (key_len > 0 && space != 0) /* trailing space */
   {
-      --key_len, --new_key;
+      --key_len; --new_key;
      if (bad_character == 0)
         bad_character = 32;
   }
@@ -1,8 +1,8 @@

 /* pngstruct.h - header file for PNG reference library
 *
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.32 [August 24, 2017]
+ * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson
 * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
 * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
 *
@@ -263,7 +263,7 @@ struct png_struct_def
                              /* pixel depth used for the row buffers */
   png_byte transformed_pixel_depth;
                              /* pixel depth after read/write transforms */
-#if PNG_ZLIB_VERNUM >= 0x1240
+#if ZLIB_VERNUM >= 0x1240
   png_byte zstream_start;    /* at start of an input zlib stream */
 #endif /* Zlib >= 1.2.4 */
 #if defined(PNG_READ_FILLER_SUPPORTED) || defined(PNG_WRITE_FILLER_SUPPORTED)
@@ -353,7 +353,7 @@ struct png_struct_def

 /* Options */
 #ifdef PNG_SET_OPTION_SUPPORTED
-   png_byte options;           /* On/off state (up to 4 options) */
+   png_uint_32 options;           /* On/off state (up to 16 options) */
 #endif

 #if PNG_LIBPNG_VER < 10700
--- a/Show More
+++ b/Show More