[PATCH v9 0/1] MR9061: cabinet: Add LZX compression support to FCI

[Eric Lasota (＠ejlasota)]

This adds LZX compression support to FCI.

Most of the work for this was done here: https://github.com/elasota/liblzx

The LZX compression code is based on the LZX compression code from wimlib with modifications to support the cabinet variant and streaming compression.  Eric Biggers (the author of wimlib) has given permission to redistribute the modified LZX code under LGPLv2 terms: https://wimlib.net/forums/viewtopic.php?t=854

I've been testing this against some large data sets (dumped game assets, Visual Studio installation) with a test program and decompressing with 7-zip and diffs have been coming back OK.

--
  v9: cabinet: LZX compression support

https://gitlab.winehq.org/wine/wine/-/merge_requests/9061

[PATCH v9 1/1] cabinet: LZX compression support

[elasota]

From: elasota

---
 dlls/cabinet/Makefile.in                 |    5 +-
 dlls/cabinet/fci.c                       |  253 +-
 dlls/cabinet/liblzx.h                    |  141 +
 dlls/cabinet/liblzx_bitops.h             |  156 +
 dlls/cabinet/liblzx_bt_matchfinder.h     |  446 +++
 dlls/cabinet/liblzx_compiler.h           |  214 ++
 dlls/cabinet/liblzx_compress_common.c    |  673 ++++
 dlls/cabinet/liblzx_compress_common.h    |   19 +
 dlls/cabinet/liblzx_config.h             |   29 +
 dlls/cabinet/liblzx_endianness.h         |  136 +
 dlls/cabinet/liblzx_error.h              |   11 +
 dlls/cabinet/liblzx_hc_matchfinder.h     |  432 +++
 dlls/cabinet/liblzx_lzx_common.c         |  325 ++
 dlls/cabinet/liblzx_lzx_common.h         |   29 +
 dlls/cabinet/liblzx_lzx_compress.c       | 3662 ++++++++++++++++++++++
 dlls/cabinet/liblzx_lzx_constants.h      |  108 +
 dlls/cabinet/liblzx_matchfinder_common.h |  131 +
 dlls/cabinet/liblzx_minmax.h             |  122 +
 dlls/cabinet/liblzx_types.h              |   33 +
 dlls/cabinet/liblzx_unaligned.h          |  134 +
 dlls/cabinet/liblzx_util.h               |   20 +
 dlls/cabinet/tests/extract.c             |  240 +-
 22 files changed, 7181 insertions(+), 138 deletions(-)
 create mode 100644 dlls/cabinet/liblzx.h
 create mode 100644 dlls/cabinet/liblzx_bitops.h
 create mode 100644 dlls/cabinet/liblzx_bt_matchfinder.h
 create mode 100644 dlls/cabinet/liblzx_compiler.h
 create mode 100644 dlls/cabinet/liblzx_compress_common.c
 create mode 100644 dlls/cabinet/liblzx_compress_common.h
 create mode 100644 dlls/cabinet/liblzx_config.h
 create mode 100644 dlls/cabinet/liblzx_endianness.h
 create mode 100644 dlls/cabinet/liblzx_error.h
 create mode 100644 dlls/cabinet/liblzx_hc_matchfinder.h
 create mode 100644 dlls/cabinet/liblzx_lzx_common.c
 create mode 100644 dlls/cabinet/liblzx_lzx_common.h
 create mode 100644 dlls/cabinet/liblzx_lzx_compress.c
 create mode 100644 dlls/cabinet/liblzx_lzx_constants.h
 create mode 100644 dlls/cabinet/liblzx_matchfinder_common.h
 create mode 100644 dlls/cabinet/liblzx_minmax.h
 create mode 100644 dlls/cabinet/liblzx_types.h
 create mode 100644 dlls/cabinet/liblzx_unaligned.h
 create mode 100644 dlls/cabinet/liblzx_util.h

diff --git a/dlls/cabinet/Makefile.in b/dlls/cabinet/Makefile.in
index a72e5a31be6..40db872bbfd 100644
--- a/dlls/cabinet/Makefile.in
+++ b/dlls/cabinet/Makefile.in
@@ -7,4 +7,7 @@ SOURCES = \
    cabinet.rc \
    cabinet_main.c \
    fci.c \
-	fdi.c
+	fdi.c \
+	liblzx_compress_common.c \
+	liblzx_lzx_common.c \
+	liblzx_lzx_compress.c
diff --git a/dlls/cabinet/fci.c b/dlls/cabinet/fci.c
index e62399db1ba..87a29623656 100644
--- a/dlls/cabinet/fci.c
+++ b/dlls/cabinet/fci.c
@@ -47,6 +47,8 @@ There is still some work to be done:
 #include "wine/list.h"
 #include "wine/debug.h"

+#include "liblzx.h"
+
 WINE_DEFAULT_DEBUG_CHANNEL(cabinet);

#ifdef WORDS_BIGENDIAN
@@ -165,11 +167,13 @@ typedef struct FCI_Int
   char               szPrevDisk[CB_MAX_DISK_NAME];   /* disk name of previous cabinet */
   unsigned char      data_in[CAB_BLOCKMAX];          /* uncompressed data blocks */
   unsigned char      data_out[2 * CAB_BLOCKMAX];     /* compressed data blocks */
+  BOOL               have_data_out;
   cab_UWORD          cdata_in;
   ULONG              cCompressedBytesInFolder;
   cab_UWORD          cFolders;
   cab_UWORD          cFiles;
-  cab_ULONG          cDataBlocks;
+  cab_ULONG          cDataBlocksIn;
+  cab_ULONG          cDataBlocksOut;
   cab_ULONG          cbFileRemainder; /* uncompressed, yet to be written data */
                /* of spanned file of a spanning folder of a spanning cabinet */
   struct temp_file   data;
@@ -185,6 +189,9 @@ typedef struct FCI_Int
   cab_ULONG          folders_data_size;   /* total size of data contained in the current folders */
   TCOMP              compression;
   cab_UWORD        (*compress)(struct FCI_Int *);
+  cab_UWORD        (*flush)(struct FCI_Int *);
+  void             (*compress_shutdown)(struct FCI_Int *);
+  struct liblzx_compressor *lzx_compressor;
 } FCI_Int;

#define FCI_INT_MAGIC 0xfcfcfc05
@@ -274,7 +281,7 @@ static struct file *add_file( FCI_Int *fci, const char *filename )
         return NULL;
     }
     file->size    = 0;
-    file->offset  = fci->cDataBlocks * CAB_BLOCKMAX + fci->cdata_in;
+    file->offset  = fci->cDataBlocksIn * CAB_BLOCKMAX + fci->cdata_in;
     file->folder  = fci->cFolders;
     file->date    = 0;
     file->time    = 0;
@@ -305,43 +312,82 @@ static void free_file( FCI_Int *fci, struct file *file )
     fci->free( file );
 }

-/* create a new data block for the data in fci->data_in */
-static BOOL add_data_block( FCI_Int *fci, PFNFCISTATUS status_callback )
+/* creates new data blocks for the data in fci->data_in */
+static BOOL add_data_blocks( FCI_Int *fci, BOOL is_last_block, PFNFCISTATUS status_callback )
 {
     int err;
     struct data_block *block;
+    cab_UWORD compressed_size = 0;
+    cab_UWORD uncompressed_size = fci->cdata_in;

-    if (!fci->cdata_in) return TRUE;
+    if (!uncompressed_size)
+    {
+        if (fci->cDataBlocksIn == 0 || !is_last_block) return TRUE;
+    }

if (fci->data.handle == -1 && !create_temp_file( fci, &fci->data )) return FALSE;

-    if (!(block = fci->alloc( sizeof(*block) )))
+    if (uncompressed_size)
     {
-        set_error( fci, FCIERR_ALLOC_FAIL, ERROR_NOT_ENOUGH_MEMORY );
-        return FALSE;
+        compressed_size = fci->compress( fci );
+
+        fci->cdata_in = 0;
+        fci->cDataBlocksIn++;
     }
-    block->uncompressed = fci->cdata_in;
-    block->compressed   = fci->compress( fci );

-    if (fci->write( fci->data.handle, fci->data_out,
-                    block->compressed, &err, fci->pv ) != block->compressed)
+    if (compressed_size == 0 && is_last_block && fci->flush)
     {
-        set_error( fci, FCIERR_TEMP_FILE, err );
-        fci->free( block );
-        return FALSE;
+        compressed_size = fci->flush( fci );
     }

-    fci->cdata_in = 0;
-    fci->pending_data_size += sizeof(CFDATA) + fci->ccab.cbReserveCFData + block->compressed;
-    fci->cCompressedBytesInFolder += block->compressed;
-    fci->cDataBlocks++;
-    list_add_tail( &fci->blocks_list, &block->entry );
-
-    if (status_callback( statusFile, block->compressed, block->uncompressed, fci->pv ) == -1)
+    while (compressed_size > 0)
     {
-        set_error( fci, FCIERR_USER_ABORT, 0 );
-        return FALSE;
+        if (!(block = fci->alloc( sizeof(*block) )))
+        {
+            set_error( fci, FCIERR_ALLOC_FAIL, ERROR_NOT_ENOUGH_MEMORY );
+            return FALSE;
+        }
+
+        if (is_last_block && fci->cDataBlocksIn - 1 == fci->cDataBlocksOut)
+        {
+            block->uncompressed = uncompressed_size;
+        }
+        else
+        {
+            block->uncompressed = CAB_BLOCKMAX;
+        }
+
+        block->compressed = compressed_size;
+
+        if (fci->write( fci->data.handle, fci->data_out,
+                        block->compressed, &err, fci->pv ) != block->compressed)
+        {
+            set_error( fci, FCIERR_TEMP_FILE, err );
+            fci->free( block );
+            return FALSE;
+        }
+
+        fci->pending_data_size += sizeof(CFDATA) + fci->ccab.cbReserveCFData + block->compressed;
+        fci->cCompressedBytesInFolder += block->compressed;
+        fci->cDataBlocksOut++;
+        list_add_tail( &fci->blocks_list, &block->entry );
+
+        if (status_callback( statusFile, block->compressed, block->uncompressed, fci->pv ) == -1)
+        {
+            set_error( fci, FCIERR_USER_ABORT, 0 );
+            return FALSE;
+        }
+
+        if (is_last_block && fci->flush)
+        {
+            compressed_size = fci->flush( fci );
+        }
+        else
+        {
+            compressed_size = 0;
+        }
     }
+
     return TRUE;
 }

@@ -377,7 +423,7 @@ static BOOL add_file_data( FCI_Int *fci, char *sourcefile, char *filename, BOOL
         }
         file->size += len;
         fci->cdata_in += len;
-        if (fci->cdata_in == CAB_BLOCKMAX && !add_data_block( fci, status_callback )) return FALSE;
+        if (fci->cdata_in == CAB_BLOCKMAX && !add_data_blocks( fci, FALSE, status_callback )) return FALSE;
     }
     fci->close( handle, &err, fci->pv );
     return TRUE;
@@ -824,7 +870,8 @@ static BOOL add_data_to_folder( FCI_Int *fci, struct folder *folder, cab_ULONG *
         }
         if (split_block) break;
         free_data_block( fci, block );
-        fci->cDataBlocks--;
+        fci->cDataBlocksIn--;
+        fci->cDataBlocksOut--;
     }

if (list_empty( &fci->blocks_list )) return TRUE;
@@ -905,6 +952,10 @@ static cab_UWORD compress_NONE( FCI_Int *fci )
     return fci->cdata_in;
 }

+static void shutdown_NONE(FCI_Int *fci)
+{
+}
+
 static void *zalloc( void *opaque, unsigned int items, unsigned int size )
 {
     FCI_Int *fci = opaque;
@@ -938,9 +989,118 @@ static cab_UWORD compress_MSZIP( FCI_Int *fci )
     fci->data_out[1] = 'K';
     deflate( &stream, Z_FINISH );
     deflateEnd( &stream );
+    fci->have_data_out = TRUE;
     return stream.total_out + 2;
 }

+static void shutdown_MSZIP( FCI_Int *fci )
+{
+}
+
+static void shutdown_LZX(FCI_Int *fci)
+{
+    liblzx_compress_destroy(fci->lzx_compressor);
+    fci->lzx_compressor = NULL;
+}
+
+static void *compress_LZX_alloc_callback(void *userdata, size_t size)
+{
+    FCI_Int *fci = (FCI_Int *)userdata;
+
+    return fci->alloc((ULONG)size);
+}
+
+static void compress_LZX_free_callback(void *userdata, void *ptr)
+{
+    FCI_Int *fci = (FCI_Int *)userdata;
+
+    fci->free(ptr);
+}
+
+static cab_UWORD compress_LZX(FCI_Int *fci)
+{
+    size_t in_digested = 0;
+    size_t compressed_size = 0;
+    const liblzx_output_chunk_t *out_chunk = NULL;
+
+    if (fci->cDataBlocksIn == 0) {
+        /* First block, restart compression */
+        int window_size_bits = LZXCompressionWindowFromTCOMP(fci->compression);
+        liblzx_compress_properties_t props;
+
+        if (fci->lzx_compressor)
+        {
+            liblzx_compress_destroy(fci->lzx_compressor);
+            fci->lzx_compressor = NULL;
+        }
+
+        memset(&props, 0, sizeof(props));
+        props.lzx_variant = LIBLZX_VARIANT_CAB_DELTA;
+        props.window_size = 1 << window_size_bits;
+        props.chunk_granularity = CAB_BLOCKMAX;
+        props.compression_level = 70;
+        props.e8_file_size = LIBLZX_CONST_DEFAULT_E8_FILE_SIZE;
+        props.alloc_func = compress_LZX_alloc_callback;
+        props.free_func = compress_LZX_free_callback;
+        props.userdata = fci;
+
+        fci->lzx_compressor = liblzx_compress_create(&props);
+
+        if (!fci->lzx_compressor)
+        {
+            set_error(fci, FCIERR_ALLOC_FAIL, ERROR_OUTOFMEMORY);
+            return 0;
+        }
+    }
+
+    while (in_digested < fci->cdata_in)
+    {
+        in_digested += liblzx_compress_add_input(fci->lzx_compressor, fci->data_in + in_digested, fci->cdata_in - in_digested);
+
+        if (out_chunk)
+        {
+            /* After producing an output chunk, all data should be digestable. */
+            assert(in_digested == fci->cdata_in);
+            break;
+        }
+
+        out_chunk = liblzx_compress_get_next_chunk(fci->lzx_compressor);
+
+        if (out_chunk)
+        {
+            compressed_size = out_chunk->size;
+            memcpy(fci->data_out, out_chunk->data, compressed_size);
+            liblzx_compress_release_next_chunk(fci->lzx_compressor);
+
+            fci->have_data_out = TRUE;
+        }
+    }
+
+    return compressed_size;
+}
+
+cab_UWORD flush_LZX(FCI_Int *fci)
+{
+    const liblzx_output_chunk_t *out_chunk = NULL;
+    cab_UWORD compressed_size = 0;
+
+    liblzx_compress_end_input(fci->lzx_compressor);
+    out_chunk = liblzx_compress_get_next_chunk(fci->lzx_compressor);
+
+    if (out_chunk == NULL)
+    {
+        return 0;
+    }
+
+    compressed_size = out_chunk->size;
+    memcpy(fci->data_out, out_chunk->data, out_chunk->size);
+
+    liblzx_compress_release_next_chunk(fci->lzx_compressor);
+
+    fci->have_data_out = TRUE;
+
+    return compressed_size;
+}

/***********************************************************************
  *		FCICreate (CABINET.10)
@@ -1046,6 +1206,7 @@ HFCI __cdecl FCICreate(
   p_fci_internal->pv = pv;
   p_fci_internal->data.handle = -1;
   p_fci_internal->compress = compress_NONE;
+  p_fci_internal->compress_shutdown = shutdown_NONE;

list_init( &p_fci_internal->folders_list );
   list_init( &p_fci_internal->files_list );
@@ -1102,11 +1263,12 @@ static BOOL fci_flush_folder( FCI_Int *p_fci_internal,
   p_fci_internal->fSplitFolder=FALSE;

/* START of COPY */
-  if (!add_data_block( p_fci_internal, pfnfcis )) return FALSE;
+  if (!add_data_blocks( p_fci_internal, TRUE, pfnfcis )) return FALSE;

/* reset to get the number of data blocks of this folder which are */
   /* actually in this cabinet ( at least partially ) */
-  p_fci_internal->cDataBlocks=0;
+  p_fci_internal->cDataBlocksIn = 0;
+  p_fci_internal->cDataBlocksOut = 0;

p_fci_internal->statusFolderTotal = get_header_size( p_fci_internal ) +
       sizeof(CFFOLDER) + p_fci_internal->ccab.cbReserveCFFolder +
@@ -1211,7 +1373,8 @@ static BOOL fci_flush_folder( FCI_Int *p_fci_internal,
   if (!add_files_to_folder( p_fci_internal, folder, payload )) return FALSE;

/* reset CFFolder specific information */
-  p_fci_internal->cDataBlocks=0;
+  p_fci_internal->cDataBlocksIn=0;
+  p_fci_internal->cDataBlocksOut=0;
   p_fci_internal->cCompressedBytesInFolder=0;

return TRUE;
@@ -1409,19 +1572,41 @@ BOOL __cdecl FCIAddFile(

if (typeCompress != p_fci_internal->compression)
   {
+      if ((typeCompress & tcompMASK_TYPE) == tcompTYPE_LZX) {
+          TCOMP window_size_bits = (typeCompress & tcompMASK_LZX_WINDOW);
+
+          if (window_size_bits < tcompLZX_WINDOW_LO || window_size_bits > tcompLZX_WINDOW_HI) {
+              set_error(p_fci_internal, FCIERR_BAD_COMPR_TYPE, ERROR_BAD_ARGUMENTS);
+              return FALSE;
+          }
+      }
+
       if (!FCIFlushFolder( hfci, pfnfcignc, pfnfcis )) return FALSE;
-      switch (typeCompress)
+
+      p_fci_internal->compress_shutdown(p_fci_internal);
+
+      switch (typeCompress & tcompMASK_TYPE)
       {
       case tcompTYPE_MSZIP:
-          p_fci_internal->compression = tcompTYPE_MSZIP;
-          p_fci_internal->compress    = compress_MSZIP;
+          p_fci_internal->compression       = tcompTYPE_MSZIP;
+          p_fci_internal->compress          = compress_MSZIP;
+          p_fci_internal->flush             = NULL;
+          p_fci_internal->compress_shutdown = shutdown_MSZIP;
+          break;
+      case tcompTYPE_LZX:
+          p_fci_internal->compression       = typeCompress;
+          p_fci_internal->compress          = compress_LZX;
+          p_fci_internal->flush             = flush_LZX;
+          p_fci_internal->compress_shutdown = shutdown_LZX;
           break;
       default:
           FIXME( "compression %x not supported, defaulting to none\n", typeCompress );
           /* fall through */
       case tcompTYPE_NONE:
-          p_fci_internal->compression = tcompTYPE_NONE;
-          p_fci_internal->compress    = compress_NONE;
+          p_fci_internal->compression       = tcompTYPE_NONE;
+          p_fci_internal->compress          = compress_NONE;
+          p_fci_internal->flush             = NULL;
+          p_fci_internal->compress_shutdown = shutdown_NONE;
           break;
       }
   }
diff --git a/dlls/cabinet/liblzx.h b/dlls/cabinet/liblzx.h
new file mode 100644
index 00000000000..504a3289bac
--- /dev/null
+++ b/dlls/cabinet/liblzx.h
@@ -0,0 +1,141 @@
+/*
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright (C) 2012-2017 Eric Biggers
+ *
+ * This file is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU Lesser General Public License as published by the Free
+ * Software Foundation; either version 2.1 of the License, or (at your option) any
+ * later version.
+ *
+ * This file is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * along with this file; if not, see https://www.gnu.org/licenses/.
+ */
+
+#pragma once
+
+#ifndef __LIBLZX_H__
+#define __LIBLZX_H__
+
+#include <stddef.h>
+#include <stdint.h>
+
+#include "liblzx_error.h"
+
+typedef struct liblzx_internal liblzx_internal_t;
+typedef struct liblzx_compress_properties liblzx_compress_properties_t;
+typedef struct liblzx_compressor liblzx_compressor_t;
+typedef struct liblzx_output_chunk liblzx_output_chunk_t;
+
+typedef void *(*liblzx_alloc_func_t)(void *opaque, size_t size);
+typedef void (*liblzx_free_func_t)(void *opaque, void *ptr);
+
+enum liblzx_variant {
+        /* LZX variant used by CAB files and LZX DELTA */
+        LIBLZX_VARIANT_CAB_DELTA,
+
+        /* LZX variant used by WIM */
+        LIBLZX_VARIANT_WIM,
+};
+
+typedef enum liblzx_variant liblzx_variant_t;
+
+enum liblzx_constant {
+        LIBLZX_CONST_DEFAULT_CHUNK_SIZE = 32768,
+        LIBLZX_CONST_DEFAULT_E8_FILE_SIZE = 12 * 1024 * 1024,
+        LIBLZX_CONST_MAX_WINDOW_SIZE = 64 * 1024 * 1024,
+};
+
+struct liblzx_output_chunk {
+        const void *data;
+        size_t size;
+};
+
+struct liblzx_compress_properties {
+        /* LZX variant to use */
+        liblzx_variant_t lzx_variant;
+
+        /* Source file size for LZX DELTA.  Ignored for WIM.
+         * When using this, use liblzx_compress_add_input to add the source
+         * file's data before adding the new file's data.  For compression
+         * only, set this to 0.
+         */
+        size_t delta_source_size;
+
+        /* Compression window size. */
+        uint32_t window_size;
+
+        /* Granularity of a chunk.  Should generally be set to
+         * LIBLZX_CONST_DEFAULT_CHUNK_SIZE.
+         */
+        uint32_t chunk_granularity;
+
+        /* Compression level.  Can be set arbitrarily high. */
+        uint16_t compression_level;
+
+        /* E8 file size parameter.  For WIM, this is ignored.  For other
+         * variants, this value is expected to be user-controllable and
+         * is sent outside of the LZX data stream.
+         */
+        uint32_t e8_file_size;
+
+        /* Memory allocation function. */
+        liblzx_alloc_func_t alloc_func;
+
+        /* Memory free function. */
+        liblzx_free_func_t free_func;
+
+        /* Userdata parameter to pass to alloc function. */
+        void *userdata;
+};
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Creates a compressor object and returns a pointer to it. */
+liblzx_compressor_t *
+liblzx_compress_create(const liblzx_compress_properties_t *props);
+
+/* Destroys a compressor object and releases all resources. */
+void
+liblzx_compress_destroy(liblzx_compressor_t *stream);
+
+/* Resets a compressor to its initial state. */
+void
+liblzx_compress_reset(liblzx_compressor_t *stream);
+
+/* Adds input data to the compression stream and returns the number of bytes
+ * digested.  The return value will never exceed in_data_size.  If this
+ * returns a value smaller than in_data_size, then a compressed block was
+ * produced and must be released with liblzx_compress_release_next_block
+ * before more data can be added.
+ */
+size_t
+liblzx_compress_add_input(liblzx_compressor_t *stream, const void *in_data,
+                          size_t in_data_size);
+
+/* Returns the next compressed chunk.  This doesn't consume the chunk in the
+ * process, so repeated calls will keep returning the same chunk.  If no chunk
+ * is available, returns NULL.
+ */
+const liblzx_output_chunk_t *
+liblzx_compress_get_next_chunk(const liblzx_compressor_t *stream);
+
+/* Releases the next compressed chunk, allowing compression to continue. */
+void
+liblzx_compress_release_next_chunk(liblzx_compressor_t *stream);
+
+/* Ends the compression stream. */
+void
+liblzx_compress_end_input(liblzx_compressor_t *stream);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
diff --git a/dlls/cabinet/liblzx_bitops.h b/dlls/cabinet/liblzx_bitops.h
new file mode 100644
index 00000000000..976a294fe54
--- /dev/null
+++ b/dlls/cabinet/liblzx_bitops.h
@@ -0,0 +1,156 @@
+/*
+ * bitops.h - inline functions for bit manipulation
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _LIBLZX_BITOPS_H
+#define _LIBLZX_BITOPS_H
+
+#include "liblzx_compiler.h"
+#include "liblzx_types.h"
+
+#if LIBLZX_IS_MSVC_COMPILER
+#include <intrin.h>
+#endif
+
+/*
+ * Bit Scan Reverse (BSR) - find the 0-based index (relative to the least
+ * significant bit) of the *most* significant 1 bit in the input value.  The
+ * input value must be nonzero!
+ */
+
+static attrib_forceinline unsigned
+bsr32(uint32_t v)
+{
+#if LIBLZX_IS_MSVC_COMPILER
+        unsigned long result;
+        _BitScanReverse(&result, v);
+        return result;
+#else
+        return 31 - __builtin_clz(v);
+#endif
+}
+
+static attrib_forceinline unsigned
+bsr64(uint64_t v)
+{
+#if LIBLZX_IS_MSVC_COMPILER
+#  ifdef _M_AMD64
+        unsigned long result;
+        _BitScanReverse64(&result, v);
+        return result;
+#  else
+        unsigned long index;
+        if (_BitScanReverse(&index, v >> 32))
+                return index + 32;
+
+        _BitScanReverse(&index, v & 0xffffffffu);
+
+        return index;
+#  endif
+#else
+        return 63 - __builtin_clzll(v);
+#endif
+}
+
+static attrib_forceinline unsigned
+bsrw(machine_word_t v)
+{
+        STATIC_ASSERT(WORDBITS == 32 || WORDBITS == 64);
+        if (WORDBITS == 32)
+                return bsr32((uint32_t)v);
+        else
+                return bsr64(v);
+}
+
+/*
+ * Bit Scan Forward (BSF) - find the 0-based index (relative to the least
+ * significant bit) of the *least* significant 1 bit in the input value.  The
+ * input value must be nonzero!
+ */
+
+static attrib_forceinline unsigned
+bsf32(uint32_t v)
+{
+#if LIBLZX_IS_MSVC_COMPILER
+        unsigned long result;
+        _BitScanForward(&result, v);
+        return result;
+#else
+        return __builtin_ctz(v);
+#endif
+}
+
+static attrib_forceinline unsigned
+bsf64(uint64_t v)
+{
+#if LIBLZX_IS_MSVC_COMPILER
+#  ifdef _M_AMD64
+        unsigned long result;
+        _BitScanForward64(&result, v);
+        return result;
+#  else
+        unsigned long index;
+        if (_BitScanForward(&index, v & 0xffffffffu))
+                return index;
+
+        if (_BitScanForward(&index, v >> 32))
+                index += 32;
+
+        return -1;
+#  endif
+#else
+        return __builtin_ctzll(v);
+#endif
+}
+
+static attrib_forceinline unsigned
+bsfw(machine_word_t v)
+{
+        STATIC_ASSERT(WORDBITS == 32 || WORDBITS == 64);
+        if (WORDBITS == 32)
+                return bsf32(v);
+        else
+                return bsf64(v);
+}
+
+/* Return the log base 2 of 'n', rounded up to the nearest integer. */
+static attrib_forceinline unsigned
+ilog2_ceil(size_t n)
+{
+        if (n <= 1)
+                return 0;
+        return 1 + bsrw(n - 1);
+}
+
+/* Round 'n' up to the nearest power of 2 */
+static attrib_forceinline size_t
+roundup_pow_of_2(size_t n)
+{
+        return (size_t)1 << ilog2_ceil(n);
+}
+
+#endif /* _LIBLZX_BITOPS_H */
diff --git a/dlls/cabinet/liblzx_bt_matchfinder.h b/dlls/cabinet/liblzx_bt_matchfinder.h
new file mode 100644
index 00000000000..d601e110aca
--- /dev/null
+++ b/dlls/cabinet/liblzx_bt_matchfinder.h
@@ -0,0 +1,446 @@
+/*
+ * bt_matchfinder.h - Lempel-Ziv matchfinding with a hash table of binary trees
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * ----------------------------------------------------------------------------
+ *
+ * This is a Binary Trees (bt) based matchfinder.
+ *
+ * The main data structure is a hash table where each hash bucket contains a
+ * binary tree of sequences whose first 4 bytes share the same hash code.  Each
+ * sequence is identified by its starting position in the input buffer.  Each
+ * binary tree is always sorted such that each left child represents a sequence
+ * lexicographically lesser than its parent and each right child represents a
+ * sequence lexicographically greater than its parent.
+ *
+ * The algorithm processes the input buffer sequentially.  At each byte
+ * position, the hash code of the first 4 bytes of the sequence beginning at
+ * that position (the sequence being matched against) is computed.  This
+ * identifies the hash bucket to use for that position.  Then, a new binary tree
+ * node is created to represent the current sequence.  Then, in a single tree
+ * traversal, the hash bucket's binary tree is searched for matches and is
+ * re-rooted at the new node.
+ *
+ * Compared to the simpler algorithm that uses linked lists instead of binary
+ * trees (see hc_matchfinder.h), the binary tree version gains more information
+ * at each node visitation.  Ideally, the binary tree version will examine only
+ * 'log(n)' nodes to find the same matches that the linked list version will
+ * find by examining 'n' nodes.  In addition, the binary tree version can
+ * examine fewer bytes at each node by taking advantage of the common prefixes
+ * that result from the sort order, whereas the linked list version may have to
+ * examine up to the full length of the match at each node.
+ *
+ * However, it is not always best to use the binary tree version.  It requires
+ * nearly twice as much memory as the linked list version, and it takes time to
+ * keep the binary trees sorted, even at positions where the compressor does not
+ * need matches.  Generally, when doing fast compression on small buffers,
+ * binary trees are the wrong approach.  They are best suited for thorough
+ * compression and/or large buffers.
+ *
+ * ----------------------------------------------------------------------------
+ */
+
+
+#include <string.h>
+
+#include "liblzx_matchfinder_common.h"
+
+#define BT_MATCHFINDER_HASH3_ORDER 15
+#define BT_MATCHFINDER_HASH3_WAYS  2
+#define BT_MATCHFINDER_HASH4_ORDER 16
+
+/* TEMPLATED functions and structures have MF_SUFFIX appended to their name.  */
+#undef TEMPLATED
+#define TEMPLATED(name)                CONCAT(name, MF_SUFFIX)
+
+#ifndef _LIBLZX_BT_MATCHFINDER_H
+#define _LIBLZX_BT_MATCHFINDER_H
+
+/* Non-templated definitions  */
+
+/* Representation of a match found by the bt_matchfinder  */
+struct lz_match {
+
+        /* The number of bytes matched.  */
+        uint32_t length;
+
+        /* The offset back from the current position that was matched.  */
+        uint32_t offset;
+};
+
+#endif /* _LIBLZX_BT_MATCHFINDER_H */
+
+struct TEMPLATED(bt_matchfinder) {
+
+        /* The hash table for finding length 2 matches, if enabled  */
+#ifdef BT_MATCHFINDER_HASH2_ORDER
+        mf_pos_t hash2_tab[1UL << BT_MATCHFINDER_HASH2_ORDER];
+#endif
+
+        /* The hash table for finding length 3 matches  */
+        mf_pos_t hash3_tab[1UL << BT_MATCHFINDER_HASH3_ORDER][BT_MATCHFINDER_HASH3_WAYS];
+
+        /* The hash table which contains the roots of the binary trees for
+         * finding length 4+ matches  */
+        mf_pos_t hash4_tab[1UL << BT_MATCHFINDER_HASH4_ORDER];
+
+        /* The child node references for the binary trees.  The left and right
+         * children of the node for the sequence with position 'pos' are
+         * 'child_tab[pos * 2]' and 'child_tab[pos * 2 + 1]', respectively.  */
+        mf_pos_t child_tab[];
+};
+
+static attrib_forceinline bool
+TEMPLATED(matchfinder_is_valid_pos)(mf_pos_t pos, mf_pos_t min_pos)
+{
+        return ((pos + 1) & MF_INVALID_POS) > min_pos;
+}
+
+static attrib_forceinline void
+TEMPLATED(matchfinder_rebase)(mf_pos_t * mf_base, size_t count,
+                              mf_pos_t cull_amount)
+{
+        /* The invalid value points to the last element of the buffer. */
+        /* Since no match can start from that byte, it is always invalid. */
+        while (count > 0) {
+                mf_pos_t pos = *mf_base;
+
+                if (pos < cull_amount || pos == MF_INVALID_POS) {
+                        *mf_base = MF_INVALID_POS;
+                } else {
+                        *mf_base -= cull_amount;
+                }
+
+                mf_base++;
+                count--;
+        }
+}
+
+/* Return the number of bytes that must be allocated for a 'bt_matchfinder' that
+ * can work with buffers up to the specified size.  */
+static attrib_forceinline size_t
+TEMPLATED(bt_matchfinder_size)(size_t max_bufsize, bool streaming)
+{
+        const size_t streaming_bufsize_mul = streaming ? 4 : 2;
+
+        const size_t base_size =
+            sizeof(struct TEMPLATED(bt_matchfinder)) +
+            (streaming_bufsize_mul * max_bufsize * sizeof(mf_pos_t));
+
+        return base_size;
+}
+
+/* Prepare the matchfinder for a new input buffer.  */
+static attrib_forceinline void
+TEMPLATED(bt_matchfinder_init)(struct TEMPLATED(bt_matchfinder) *mf)
+{
+        memset(mf, 0xFF, sizeof(*mf));
+}
+
+static attrib_forceinline mf_pos_t *
+TEMPLATED(bt_left_child)(struct TEMPLATED(bt_matchfinder) *mf, uint32_t node)
+{
+        return &mf->child_tab[(node << 1) + 0];
+}
+
+static attrib_forceinline mf_pos_t *
+TEMPLATED(bt_right_child)(struct TEMPLATED(bt_matchfinder) *mf, uint32_t node)
+{
+        return &mf->child_tab[(node << 1) + 1];
+}
+
+/* The minimum permissible value of 'max_len' for bt_matchfinder_get_matches()
+ * and bt_matchfinder_skip_byte().  There must be sufficiently many bytes
+ * remaining to load a 32-bit integer from the *next* position.  */
+#define BT_MATCHFINDER_REQUIRED_NBYTES        5
+
+/* Advance the binary tree matchfinder by one byte, optionally recording
+ * matches.  @record_matches should be a compile-time constant.  */
+static attrib_forceinline struct lz_match *
+TEMPLATED(bt_matchfinder_advance_one_byte)(struct TEMPLATED(bt_matchfinder) * const mf,
+                                           const uint8_t * const in_begin,
+                                           mf_pos_t in_min_pos,
+                                           const ptrdiff_t cur_pos,
+                                           const uint32_t max_find_len,
+                                           const uint32_t max_produce_len,
+                                           const uint32_t nice_len,
+                                           const uint32_t max_search_depth,
+                                           uint32_t * const next_hashes,
+                                           uint32_t * const best_len_ret,
+                                           struct lz_match *lz_matchptr,
+                                           const bool record_matches)
+{
+        const uint8_t *in_next = in_begin + cur_pos;
+        uint32_t depth_remaining = max_search_depth;
+        uint32_t next_hashseq;
+        uint32_t hash3;
+        uint32_t hash4;
+#ifdef BT_MATCHFINDER_HASH2_ORDER
+        uint16_t seq2;
+        uint32_t hash2;
+#endif
+        STATIC_ASSERT(BT_MATCHFINDER_HASH3_WAYS >= 1 &&
+                      BT_MATCHFINDER_HASH3_WAYS <= 2);
+        uint32_t cur_node;
+#if BT_MATCHFINDER_HASH3_WAYS >= 2
+        uint32_t cur_node_2;
+#endif
+        const uint8_t *matchptr;
+        mf_pos_t *pending_lt_ptr, *pending_gt_ptr;
+        uint32_t best_lt_len, best_gt_len;
+        uint32_t len;
+        uint32_t best_len = 3;
+
+        next_hashseq = get_unaligned_le32(in_next + 1);
+
+        hash3 = next_hashes[0];
+        hash4 = next_hashes[1];
+
+        next_hashes[0] = lz_hash(next_hashseq & 0xFFFFFF, BT_MATCHFINDER_HASH3_ORDER);
+        next_hashes[1] = lz_hash(next_hashseq, BT_MATCHFINDER_HASH4_ORDER);
+        prefetchw(&mf->hash3_tab[next_hashes[0]]);
+        prefetchw(&mf->hash4_tab[next_hashes[1]]);
+
+#ifdef BT_MATCHFINDER_HASH2_ORDER
+        seq2 = load_u16_unaligned(in_next);
+        hash2 = lz_hash(seq2, BT_MATCHFINDER_HASH2_ORDER);
+        cur_node = mf->hash2_tab[hash2];
+        mf->hash2_tab[hash2] = cur_pos;
+        if (record_matches &&
+            TEMPLATED(matchfinder_is_valid_pos)(cur_node, in_min_pos) &&
+            seq2 == load_u16_unaligned(&in_begin[cur_node]))
+        {
+                lz_matchptr->length = 2;
+                lz_matchptr->offset = in_next - &in_begin[cur_node];
+                lz_matchptr++;
+        }
+#endif
+
+        cur_node = mf->hash3_tab[hash3][0];
+        mf->hash3_tab[hash3][0] = cur_pos;
+#if BT_MATCHFINDER_HASH3_WAYS >= 2
+        cur_node_2 = mf->hash3_tab[hash3][1];
+        mf->hash3_tab[hash3][1] = cur_node;
+#endif
+        if (record_matches &&
+            TEMPLATED(matchfinder_is_valid_pos)(cur_node, in_min_pos)) {
+                uint32_t seq3 = load_u24_unaligned(in_next);
+                if (seq3 == load_u24_unaligned(&in_begin[cur_node]) &&
+                        likely(cur_node >= in_min_pos)) {
+                        lz_matchptr->length = 3;
+                        lz_matchptr->offset = in_next - &in_begin[cur_node];
+                        lz_matchptr++;
+                }
+        #if BT_MATCHFINDER_HASH3_WAYS >= 2
+                else if (TEMPLATED(matchfinder_is_valid_pos)(cur_node_2,
+                                                             in_min_pos) &&
+                        seq3 == load_u24_unaligned(&in_begin[cur_node_2])) {
+                        lz_matchptr->length = 3;
+                        lz_matchptr->offset = in_next - &in_begin[cur_node_2];
+                        lz_matchptr++;
+                }
+        #endif
+        }
+
+        cur_node = mf->hash4_tab[hash4];
+        mf->hash4_tab[hash4] = cur_pos;
+
+        pending_lt_ptr = TEMPLATED(bt_left_child)(mf, cur_pos);
+        pending_gt_ptr = TEMPLATED(bt_right_child)(mf, cur_pos);
+
+        if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node, in_min_pos)) {
+                *pending_lt_ptr = MF_INVALID_POS;
+                *pending_gt_ptr = MF_INVALID_POS;
+                *best_len_ret = best_len;
+                return lz_matchptr;
+        }
+
+        best_lt_len = 0;
+        best_gt_len = 0;
+        len = 0;
+
+        for (;;) {
+                matchptr = &in_begin[cur_node];
+
+                if (matchptr[len] == in_next[len]) {
+                        len = lz_extend(in_next, matchptr, len + 1, max_find_len);
+                        if (!record_matches || len > best_len) {
+                                if (record_matches) {
+                                        best_len = len;
+                                        lz_matchptr->length = min_u32(len, max_produce_len);
+                                        lz_matchptr->offset =
+                                                in_next - matchptr;
+                                        lz_matchptr++;
+                                }
+                                if (len >= nice_len) {
+                                        *pending_lt_ptr =
+                                                *TEMPLATED(bt_left_child)(mf, cur_node);
+                                        *pending_gt_ptr =
+                                                *TEMPLATED(bt_right_child)(mf, cur_node);
+                                        *best_len_ret = best_len;
+                                        return lz_matchptr;
+                                }
+                        }
+                }
+
+                if (matchptr[len] < in_next[len]) {
+                        *pending_lt_ptr = cur_node;
+                        pending_lt_ptr = TEMPLATED(bt_right_child)(mf, cur_node);
+                        cur_node = *pending_lt_ptr;
+                        best_lt_len = len;
+                        if (best_gt_len < len)
+                                len = best_gt_len;
+                } else {
+                        *pending_gt_ptr = cur_node;
+                        pending_gt_ptr = TEMPLATED(bt_left_child)(mf, cur_node);
+                        cur_node = *pending_gt_ptr;
+                        best_gt_len = len;
+                        if (best_lt_len < len)
+                                len = best_lt_len;
+                }
+
+                if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node,
+                                                         in_min_pos) ||
+                    !--depth_remaining) {
+                        *pending_lt_ptr = MF_INVALID_POS;
+                        *pending_gt_ptr = MF_INVALID_POS;
+                        *best_len_ret = best_len;
+
+                        return lz_matchptr;
+                }
+        }
+}
+
+/*
+ * Retrieve a list of matches with the current position.
+ *
+ * @mf
+ *        The matchfinder structure.
+ * @in_begin
+ *        Pointer to the beginning of the input buffer.
+ * @in_abs_pos
+ *        Absolute position of in_begin in the file
+ * @cur_pos
+ *        The current position in the input buffer relative to @in_begin (the
+ *        position of the sequence being matched against).
+ * @max_len
+ *        The maximum permissible match length at this position.  Must be >=
+ *        BT_MATCHFINDER_REQUIRED_NBYTES.
+ * @nice_len
+ *        Stop searching if a match of at least this length is found.
+ *        Must be <= @max_len.
+ * @max_search_depth
+ *        Limit on the number of potential matches to consider.  Must be >= 1.
+ * @next_hashes
+ *        The precomputed hash codes for the sequence beginning at @in_next.
+ *        These will be used and then updated with the precomputed hashcodes for
+ *        the sequence beginning at @in_next + 1.
+ * @best_len_ret
+ *        If a match of length >= 4 was found, then the length of the longest such
+ *        match is written here; otherwise 3 is written here.  (Note: this is
+ *        redundant with the 'struct lz_match' array, but this is easier for the
+ *        compiler to optimize when inlined and the caller immediately does a
+ *        check against 'best_len'.)
+ * @lz_matchptr
+ *        An array in which this function will record the matches.  The recorded
+ *        matches will be sorted by strictly increasing length and (non-strictly)
+ *        increasing offset.  The maximum number of matches that may be found is
+ *        'nice_len - 1', or one less if length 2 matches are disabled.
+ *
+ * The return value is a pointer to the next available slot in the @lz_matchptr
+ * array.  (If no matches were found, this will be the same as @lz_matchptr.)
+ */
+static attrib_forceinline struct lz_match *
+TEMPLATED(bt_matchfinder_get_matches)(struct TEMPLATED(bt_matchfinder) *mf,
+                                      const uint8_t *in_begin,
+                                      uint32_t in_min_pos,
+                                      ptrdiff_t cur_pos,
+                                      uint32_t max_find_len,
+                                      uint32_t max_produce_len,
+                                      uint32_t nice_len,
+                                      uint32_t max_search_depth,
+                                      uint32_t next_hashes[2],
+                                      uint32_t *best_len_ret,
+                                      struct lz_match *lz_matchptr)
+{
+        return TEMPLATED(bt_matchfinder_advance_one_byte)(mf,
+                                                          in_begin,
+                                                          in_min_pos,
+                                                          cur_pos,
+                                                          max_find_len,
+                                                          max_produce_len,
+                                                          nice_len,
+                                                          max_search_depth,
+                                                          next_hashes,
+                                                          best_len_ret,
+                                                          lz_matchptr,
+                                                          true);
+}
+
+/*
+ * Advance the matchfinder, but don't record any matches.
+ *
+ * This is very similar to bt_matchfinder_get_matches() because both functions
+ * must do hashing and tree re-rooting.
+ */
+static attrib_forceinline void
+TEMPLATED(bt_matchfinder_skip_byte)(struct TEMPLATED(bt_matchfinder) *mf,
+                                    const uint8_t *in_begin,
+                                    uint32_t in_min_pos,
+                                    ptrdiff_t cur_pos,
+                                    uint32_t nice_len,
+                                    uint32_t max_search_depth,
+                                    uint32_t next_hashes[2])
+{
+        uint32_t best_len;
+        TEMPLATED(bt_matchfinder_advance_one_byte)(mf,
+                                                   in_begin,
+                                                   in_min_pos,
+                                                   cur_pos,
+                                                   nice_len,
+                                                   nice_len,
+                                                   nice_len,
+                                                   max_search_depth,
+                                                   next_hashes,
+                                                   &best_len,
+                                                   NULL,
+                                                   false);
+}
+
+/*
+ * Culls any matches that are lower than a specified offset and reduces any
+ * remaining offsets by the same amount.
+ */
+static attrib_forceinline void
+TEMPLATED(bt_matchfinder_cull)(struct TEMPLATED(bt_matchfinder) * mf,
+                               uint32_t cull_size, uint32_t window_size)
+{
+        size_t mf_size = TEMPLATED(bt_matchfinder_size)(window_size, true);
+
+        const size_t mf_count = mf_size / sizeof(mf_pos_t);
+
+        TEMPLATED(matchfinder_rebase)((mf_pos_t *)mf, mf_count, cull_size);
+}
diff --git a/dlls/cabinet/liblzx_compiler.h b/dlls/cabinet/liblzx_compiler.h
new file mode 100644
index 00000000000..73e36c07a59
--- /dev/null
+++ b/dlls/cabinet/liblzx_compiler.h
@@ -0,0 +1,214 @@
+/*
+ * compiler.h
+ *
+ * Compiler-specific definitions.  Currently, only GCC and clang are supported.
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _LIBLZX_COMPILER_H
+#define _LIBLZX_COMPILER_H
+
+#if defined(_MSC_VER) && !defined(__clang__) && !defined(__GNUC__)
+#define LIBLZX_IS_MSVC_COMPILER 1
+#else
+#define LIBLZX_IS_MSVC_COMPILER 0
+#endif
+
+#if LIBLZX_IS_MSVC_COMPILER
+#include <stdint.h>
+#include <stddef.h>
+
+#pragma warning(error:4013)
+#endif
+
+#ifndef __ORDER_LITTLE_ENDIAN__
+#define __ORDER_LITTLE_ENDIAN__ 1
+#endif
+
+#ifndef __ORDER_BIG_ENDIAN__
+#define __ORDER_BIG_ENDIAN__ 2
+#endif
+
+
+/* Is the compiler GCC of the specified version or later?  This always returns
+ * false for clang, since clang is "frozen" at GNUC 4.2.  The __has_*
+ * feature-test macros should be used to detect clang functionality instead.  */
+#define GCC_PREREQ(major, minor)                                        \
+        (!defined(__clang__) && !defined(__INTEL_COMPILER) &&                \
+         (__GNUC__ > major ||                                                \
+          (__GNUC__ == major && __GNUC_MINOR__ >= minor)))
+
+/* Feature-test macros defined by recent versions of clang.  */
+#ifndef __has_attribute
+#  define __has_attribute(attribute)    0
+#endif
+#ifndef __has_feature
+#  define __has_feature(feature)        0
+#endif
+#ifndef __has_builtin
+#  define __has_builtin(builtin)        0
+#endif
+
+/* Declare that the annotated function should always be inlined.  This might be
+ * desirable in highly tuned code, e.g. compression codecs.  */
+#if LIBLZX_IS_MSVC_COMPILER
+#define attrib_forceinline        __forceinline
+#else
+#define attrib_forceinline        inline __attribute__((always_inline))
+#endif
+
+/* Declare that the annotated function should *not* be inlined.  */
+#if LIBLZX_IS_MSVC_COMPILER
+#define attrib_noinline                __declspec(noinline)
+#else
+#define attrib_noinline                __attribute__((noinline))
+#endif
+
+/* Declare that the annotated function is unlikely to be executed */
+#if LIBLZX_IS_MSVC_COMPILER
+#define attrib_cold
+#else
+#define attrib_cold __attribute__((cold))
+#endif
+
+/* Declare that the annotated type or variable is aligned */
+#if LIBLZX_IS_MSVC_COMPILER
+#define attrib_aligned(alignment)        __declspec(align(alignment))
+#else
+#define attrib_aligned(alignment)        __attribute__((aligned(alignment)))
+#endif
+
+/* Functionally the same as 'attrib_noinline', but documents that the reason
+ * for not inlining is to prevent the annotated function from being inlined
+ * into a recursive function, thereby increasing its stack usage.  */
+#define attrib_noinline_for_stack attrib_noinline
+
+/* Hint that the expression is usually true.  */
+#if LIBLZX_IS_MSVC_COMPILER
+#define likely(expr)                (expr)
+#else
+#define likely(expr)                __builtin_expect(!!(expr), 1)
+#endif
+
+/* Hint that the expression is usually false.  */
+#if LIBLZX_IS_MSVC_COMPILER
+#define unlikely(expr)                (expr)
+#else
+#define unlikely(expr)                __builtin_expect(!!(expr), 0)
+#endif
+
+/* Prefetch into L1 cache for read.  */
+#if LIBLZX_IS_MSVC_COMPILER
+#define prefetchr(addr)                _mm_prefetch((const char *)(addr), _MM_HINT_T0)
+#else
+#define prefetchr(addr)                __builtin_prefetch((addr), 0)
+#endif
+
+/* Prefetch into L1 cache for write.  */
+#if LIBLZX_IS_MSVC_COMPILER
+#define prefetchw(addr)                _mm_prefetch((const char *)(addr), _MM_HINT_T0)
+#else
+#define prefetchw(addr)                __builtin_prefetch((addr), 1)
+#endif
+
+/* Hint that the annotated function takes a printf()-like format string and
+ * arguments.  This is currently disabled on Windows because MinGW does not
+ * support this attribute on functions taking wide-character strings.  */
+#ifdef _WIN32
+#  define _format_attribute(type, format_str, format_start)
+#else
+#  define _format_attribute(type, format_str, format_start)        \
+                        __attribute__((format(type, format_str, format_start)))
+#endif
+
+/* Endianness definitions.  Either CPU_IS_BIG_ENDIAN() or CPU_IS_LITTLE_ENDIAN()
+ * evaluates to 1.  The other evaluates to 0.  Note that newer gcc supports
+ * __BYTE_ORDER__ for easily determining the endianness; older gcc doesn't.  In
+ * the latter case we fall back to a configure-time check.  */
+#ifdef __BYTE_ORDER__
+#  define CPU_IS_BIG_ENDIAN()        (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
+#elif defined(HAVE_CONFIG_H)
+#  include "config.h"
+#  ifdef WORDS_BIGENDIAN
+#    define CPU_IS_BIG_ENDIAN()        1
+#  else
+#    define CPU_IS_BIG_ENDIAN()        0
+#  endif
+#endif
+#define CPU_IS_LITTLE_ENDIAN() (!CPU_IS_BIG_ENDIAN())
+
+/* UNALIGNED_ACCESS_IS_FAST should be defined to 1 if unaligned memory accesses
+ * can be performed efficiently on the target platform.  */
+#if defined(__x86_64__) || defined(__i386__) || \
+        defined(__ARM_FEATURE_UNALIGNED) || defined(__powerpc64__)
+#  define UNALIGNED_ACCESS_IS_FAST 1
+#else
+#  define UNALIGNED_ACCESS_IS_FAST 0
+#endif
+
+/* Swap the values of two variables, without multiple evaluation.  */
+#ifndef swap
+#  define swap(a, b) do { typeof(a) _a = (a); (a) = (b); (b) = _a; } while(0)
+#endif
+#define SWAP(a, b)        swap((a), (b))
+
+/* Optional definitions for checking with 'sparse'.  */
+#ifdef __CHECKER__
+#  define _bitwise_attr        __attribute__((bitwise))
+#  define _force_attr        __attribute__((force))
+#else
+#  define _bitwise_attr
+#  define _force_attr
+#endif
+
+/* STATIC_ASSERT() - verify the truth of an expression at compilation time.  */
+#ifdef __CHECKER__
+#  define STATIC_ASSERT(expr)
+#  define STATIC_ASSERT_STMT(expr)
+#elif __STDC_VERSION__ >= 201112L
+#  define STATIC_ASSERT(expr)        _Static_assert((expr), "")
+#  define STATIC_ASSERT_STMT(expr)   do {_Static_assert((expr), "");} while(0)
+#else
+#  define STATIC_ASSERT(expr)        ((void)sizeof(char[1 - 2 * !(expr)]))
+#  define STATIC_ASSERT_STMT(expr)   STATIC_ASSERT(expr)
+#endif
+
+/* STATIC_ASSERT_ZERO() - verify the truth of an expression at compilation time
+ * and also produce a result of value '0' to be used in constant expressions */
+#define STATIC_ASSERT_ZERO(expr) ((int)sizeof(char[-!(expr)]))
+
+#define CONCAT_IMPL(s1, s2)        s1##s2
+
+/* CONCAT() - concatenate two tokens at preprocessing time.  */
+#define CONCAT(s1, s2)                CONCAT_IMPL(s1, s2)
+
+#if LIBLZX_IS_MSVC_COMPILER
+#define __builtin_constant_p(n) (0)
+
+typedef ptrdiff_t ssize_t;
+#endif
+
+#endif /* _LIBLZX_COMPILER_H */
diff --git a/dlls/cabinet/liblzx_compress_common.c b/dlls/cabinet/liblzx_compress_common.c
new file mode 100644
index 00000000000..be88df26b85
--- /dev/null
+++ b/dlls/cabinet/liblzx_compress_common.c
@@ -0,0 +1,673 @@
+/*
+ * compress_common.c
+ *
+ * Code for compression shared among multiple compression formats.
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifdef HAVE_CONFIG_H
+#  include "config.h"
+#endif
+
+#include <string.h>
+
+#include <assert.h>
+#include "liblzx_compress_common.h"
+#include "liblzx_minmax.h"
+#include "liblzx_util.h"
+
+/*
+ * Given the binary tree node A[subtree_idx] whose children already satisfy the
+ * maxheap property, swap the node with its greater child until it is greater
+ * than or equal to both of its children, so that the maxheap property is
+ * satisfied in the subtree rooted at A[subtree_idx].  'A' uses 1-based indices.
+ */
+static void
+heapify_subtree(uint32_t A[], unsigned length, unsigned subtree_idx)
+{
+        unsigned parent_idx;
+        unsigned child_idx;
+        uint32_t v;
+
+        v = A[subtree_idx];
+        parent_idx = subtree_idx;
+        while ((child_idx = parent_idx * 2) <= length) {
+                if (child_idx < length && A[child_idx + 1] > A[child_idx])
+                        child_idx++;
+                if (v >= A[child_idx])
+                        break;
+                A[parent_idx] = A[child_idx];
+                parent_idx = child_idx;
+        }
+        A[parent_idx] = v;
+}
+
+/*
+ * Rearrange the array 'A' so that it satisfies the maxheap property.
+ * 'A' uses 1-based indices, so the children of A[i] are A[i*2] and A[i*2 + 1].
+ */
+static void
+heapify_array(uint32_t A[], unsigned length)
+{
+        unsigned subtree_idx;
+
+        for (subtree_idx = length / 2; subtree_idx >= 1; subtree_idx--)
+                heapify_subtree(A, length, subtree_idx);
+}
+
+/*
+ * Sort the array 'A', which contains 'length' unsigned 32-bit integers.
+ *
+ * Note: name this function heap_sort() instead of heapsort() to avoid colliding
+ * with heapsort() from stdlib.h on BSD-derived systems --- though this isn't
+ * necessary when compiling with -D_ANSI_SOURCE, which is the better solution.
+ */
+static void
+heap_sort(uint32_t A[], unsigned length)
+{
+        A--; /* Use 1-based indices  */
+
+        heapify_array(A, length);
+
+        while (length >= 2) {
+                uint32_t tmp = A[length];
+
+                A[length] = A[1];
+                A[1] = tmp;
+                length--;
+                heapify_subtree(A, length, 1);
+        }
+}
+
+#define NUM_SYMBOL_BITS 10
+#define NUM_FREQ_BITS        (32 - NUM_SYMBOL_BITS)
+#define SYMBOL_MASK        ((1 << NUM_SYMBOL_BITS) - 1)
+#define FREQ_MASK        (~SYMBOL_MASK)
+
+#define GET_NUM_COUNTERS(num_syms)        (num_syms)
+
+/*
+ * Sort the symbols primarily by frequency and secondarily by symbol value.
+ * Discard symbols with zero frequency and fill in an array with the remaining
+ * symbols, along with their frequencies.  The low NUM_SYMBOL_BITS bits of each
+ * array entry will contain the symbol value, and the remaining bits will
+ * contain the frequency.
+ *
+ * @num_syms
+ *        Number of symbols in the alphabet, at most 1 << NUM_SYMBOL_BITS.
+ *
+ * @freqs[num_syms]
+ *        Frequency of each symbol, summing to at most (1 << NUM_FREQ_BITS) - 1.
+ *
+ * @lens[num_syms]
+ *        An array that eventually will hold the length of each codeword.  This
+ *        function only fills in the codeword lengths for symbols that have zero
+ *        frequency, which are not well defined per se but will be set to 0.
+ *
+ * @symout[num_syms]
+ *        The output array, described above.
+ *
+ * Returns the number of entries in 'symout' that were filled.  This is the
+ * number of symbols that have nonzero frequency.
+ */
+static unsigned
+sort_symbols(unsigned num_syms, const uint32_t freqs[], uint8_t lens[], uint32_t symout[])
+{
+        unsigned sym;
+        unsigned i;
+        unsigned num_used_syms;
+        unsigned num_counters;
+        unsigned counters[GET_NUM_COUNTERS(MAX_NUM_SYMS)];
+
+        /*
+         * We use heapsort, but with an added optimization.  Since often most
+         * symbol frequencies are low, we first do a count sort using a limited
+         * number of counters.  High frequencies are counted in the last
+         * counter, and only they will be sorted with heapsort.
+         *
+         * Note: with more symbols, it is generally beneficial to have more
+         * counters.  About 1 counter per symbol seems fastest.
+         */
+
+        num_counters = GET_NUM_COUNTERS(num_syms);
+
+        memset(counters, 0, num_counters * sizeof(counters[0]));
+
+        /* Count the frequencies. */
+        for (sym = 0; sym < num_syms; sym++)
+                counters[min_size(freqs[sym], num_counters - 1)]++;
+
+        /*
+         * Make the counters cumulative, ignoring the zero-th, which counted
+         * symbols with zero frequency.  As a side effect, this calculates the
+         * number of symbols with nonzero frequency.
+         */
+        num_used_syms = 0;
+        for (i = 1; i < num_counters; i++) {
+                unsigned count = counters[i];
+
+                counters[i] = num_used_syms;
+                num_used_syms += count;
+        }
+
+        /*
+         * Sort nonzero-frequency symbols using the counters.  At the same time,
+         * set the codeword lengths of zero-frequency symbols to 0.
+         */
+        for (sym = 0; sym < num_syms; sym++) {
+                uint32_t freq = freqs[sym];
+
+                if (freq != 0) {
+                        symout[counters[min_size(freq, num_counters - 1)]++] =
+                                sym | (freq << NUM_SYMBOL_BITS);
+                } else {
+                        lens[sym] = 0;
+                }
+        }
+
+        /* Sort the symbols counted in the last counter. */
+        heap_sort(symout + counters[num_counters - 2],
+                  counters[num_counters - 1] - counters[num_counters - 2]);
+
+        return num_used_syms;
+}
+
+/*
+ * Build a Huffman tree.
+ *
+ * This is an optimized implementation that
+ *        (a) takes advantage of the frequencies being already sorted;
+ *        (b) only generates non-leaf nodes, since the non-leaf nodes of a Huffman
+ *            tree are sufficient to generate a canonical code;
+ *        (c) Only stores parent pointers, not child pointers;
+ *        (d) Produces the nodes in the same memory used for input frequency
+ *            information.
+ *
+ * Array 'A', which contains 'sym_count' entries, is used for both input and
+ * output.  For this function, 'sym_count' must be at least 2.
+ *
+ * For input, the array must contain the frequencies of the symbols, sorted in
+ * increasing order.  Specifically, each entry must contain a frequency left
+ * shifted by NUM_SYMBOL_BITS bits.  Any data in the low NUM_SYMBOL_BITS bits of
+ * the entries will be ignored by this function.  Although these bits will, in
+ * fact, contain the symbols that correspond to the frequencies, this function
+ * is concerned with frequencies only and keeps the symbols as-is.
+ *
+ * For output, this function will produce the non-leaf nodes of the Huffman
+ * tree.  These nodes will be stored in the first (sym_count - 1) entries of the
+ * array.  Entry A[sym_count - 2] will represent the root node.  Each other node
+ * will contain the zero-based index of its parent node in 'A', left shifted by
+ * NUM_SYMBOL_BITS bits.  The low NUM_SYMBOL_BITS bits of each entry in A will
+ * be kept as-is.  Again, note that although these low bits will, in fact,
+ * contain a symbol value, this symbol will have *no relationship* with the
+ * Huffman tree node that happens to occupy the same slot.  This is because this
+ * implementation only generates the non-leaf nodes of the tree.
+ */
+static void
+build_tree(uint32_t A[], unsigned sym_count)
+{
+        const unsigned last_idx = sym_count - 1;
+
+        /* Index of the next lowest frequency leaf that still needs a parent */
+        unsigned i = 0;
+
+        /*
+         * Index of the next lowest frequency non-leaf that still needs a
+         * parent, or 'e' if there is currently no such node
+         */
+        unsigned b = 0;
+
+        /* Index of the next spot for a non-leaf (will overwrite a leaf) */
+        unsigned e = 0;
+
+        do {
+                uint32_t new_freq;
+
+                /*
+                 * Select the next two lowest frequency nodes among the leaves
+                 * A[i] and non-leaves A[b], and create a new node A[e] to be
+                 * their parent.  Set the new node's frequency to the sum of the
+                 * frequencies of its two children.
+                 *
+                 * Usually the next two lowest frequency nodes are of the same
+                 * type (leaf or non-leaf), so check those cases first.
+                 */
+                if (i + 1 <= last_idx &&
+                    (b == e || (A[i + 1] & FREQ_MASK) <= (A[b] & FREQ_MASK))) {
+                        /* Two leaves */
+                        new_freq = (A[i] & FREQ_MASK) + (A[i + 1] & FREQ_MASK);
+                        i += 2;
+                } else if (b + 2 <= e &&
+                           (i > last_idx ||
+                            (A[b + 1] & FREQ_MASK) < (A[i] & FREQ_MASK))) {
+                        /* Two non-leaves */
+                        new_freq = (A[b] & FREQ_MASK) + (A[b + 1] & FREQ_MASK);
+                        A[b] = (e << NUM_SYMBOL_BITS) | (A[b] & SYMBOL_MASK);
+                        A[b + 1] = (e << NUM_SYMBOL_BITS) |
+                                   (A[b + 1] & SYMBOL_MASK);
+                        b += 2;
+                } else {
+                        /* One leaf and one non-leaf */
+                        new_freq = (A[i] & FREQ_MASK) + (A[b] & FREQ_MASK);
+                        A[b] = (e << NUM_SYMBOL_BITS) | (A[b] & SYMBOL_MASK);
+                        i++;
+                        b++;
+                }
+                A[e] = new_freq | (A[e] & SYMBOL_MASK);
+                /*
+                 * A binary tree with 'n' leaves has 'n - 1' non-leaves, so the
+                 * tree is complete once we've created 'n - 1' non-leaves.
+                 */
+        } while (++e < last_idx);
+}
+
+/*
+ * Given the stripped-down Huffman tree constructed by build_tree(), determine
+ * the number of codewords that should be assigned each possible length, taking
+ * into account the length-limited constraint.
+ *
+ * @A
+ *        The array produced by build_tree(), containing parent index information
+ *        for the non-leaf nodes of the Huffman tree.  Each entry in this array is
+ *        a node; a node's parent always has a greater index than that node
+ *        itself.  This function will overwrite the parent index information in
+ *        this array, so essentially it will destroy the tree.  However, the data
+ *        in the low NUM_SYMBOL_BITS of each entry will be preserved.
+ *
+ * @root_idx
+ *        The 0-based index of the root node in 'A', and consequently one less
+ *        than the number of tree node entries in 'A'.  (Or, really 2 less than
+ *        the actual length of 'A'.)
+ *
+ * @len_counts
+ *        An array of length ('max_codeword_len' + 1) in which the number of
+ *        codewords having each length <= max_codeword_len will be returned.
+ *
+ * @max_codeword_len
+ *        The maximum permissible codeword length.
+ */
+static void
+compute_length_counts(uint32_t A[], unsigned root_idx, unsigned len_counts[],
+                      unsigned max_codeword_len)
+{
+        unsigned len;
+        int node;
+
+        /*
+         * The key observations are:
+         *
+         * (1) We can traverse the non-leaf nodes of the tree, always visiting a
+         *     parent before its children, by simply iterating through the array
+         *     in reverse order.  Consequently, we can compute the depth of each
+         *     node in one pass, overwriting the parent indices with depths.
+         *
+         * (2) We can initially assume that in the real Huffman tree, both
+         *     children of the root are leaves.  This corresponds to two
+         *     codewords of length 1.  Then, whenever we visit a (non-leaf) node
+         *     during the traversal, we modify this assumption to account for
+         *     the current node *not* being a leaf, but rather its two children
+         *     being leaves.  This causes the loss of one codeword for the
+         *     current depth and the addition of two codewords for the current
+         *     depth plus one.
+         *
+         * (3) We can handle the length-limited constraint fairly easily by
+         *     simply using the largest length available when a depth exceeds
+         *     max_codeword_len.
+         */
+
+        for (len = 0; len <= max_codeword_len; len++)
+                len_counts[len] = 0;
+        len_counts[1] = 2;
+
+        /* Set the root node's depth to 0. */
+        A[root_idx] &= SYMBOL_MASK;
+
+        for (node = root_idx - 1; node >= 0; node--) {
+
+                /* Calculate the depth of this node. */
+
+                unsigned parent = A[node] >> NUM_SYMBOL_BITS;
+                unsigned parent_depth = A[parent] >> NUM_SYMBOL_BITS;
+                unsigned depth = parent_depth + 1;
+                unsigned len = depth;
+
+                /*
+                 * Set the depth of this node so that it is available when its
+                 * children (if any) are processed.
+                 */
+                A[node] = (A[node] & SYMBOL_MASK) | (depth << NUM_SYMBOL_BITS);
+
+                /*
+                 * If needed, decrease the length to meet the length-limited
+                 * constraint.  This is not the optimal method for generating
+                 * length-limited Huffman codes!  But it should be good enough.
+                 */
+                if (len >= max_codeword_len) {
+                        len = max_codeword_len;
+                        do {
+                                len--;
+                        } while (len_counts[len] == 0);
+                }
+
+                /*
+                 * Account for the fact that we have a non-leaf node at the
+                 * current depth.
+                 */
+                len_counts[len]--;
+                len_counts[len + 1] += 2;
+        }
+}
+
+/*
+ * Generate the codewords for a canonical Huffman code.
+ *
+ * @A
+ *        The output array for codewords.  In addition, initially this
+ *        array must contain the symbols, sorted primarily by frequency and
+ *        secondarily by symbol value, in the low NUM_SYMBOL_BITS bits of
+ *        each entry.
+ *
+ * @len
+ *        Output array for codeword lengths.
+ *
+ * @len_counts
+ *        An array that provides the number of codewords that will have
+ *        each possible length <= max_codeword_len.
+ *
+ * @max_codeword_len
+ *        Maximum length, in bits, of each codeword.
+ *
+ * @num_syms
+ *        Number of symbols in the alphabet, including symbols with zero
+ *        frequency.  This is the length of the 'A' and 'len' arrays.
+ */
+static void
+gen_codewords(uint32_t A[], uint8_t lens[], const unsigned len_counts[],
+              unsigned max_codeword_len, unsigned num_syms)
+{
+        uint32_t next_codewords[MAX_CODEWORD_LEN + 1];
+        unsigned i;
+        unsigned len;
+        unsigned sym;
+
+        /*
+         * Given the number of codewords that will have each length, assign
+         * codeword lengths to symbols.  We do this by assigning the lengths in
+         * decreasing order to the symbols sorted primarily by increasing
+         * frequency and secondarily by increasing symbol value.
+         */
+        for (i = 0, len = max_codeword_len; len >= 1; len--) {
+                unsigned count = len_counts[len];
+
+                while (count--)
+                        lens[A[i++] & SYMBOL_MASK] = len;
+        }
+
+        /*
+         * Generate the codewords themselves.  We initialize the
+         * 'next_codewords' array to provide the lexicographically first
+         * codeword of each length, then assign codewords in symbol order.  This
+         * produces a canonical code.
+         */
+        next_codewords[0] = 0;
+        next_codewords[1] = 0;
+        for (len = 2; len <= max_codeword_len; len++)
+                next_codewords[len] =
+                        (next_codewords[len - 1] + len_counts[len - 1]) << 1;
+
+        for (sym = 0; sym < num_syms; sym++)
+                A[sym] = next_codewords[lens[sym]]++;
+}
+
+/*
+ * ---------------------------------------------------------------------
+ *                        make_canonical_huffman_code()
+ * ---------------------------------------------------------------------
+ *
+ * Given an alphabet and the frequency of each symbol in it, construct a
+ * length-limited canonical Huffman code.
+ *
+ * @num_syms
+ *        The number of symbols in the alphabet.  The symbols are the integers in
+ *        the range [0, num_syms - 1].  This parameter must be at least 2 and
+ *        must not exceed (1 << NUM_SYMBOL_BITS).
+ *
+ * @max_codeword_len
+ *        The maximum permissible codeword length.
+ *
+ * @freqs
+ *        An array of length @num_syms that gives the frequency of each symbol.
+ *        It is valid for some, none, or all of the frequencies to be 0.  The sum
+ *        of frequencies must not exceed (1 << NUM_FREQ_BITS) - 1.
+ *
+ * @lens
+ *        An array of @num_syms entries in which this function will return the
+ *        length, in bits, of the codeword assigned to each symbol.  Symbols with
+ *        0 frequency will not have codewords per se, but their entries in this
+ *        array will be set to 0.  No lengths greater than @max_codeword_len will
+ *        be assigned.
+ *
+ * @codewords
+ *        An array of @num_syms entries in which this function will return the
+ *        codeword for each symbol, right-justified and padded on the left with
+ *        zeroes.  Codewords for symbols with 0 frequency will be undefined.
+ *
+ * ---------------------------------------------------------------------
+ *
+ * This function builds a length-limited canonical Huffman code.
+ *
+ * A length-limited Huffman code contains no codewords longer than some
+ * specified length, and has exactly (with some algorithms) or approximately
+ * (with the algorithm used here) the minimum weighted path length from the
+ * root, given this constraint.
+ *
+ * A canonical Huffman code satisfies the properties that a longer codeword
+ * never lexicographically precedes a shorter codeword, and the lexicographic
+ * ordering of codewords of the same length is the same as the lexicographic
+ * ordering of the corresponding symbols.  A canonical Huffman code, or more
+ * generally a canonical prefix code, can be reconstructed from only a list
+ * containing the codeword length of each symbol.
+ *
+ * The classic algorithm to generate a Huffman code creates a node for each
+ * symbol, then inserts these nodes into a min-heap keyed by symbol frequency.
+ * Then, repeatedly, the two lowest-frequency nodes are removed from the
+ * min-heap and added as the children of a new node having frequency equal to
+ * the sum of its two children, which is then inserted into the min-heap.  When
+ * only a single node remains in the min-heap, it is the root of the Huffman
+ * tree.  The codeword for each symbol is determined by the path needed to reach
+ * the corresponding node from the root.  Descending to the left child appends a
+ * 0 bit, whereas descending to the right child appends a 1 bit.
+ *
+ * The classic algorithm is relatively easy to understand, but it is subject to
+ * a number of inefficiencies.  In practice, it is fastest to first sort the
+ * symbols by frequency.  (This itself can be subject to an optimization based
+ * on the fact that most frequencies tend to be low.)  At the same time, we sort
+ * secondarily by symbol value, which aids the process of generating a canonical
+ * code.  Then, during tree construction, no heap is necessary because both the
+ * leaf nodes and the unparented non-leaf nodes can be easily maintained in
+ * sorted order.  Consequently, there can never be more than two possibilities
+ * for the next-lowest-frequency node.
+ *
+ * In addition, because we're generating a canonical code, we actually don't
+ * need the leaf nodes of the tree at all, only the non-leaf nodes.  This is
+ * because for canonical code generation we don't need to know where the symbols
+ * are in the tree.  Rather, we only need to know how many leaf nodes have each
+ * depth (codeword length).  And this information can, in fact, be quickly
+ * generated from the tree of non-leaves only.
+ *
+ * Furthermore, we can build this stripped-down Huffman tree directly in the
+ * array in which the codewords are to be generated, provided that these array
+ * slots are large enough to hold a symbol and frequency value.
+ *
+ * Still furthermore, we don't even need to maintain explicit child pointers.
+ * We only need the parent pointers, and even those can be overwritten in-place
+ * with depth information as part of the process of extracting codeword lengths
+ * from the tree.  So in summary, we do NOT need a big structure like:
+ *
+ *        struct huffman_tree_node {
+ *                unsigned int symbol;
+ *                unsigned int frequency;
+ *                unsigned int depth;
+ *                struct huffman_tree_node *left_child;
+ *                struct huffman_tree_node *right_child;
+ *        };
+ *
+ *
+ * ... which often gets used in "naive" implementations of Huffman code
+ * generation.
+ *
+ * Many of these optimizations are based on the implementation in 7-Zip (source
+ * file: C/HuffEnc.c), which was placed in the public domain by Igor Pavlov.
+ *
+ * NOTE: in general, the same frequencies can be used to generate different
+ * length-limited canonical Huffman codes.  One choice we have is during tree
+ * construction, when we must decide whether to prefer a leaf or non-leaf when
+ * there is a tie in frequency.  Another choice we have is how to deal with
+ * codewords that would exceed @max_codeword_len bits in length.  Both of these
+ * choices affect the resulting codeword lengths, which otherwise can be mapped
+ * uniquely onto the resulting canonical Huffman code.
+ *
+ * Normally, there is no problem with choosing one valid code over another,
+ * provided that they produce similar compression ratios.  However, the LZMS
+ * compression format uses adaptive Huffman coding.  It requires that both the
+ * decompressor and compressor build a canonical code equivalent to that which
+ * can be generated by using the classic Huffman tree construction algorithm and
+ * always processing leaves before non-leaves when there is a frequency tie.
+ * Therefore, we make sure to do this.  This method also has the advantage of
+ * sometimes shortening the longest codeword that is generated.
+ *
+ * There also is the issue of how codewords longer than @max_codeword_len are
+ * dealt with.  Fortunately, for LZMS this is irrelevant because for the LZMS
+ * alphabets no codeword can ever exceed LZMS_MAX_CODEWORD_LEN (= 15).  Since
+ * the LZMS algorithm regularly halves all frequencies, the frequencies cannot
+ * become high enough for a length 16 codeword to be generated.  Specifically, I
+ * think that if ties are broken in favor of non-leaves (as we do), the lowest
+ * total frequency that would give a length-16 codeword would be the sum of the
+ * frequencies 1 1 1 3 4 7 11 18 29 47 76 123 199 322 521 843 1364, which is
+ * 3570.  And in LZMS we can't get a frequency that high based on the alphabet
+ * sizes, rebuild frequencies, and scaling factors.  This worst-case scenario is
+ * based on the following degenerate case (only the bottom of the tree shown):
+ *
+ *                          ...
+ *                        17
+ *                       /  \
+ *                      10   7
+ *                     / \
+ *                    6   4
+ *                   / \
+ *                  3   3
+ *                 / \
+ *                2   1
+ *               / \
+ *              1   1
+ *
+ * Excluding the first leaves (those with value 1), each leaf value must be
+ * greater than the non-leaf up 1 and down 2 from it; otherwise that leaf would
+ * have taken precedence over that non-leaf and been combined with the leaf
+ * below, thereby decreasing the height compared to that shown.
+ *
+ * Interesting fact: if we were to instead prioritize non-leaves over leaves,
+ * then the worst case frequencies would be the Fibonacci sequence, plus an
+ * extra frequency of 1.  In this hypothetical scenario, it would be slightly
+ * easier for longer codewords to be generated.
+ */
+void
+make_canonical_huffman_code(unsigned num_syms, unsigned max_codeword_len,
+                            const uint32_t freqs[], uint8_t lens[], uint32_t codewords[])
+{
+        uint32_t *A = codewords;
+        unsigned num_used_syms;
+
+        assert(num_syms <= MAX_NUM_SYMS);
+        STATIC_ASSERT_STMT(MAX_NUM_SYMS <= 1 << NUM_SYMBOL_BITS);
+        assert(max_codeword_len <= MAX_CODEWORD_LEN);
+
+        /*
+         * We begin by sorting the symbols primarily by frequency and
+         * secondarily by symbol value.  As an optimization, the array used for
+         * this purpose ('A') shares storage with the space in which we will
+         * eventually return the codewords.
+         */
+        num_used_syms = sort_symbols(num_syms, freqs, lens, A);
+
+        /*
+         * 'num_used_syms' is the number of symbols with nonzero frequency.
+         * This may be less than @num_syms.  'num_used_syms' is also the number
+         * of entries in 'A' that are valid.  Each entry consists of a distinct
+         * symbol and a nonzero frequency packed into a 32-bit integer.
+         */
+
+        /*
+         * Handle special cases where only 0 or 1 symbols were used (had nonzero
+         * frequency).
+         */
+
+        if (unlikely(num_used_syms == 0)) {
+                /*
+                 * Code is empty.  sort_symbols() already set all lengths to 0,
+                 * so there is nothing more to do.
+                 */
+                return;
+        }
+
+        if (unlikely(num_used_syms == 1)) {
+                /*
+                 * Only one symbol was used, so we only need one codeword.  But
+                 * two codewords are needed to form the smallest complete
+                 * Huffman code, which uses codewords 0 and 1.  Therefore, we
+                 * choose another symbol to which to assign a codeword.  We use
+                 * 0 (if the used symbol is not 0) or 1 (if the used symbol is
+                 * 0).  In either case, the lesser-valued symbol must be
+                 * assigned codeword 0 so that the resulting code is canonical.
+                 */
+
+                unsigned sym = A[0] & SYMBOL_MASK;
+                unsigned nonzero_idx = sym ? sym : 1;
+
+                codewords[0] = 0;
+                lens[0] = 1;
+                codewords[nonzero_idx] = 1;
+                lens[nonzero_idx] = 1;
+                return;
+        }
+
+        /*
+         * Build a stripped-down version of the Huffman tree, sharing the array
+         * 'A' with the symbol values.  Then extract length counts from the tree
+         * and use them to generate the final codewords.
+         */
+
+        build_tree(A, num_used_syms);
+
+        {
+                unsigned len_counts[MAX_CODEWORD_LEN + 1];
+
+                compute_length_counts(A, num_used_syms - 2,
+                                      len_counts, max_codeword_len);
+
+                gen_codewords(A, lens, len_counts, max_codeword_len, num_syms);
+        }
+}
diff --git a/dlls/cabinet/liblzx_compress_common.h b/dlls/cabinet/liblzx_compress_common.h
new file mode 100644
index 00000000000..5eb73086fb7
--- /dev/null
+++ b/dlls/cabinet/liblzx_compress_common.h
@@ -0,0 +1,19 @@
+/*
+ * compress_common.h
+ *
+ * Header for compression code shared by multiple compression formats.
+ */
+
+#ifndef _LIBLZX_COMPRESS_COMMON_H
+#define _LIBLZX_COMPRESS_COMMON_H
+
+#include "liblzx_types.h"
+
+#define MAX_NUM_SYMS            799        /* LZMS_MAX_NUM_SYMS */
+#define MAX_CODEWORD_LEN        16
+
+void
+make_canonical_huffman_code(unsigned num_syms, unsigned max_codeword_len,
+                            const uint32_t freqs[], uint8_t lens[], uint32_t codewords[]);
+
+#endif /* _LIBLZX_COMPRESS_COMMON_H */
diff --git a/dlls/cabinet/liblzx_config.h b/dlls/cabinet/liblzx_config.h
new file mode 100644
index 00000000000..2c070d93022
--- /dev/null
+++ b/dlls/cabinet/liblzx_config.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright (C) 2012-2017 Eric Biggers
+ *
+ * This file is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU Lesser General Public License as published by the Free
+ * Software Foundation; either version 2.1 of the License, or (at your option) any
+ * later version.
+ *
+ * This file is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * along with this file; if not, see https://www.gnu.org/licenses/.
+ */
+ 
+ #pragma once
+
+#ifndef __LIBLZX_CONFIG_H__
+#define __LIBLZX_CONFIG_H__
+
+// Set to 1 to export as a DLL
+#ifndef LIBLZX_DLL_EXPORT
+#define LIBLZX_DLL_EXPORT 0
+#endif
+
+#endif
diff --git a/dlls/cabinet/liblzx_endianness.h b/dlls/cabinet/liblzx_endianness.h
new file mode 100644
index 00000000000..2ee8fceee69
--- /dev/null
+++ b/dlls/cabinet/liblzx_endianness.h
@@ -0,0 +1,136 @@
+/*
+ * endianness.h - macros and inline functions for endianness conversion
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _LIBLZX_ENDIANNESS_H
+#define _LIBLZX_ENDIANNESS_H
+
+#include "liblzx_compiler.h"
+#include "liblzx_types.h"
+
+#if LIBLZX_IS_MSVC_COMPILER
+#include <intrin.h>
+#endif
+
+#ifdef HAVE_SYS_ENDIAN_H
+   /* Needed on NetBSD to stop system bswap macros from messing things up */
+#  include <sys/endian.h>
+#  undef bswap16
+#  undef bswap32
+#  undef bswap64
+#endif
+
+/* Watch out for conflict with ntfs-3g/endians.h ... */
+#ifndef _NTFS_ENDIANS_H
+
+#define bswap16_const(n)                        \
+        ((((uint16_t)(n) & 0x00FF) << 8)        |       \
+         (((uint16_t)(n) & 0xFF00) >> 8))
+
+#define bswap32_const(n)                                \
+        ((((uint32_t)(n) & 0x000000FF) << 24)   |       \
+         (((uint32_t)(n) & 0x0000FF00) << 8)    |       \
+         (((uint32_t)(n) & 0x00FF0000) >> 8)    |       \
+         (((uint32_t)(n) & 0xFF000000) >> 24))
+
+#define bswap64_const(n)                                \
+        ((((uint64_t)(n) & 0x00000000000000FF) << 56)   |       \
+         (((uint64_t)(n) & 0x000000000000FF00) << 40)   |       \
+         (((uint64_t)(n) & 0x0000000000FF0000) << 24)   |       \
+         (((uint64_t)(n) & 0x00000000FF000000) << 8)    |       \
+         (((uint64_t)(n) & 0x000000FF00000000) >> 8)    |       \
+         (((uint64_t)(n) & 0x0000FF0000000000) >> 24)   |       \
+         (((uint64_t)(n) & 0x00FF000000000000) >> 40)   |       \
+         (((uint64_t)(n) & 0xFF00000000000000) >> 56))
+
+static attrib_forceinline uint16_t do_bswap16(uint16_t n)
+{
+#if LIBLZX_IS_MSVC_COMPILER
+        return _byteswap_ushort(n);
+#elif GCC_PREREQ(4, 8) || __has_builtin(__builtin_bswap16)
+        return __builtin_bswap16(n);
+#else
+        return bswap16_const(n);
+#endif
+}
+
+static attrib_forceinline uint32_t do_bswap32(uint32_t n)
+{
+#if LIBLZX_IS_MSVC_COMPILER
+        return _byteswap_ulong(n);
+#elif GCC_PREREQ(4, 3) || __has_builtin(__builtin_bswap32)
+        return __builtin_bswap32(n);
+#else
+        return bswap32_const(n);
+#endif
+}
+
+static attrib_forceinline uint64_t do_bswap64(uint64_t n)
+{
+#if LIBLZX_IS_MSVC_COMPILER
+        return _byteswap_uint64(n);
+#elif GCC_PREREQ(4, 3) || __has_builtin(__builtin_bswap64)
+        return __builtin_bswap64(n);
+#else
+        return bswap64_const(n);
+#endif
+}
+
+#define bswap16(n) (__builtin_constant_p(n) ? bswap16_const(n) : do_bswap16(n))
+#define bswap32(n) (__builtin_constant_p(n) ? bswap32_const(n) : do_bswap32(n))
+#define bswap64(n) (__builtin_constant_p(n) ? bswap64_const(n) : do_bswap64(n))
+
+#if CPU_IS_BIG_ENDIAN()
+#  define cpu_to_le16(n) ((_force_attr le16_t)bswap16(n))
+#  define cpu_to_le32(n) ((_force_attr le32_t)bswap32(n))
+#  define cpu_to_le64(n) ((_force_attr le64_t)bswap64(n))
+#  define le16_to_cpu(n) bswap16((_force_attr uint16_t)(le16_t)(n))
+#  define le32_to_cpu(n) bswap32((_force_attr uint32_t)(le32_t)(n))
+#  define le64_to_cpu(n) bswap64((_force_attr uint64_t)(le64_t)(n))
+#  define cpu_to_be16(n) ((_force_attr be16_t)(uint16_t)(n))
+#  define cpu_to_be32(n) ((_force_attr be32_t)(uint32_t)(n))
+#  define cpu_to_be64(n) ((_force_attr be64_t)(uint64_t)(n))
+#  define be16_to_cpu(n) ((_force_attr uint16_t)(be16_t)(n))
+#  define be32_to_cpu(n) ((_force_attr uint32_t)(be32_t)(n))
+#  define be64_to_cpu(n) ((_force_attr uint64_t)(be64_t)(n))
+#else
+#  define cpu_to_le16(n) ((_force_attr le16_t)(uint16_t)(n))
+#  define cpu_to_le32(n) ((_force_attr le32_t)(uint32_t)(n))
+#  define cpu_to_le64(n) ((_force_attr le64_t)(uint64_t)(n))
+#  define le16_to_cpu(n) ((_force_attr uint16_t)(le16_t)(n))
+#  define le32_to_cpu(n) ((_force_attr uint32_t)(le32_t)(n))
+#  define le64_to_cpu(n) ((_force_attr uint64_t)(le64_t)(n))
+#  define cpu_to_be16(n) ((_force_attr be16_t)bswap16(n))
+#  define cpu_to_be32(n) ((_force_attr be32_t)bswap32(n))
+#  define cpu_to_be64(n) ((_force_attr be64_t)bswap64(n))
+#  define be16_to_cpu(n) bswap16((_force_attr uint16_t)(be16_t)(n))
+#  define be32_to_cpu(n) bswap32((_force_attr uint32_t)(be32_t)(n))
+#  define be64_to_cpu(n) bswap64((_force_attr uint64_t)(be64_t)(n))
+#endif
+
+#endif /* _NTFS_ENDIANS_H */
+#endif /* _LIBLZX_ENDIANNESS_H */
diff --git a/dlls/cabinet/liblzx_error.h b/dlls/cabinet/liblzx_error.h
new file mode 100644
index 00000000000..cd984b61094
--- /dev/null
+++ b/dlls/cabinet/liblzx_error.h
@@ -0,0 +1,11 @@
+#ifndef __LIBLZX_ERROR_H__
+#define __LIBLZX_ERROR_H__
+
+enum liblzx_error {
+    LIBLZX_ERR_NONE = 0,
+
+    LIBLZX_ERR_NOMEM = -1,
+    LIBLZX_ERR_INVALID_PARAM = -2,
+};
+
+#endif /* __LIBLZX_ERROR_H__ */
diff --git a/dlls/cabinet/liblzx_hc_matchfinder.h b/dlls/cabinet/liblzx_hc_matchfinder.h
new file mode 100644
index 00000000000..31d659ab1fa
--- /dev/null
+++ b/dlls/cabinet/liblzx_hc_matchfinder.h
@@ -0,0 +1,432 @@
+/*
+ * hc_matchfinder.h - Lempel-Ziv matchfinding with a hash table of linked lists
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * ---------------------------------------------------------------------------
+ *
+ *                                   Algorithm
+ *
+ * This is a Hash Chains (hc) based matchfinder.
+ *
+ * The main data structure is a hash table where each hash bucket contains a
+ * linked list (or "chain") of sequences whose first 4 bytes share the same hash
+ * code.  Each sequence is identified by its starting position in the input
+ * buffer.
+ *
+ * The algorithm processes the input buffer sequentially.  At each byte
+ * position, the hash code of the first 4 bytes of the sequence beginning at
+ * that position (the sequence being matched against) is computed.  This
+ * identifies the hash bucket to use for that position.  Then, this hash
+ * bucket's linked list is searched for matches.  Then, a new linked list node
+ * is created to represent the current sequence and is prepended to the list.
+ *
+ * This algorithm has several useful properties:
+ *
+ * - It only finds true Lempel-Ziv matches; i.e., those where the matching
+ *   sequence occurs prior to the sequence being matched against.
+ *
+ * - The sequences in each linked list are always sorted by decreasing starting
+ *   position.  Therefore, the closest (smallest offset) matches are found
+ *   first, which in many compression formats tend to be the cheapest to encode.
+ *
+ * - Although fast running time is not guaranteed due to the possibility of the
+ *   lists getting very long, the worst degenerate behavior can be easily
+ *   prevented by capping the number of nodes searched at each position.
+ *
+ * - If the compressor decides not to search for matches at a certain position,
+ *   then that position can be quickly inserted without searching the list.
+ *
+ * - The algorithm is adaptable to sliding windows: just store the positions
+ *   relative to a "base" value that is updated from time to time, and stop
+ *   searching each list when the sequences get too far away.
+ *
+ * ---------------------------------------------------------------------------
+ *
+ *                                Notes on usage
+ *
+ * Before including this header, you must define 'mf_pos_t' to an integer type
+ * that can represent all possible positions.  This can be a 16-bit or 32-bit
+ * unsigned integer.  When possible, the former should be used due to the
+ * reduced cache pressure.  This header can be included multiple times in a
+ * single .c file with different 'mf_pos_t' definitions; however, you must
+ * define a different MF_SUFFIX each time to generate different names for the
+ * matchfinder structure and functions.
+ *
+ * The number of bytes that must be allocated for a given 'struct
+ * hc_matchfinder' must be gotten by calling hc_matchfinder_size().
+ *
+ * ----------------------------------------------------------------------------
+ *
+ *                                 Optimizations
+ *
+ * The main hash table and chains handle length 4+ matches.  Length 3 matches
+ * are handled by a separate hash table with no chains.  This works well for
+ * typical "greedy" or "lazy"-style compressors, where length 3 matches are
+ * often only helpful if they have small offsets.  Instead of searching a full
+ * chain for length 3+ matches, the algorithm just checks for one close length 3
+ * match, then focuses on finding length 4+ matches.
+ *
+ * The longest_match() and skip_bytes() functions are inlined into the
+ * compressors that use them.  This isn't just about saving the overhead of a
+ * function call.  These functions are intended to be called from the inner
+ * loops of compressors, where giving the compiler more control over register
+ * allocation is very helpful.  There is also significant benefit to be gained
+ * from allowing the CPU to predict branches independently at each call site.
+ * For example, "lazy"-style compressors can be written with two calls to
+ * longest_match(), each of which starts with a different 'best_len' and
+ * therefore has significantly different performance characteristics.
+ *
+ * Although any hash function can be used, a multiplicative hash is fast and
+ * works well.
+ *
+ * On some processors, it is significantly faster to extend matches by whole
+ * words (32 or 64 bits) instead of by individual bytes.  For this to be the
+ * case, the processor must implement unaligned memory accesses efficiently and
+ * must have either a fast "find first set bit" instruction or a fast "find last
+ * set bit" instruction, depending on the processor's endianness.
+ *
+ * The code uses one loop for finding the first match and one loop for finding a
+ * longer match.  Each of these loops is tuned for its respective task and in
+ * combination are faster than a single generalized loop that handles both
+ * tasks.
+ *
+ * The code also uses a tight inner loop that only compares the last and first
+ * bytes of a potential match.  It is only when these bytes match that a full
+ * match extension is attempted.
+ *
+ * ----------------------------------------------------------------------------
+ */
+
+#include <string.h>
+
+#include "liblzx_matchfinder_common.h"
+
+#define HC_MATCHFINDER_HASH3_ORDER        15
+#define HC_MATCHFINDER_HASH4_ORDER        16
+
+/* TEMPLATED functions and structures have MF_SUFFIX appended to their name.  */
+#undef TEMPLATED
+#define TEMPLATED(name)                CONCAT(name, MF_SUFFIX)
+
+struct TEMPLATED(hc_matchfinder) {
+
+        /* The hash table for finding length 3 matches  */
+        mf_pos_t hash3_tab[1UL << HC_MATCHFINDER_HASH3_ORDER];
+
+        /* The hash table which contains the first nodes of the linked lists for
+         * finding length 4+ matches  */
+        mf_pos_t hash4_tab[1UL << HC_MATCHFINDER_HASH4_ORDER];
+
+        /* The "next node" references for the linked lists.  The "next node" of
+         * the node for the sequence with position 'pos' is 'next_tab[pos]'.  */
+        mf_pos_t next_tab[];
+};
+
+/* Return the number of bytes that must be allocated for a 'hc_matchfinder' that
+ * can work with buffers up to the specified size.  */
+static attrib_forceinline size_t
+TEMPLATED(hc_matchfinder_size)(size_t max_bufsize, bool streaming)
+{
+        const size_t streaming_mul = streaming ? 2 : 1;
+
+        return sizeof(struct TEMPLATED(hc_matchfinder)) +
+               (max_bufsize * streaming_mul * sizeof(mf_pos_t));
+}
+
+/* Prepare the matchfinder for a new input buffer.  */
+static attrib_forceinline void
+TEMPLATED(hc_matchfinder_init)(struct TEMPLATED(hc_matchfinder) * mf,
+                               size_t max_bufsize, bool streaming)
+{
+        memset(mf, 0xFF, TEMPLATED(hc_matchfinder_size)(max_bufsize, streaming));
+}
+
+/* The minimum permissible value of 'max_len' for bt_matchfinder_get_matches()
+ * and bt_matchfinder_skip_byte().  There must be sufficiently many bytes
+ * remaining to load a 32-bit integer from the *next* position.  */
+#define HC_MATCHFINDER_REQUIRED_NBYTES 5
+
+/*
+ * Find the longest match longer than 'best_len' bytes.
+ *
+ * @mf
+ *        The matchfinder structure.
+ * @in_begin
+ *        Pointer to the beginning of the input buffer.
+ * @in_next
+ *        Pointer to the next position in the input buffer, i.e. the sequence
+ *        being matched against.
+ * @best_len
+ *        Require a match longer than this length.
+ * @max_len
+ *        The maximum permissible match length at this position.
+ * @nice_len
+ *        Stop searching if a match of at least this length is found.
+ *        Must be <= @max_len.
+ * @max_search_depth
+ *        Limit on the number of potential matches to consider.  Must be >= 1.
+ * @next_hashes
+ *        The precomputed hash codes for the sequence beginning at @in_next.
+ *        These will be used and then updated with the precomputed hashcodes for
+ *        the sequence beginning at @in_next + 1.
+ * @offset_ret
+ *        If a match is found, its offset is returned in this location.
+ *
+ * Return the length of the match found, or 'best_len' if no match longer than
+ * 'best_len' was found.
+ */
+static attrib_forceinline uint32_t
+TEMPLATED(hc_matchfinder_longest_match)(struct TEMPLATED(hc_matchfinder) * const mf,
+                                        const uint8_t * const in_begin,
+                                        uint32_t in_min_pos,
+                                        const uint8_t * const in_next,
+                                        uint32_t best_len,
+                                        const uint32_t max_find_len,
+                                        const uint32_t max_produce_len,
+                                        const uint32_t nice_len,
+                                        const uint32_t max_search_depth,
+                                        uint32_t * const next_hashes,
+                                        uint32_t * const offset_ret)
+{
+        uint32_t depth_remaining = max_search_depth;
+        const uint8_t *best_matchptr = in_next;
+        mf_pos_t cur_node3, cur_node4;
+        uint32_t hash3, hash4;
+        uint32_t next_hashseq;
+        uint32_t seq4;
+        const uint8_t *matchptr;
+        uint32_t len;
+        uint32_t cur_pos = in_next - in_begin;
+
+        /* can we read 4 bytes from 'in_next + 1'? */
+        if (unlikely(max_find_len < HC_MATCHFINDER_REQUIRED_NBYTES))
+                goto out;
+
+        /* Get the precomputed hash codes.  */
+        hash3 = next_hashes[0];
+        hash4 = next_hashes[1];
+
+        /* From the hash buckets, get the first node of each linked list.  */
+        cur_node3 = mf->hash3_tab[hash3];
+        cur_node4 = mf->hash4_tab[hash4];
+
+        /* Update for length 3 matches.  This replaces the singleton node in the
+         * 'hash3' bucket with the node for the current sequence.  */
+        mf->hash3_tab[hash3] = cur_pos;
+
+        /* Update for length 4 matches.  This prepends the node for the current
+         * sequence to the linked list in the 'hash4' bucket.  */
+        mf->hash4_tab[hash4] = cur_pos;
+        mf->next_tab[cur_pos] = cur_node4;
+
+        /* Compute the next hash codes.  */
+        next_hashseq = get_unaligned_le32(in_next + 1);
+        next_hashes[0] = lz_hash(next_hashseq & 0xFFFFFF, HC_MATCHFINDER_HASH3_ORDER);
+        next_hashes[1] = lz_hash(next_hashseq, HC_MATCHFINDER_HASH4_ORDER);
+        prefetchw(&mf->hash3_tab[next_hashes[0]]);
+        prefetchw(&mf->hash4_tab[next_hashes[1]]);
+
+        if (best_len < 4) {  /* No match of length >= 4 found yet?  */
+
+                /* Check for a length 3 match if needed.  */
+
+                if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node3, in_min_pos))
+                        goto out;
+
+                seq4 = load_u32_unaligned(in_next);
+
+                if (best_len < 3) {
+                        matchptr = &in_begin[cur_node3];
+                        if (load_u24_unaligned(matchptr) == loaded_u32_to_u24(seq4)) {
+                                best_len = 3;
+                                best_matchptr = matchptr;
+                        }
+                }
+
+                /* Check for a length 4 match.  */
+
+                if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node4, in_min_pos))
+                        goto out;
+
+                for (;;) {
+                        /* No length 4 match found yet.  Check the first 4 bytes.  */
+                        matchptr = &in_begin[cur_node4];
+
+                        if (load_u32_unaligned(matchptr) == seq4)
+                                break;
+
+                        /* The first 4 bytes did not match.  Keep trying.  */
+                        cur_node4 = mf->next_tab[cur_node4];
+                        if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node4,
+                                                                 in_min_pos) ||
+                            !--depth_remaining)
+                                goto out;
+                }
+
+                /* Found a match of length >= 4.  Extend it to its full length.  */
+                best_matchptr = matchptr;
+                best_len = lz_extend(in_next, best_matchptr, 4, max_find_len);
+                if (best_len >= nice_len)
+                        goto out;
+                cur_node4 = mf->next_tab[cur_node4];
+                if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node4,
+                                                         in_min_pos) ||
+                    !--depth_remaining)
+                        goto out;
+        } else {
+                if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node4,
+                                                         in_min_pos) ||
+                    best_len >= nice_len)
+                        goto out;
+        }
+
+        /* Check for matches of length >= 5.  */
+
+        for (;;) {
+                for (;;) {
+                        matchptr = &in_begin[cur_node4];
+
+                        /* Already found a length 4 match.  Try for a longer
+                         * match; start by checking either the last 4 bytes and
+                         * the first 4 bytes, or the last byte.  (The last byte,
+                         * the one which would extend the match length by 1, is
+                         * the most important.)  */
+                #if UNALIGNED_ACCESS_IS_FAST
+                        if ((load_u32_unaligned(matchptr + best_len - 3) ==
+                             load_u32_unaligned(in_next + best_len - 3)) &&
+                            (load_u32_unaligned(matchptr) ==
+                             load_u32_unaligned(in_next)))
+                #else
+                        if (matchptr[best_len] == in_next[best_len])
+                #endif
+                                break;
+
+                        /* Continue to the next node in the list.  */
+                        cur_node4 = mf->next_tab[cur_node4];
+                        if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node4,
+                                                                 in_min_pos) ||
+                                !--depth_remaining)
+                                goto out;
+                }
+
+        #if UNALIGNED_ACCESS_IS_FAST
+                len = 4;
+        #else
+                len = 0;
+        #endif
+                len = lz_extend(in_next, matchptr, len, max_find_len);
+                if (len > best_len) {
+                        /* This is the new longest match.  */
+                        best_len = len;
+                        best_matchptr = matchptr;
+                        if (best_len >= nice_len)
+                                goto out;
+                }
+
+                /* Continue to the next node in the list.  */
+                cur_node4 = mf->next_tab[cur_node4];
+                if (!TEMPLATED(matchfinder_is_valid_pos)(cur_node4,
+                                                         in_min_pos) ||
+                    !--depth_remaining)
+                        goto out;
+        }
+out:
+        *offset_ret = in_next - best_matchptr;
+        best_len = min_u32(best_len, max_produce_len);
+        if (best_len < 2)
+                best_len = 2;
+
+        return best_len;
+}
+
+/*
+ * Advance the matchfinder, but don't search for matches.
+ *
+ * @mf
+ *        The matchfinder structure.
+ * @in_begin
+ *        Pointer to the beginning of the input buffer.
+ * @in_next
+ *        Pointer to the next position in the input buffer.
+ * @in_end
+ *        Pointer to the end of the input buffer.
+ * @count
+ *        The number of bytes to advance.  Must be > 0.
+ * @next_hashes
+ *        The precomputed hash codes for the sequence beginning at @in_next.
+ *        These will be used and then updated with the precomputed hashcodes for
+ *        the sequence beginning at @in_next + @count.
+ */
+static attrib_forceinline void
+TEMPLATED(hc_matchfinder_skip_bytes)(struct TEMPLATED(hc_matchfinder) * const mf,
+                                     const uint8_t * const in_begin,
+                                     const uint8_t *in_next,
+                                     const uint8_t * const in_end,
+                                     const uint32_t count,
+                                     uint32_t * const next_hashes)
+{
+        uint32_t cur_pos;
+        uint32_t hash3, hash4;
+        uint32_t next_hashseq;
+        uint32_t remaining = count;
+
+        if (unlikely(count + HC_MATCHFINDER_REQUIRED_NBYTES > in_end - in_next))
+                return;
+
+        cur_pos = in_next - in_begin;
+        hash3 = next_hashes[0];
+        hash4 = next_hashes[1];
+        do {
+                mf->hash3_tab[hash3] = cur_pos;
+                mf->next_tab[cur_pos] = mf->hash4_tab[hash4];
+                mf->hash4_tab[hash4] = cur_pos;
+
+                next_hashseq = get_unaligned_le32(++in_next);
+                hash3 = lz_hash(next_hashseq & 0xFFFFFF, HC_MATCHFINDER_HASH3_ORDER);
+                hash4 = lz_hash(next_hashseq, HC_MATCHFINDER_HASH4_ORDER);
+                cur_pos++;
+        } while (--remaining);
+
+        prefetchw(&mf->hash3_tab[hash3]);
+        prefetchw(&mf->hash4_tab[hash4]);
+        next_hashes[0] = hash3;
+        next_hashes[1] = hash4;
+}
+
+/*
+ * Culls any matches that are lower than a specified offset and reduces any
+ * remaining offsets by the same amount.
+ */
+static attrib_forceinline void
+TEMPLATED(hc_matchfinder_cull)(struct TEMPLATED(hc_matchfinder) * mf,
+                               uint32_t cull_size, uint32_t window_size)
+{
+        const size_t mf_count =
+            TEMPLATED(hc_matchfinder_size)(window_size, true) /
+            sizeof(mf_pos_t);
+
+        TEMPLATED(matchfinder_rebase)((mf_pos_t *)mf, mf_count, cull_size);
+}
diff --git a/dlls/cabinet/liblzx_lzx_common.c b/dlls/cabinet/liblzx_lzx_common.c
new file mode 100644
index 00000000000..d8619e0965f
--- /dev/null
+++ b/dlls/cabinet/liblzx_lzx_common.c
@@ -0,0 +1,325 @@
+/*
+ * lzx_common.c - Common code for LZX compression and decompression.
+ */
+
+/*
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright (C) 2012-2016 Eric Biggers
+ *
+ * This file is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU Lesser General Public License as published by the Free
+ * Software Foundation; either version 2.1 of the License, or (at your option) any
+ * later version.
+ *
+ * This file is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * along with this file; if not, see https://www.gnu.org/licenses/.
+ */
+
+#ifdef HAVE_CONFIG_H
+#  include "config.h"
+#endif
+
+#include "liblzx_minmax.h"
+
+#include <string.h>
+
+#ifdef __SSE2__
+#  include <emmintrin.h>
+#endif
+
+#ifdef __AVX2__
+#  include <immintrin.h>
+#endif
+
+#include "liblzx_bitops.h"
+#include "liblzx_endianness.h"
+#include "liblzx_lzx_common.h"
+#include "liblzx_unaligned.h"
+#include "liblzx_util.h"
+
+/* Mapping: offset slot => first match offset that uses that offset slot.
+ * The offset slots for repeat offsets map to "fake" offsets < 1.  */
+const int32_t lzx_offset_slot_base[LZX_MAX_OFFSET_SLOTS + 1] = {
+        -2     , -1     , 0      , 1      , 2      ,    /* 0  --- 4  */
+        4      , 6      , 10     , 14     , 22     ,    /* 5  --- 9  */
+        30     , 46     , 62     , 94     , 126    ,    /* 10 --- 14 */
+        190    , 254    , 382    , 510    , 766    ,    /* 15 --- 19 */
+        1022   , 1534   , 2046   , 3070   , 4094   ,    /* 20 --- 24 */
+        6142   , 8190   , 12286  , 16382  , 24574  ,    /* 25 --- 29 */
+        32766  , 49150  , 65534  , 98302  , 131070 ,    /* 30 --- 34 */
+        196606 , 262142 , 393214 , 524286 , 655358 ,    /* 35 --- 39 */
+        786430 , 917502 , 1048574, 1179646, 1310718,    /* 40 --- 44 */
+        1441790, 1572862, 1703934, 1835006, 1966078,    /* 45 --- 49 */
+        2097150                                         /* extra     */
+};
+
+/* Mapping: offset slot => how many extra bits must be read and added to the
+ * corresponding offset slot base to decode the match offset.  */
+const uint8_t lzx_extra_offset_bits[LZX_MAX_OFFSET_SLOTS] = {
+        0 , 0 , 0 , 0 , 1 ,
+        1 , 2 , 2 , 3 , 3 ,
+        4 , 4 , 5 , 5 , 6 ,
+        6 , 7 , 7 , 8 , 8 ,
+        9 , 9 , 10, 10, 11,
+        11, 12, 12, 13, 13,
+        14, 14, 15, 15, 16,
+        16, 17, 17, 17, 17,
+        17, 17, 17, 17, 17,
+        17, 17, 17, 17, 17,
+};
+
+/* Round the specified buffer size up to the next valid LZX window size, and
+ * return its order (log2).  Or, if the buffer size is 0 or greater than the
+ * largest valid LZX window size, return 0.  */
+unsigned
+lzx_get_window_order(size_t max_bufsize)
+{
+        if (max_bufsize == 0 || max_bufsize > LZX_MAX_WINDOW_SIZE)
+                return 0;
+
+        return max_uint(ilog2_ceil(max_bufsize), LZX_MIN_WINDOW_ORDER);
+}
+
+/* Given a valid LZX window order, return the number of symbols that will exist
+ * in the main Huffman code.  */
+unsigned
+lzx_get_num_main_syms(unsigned window_order)
+{
+        /* Note: one would expect that the maximum match offset would be
+         * 'window_size - LZX_MIN_MATCH_LEN', which would occur if the first two
+         * bytes were to match the last two bytes.  However, the format
+         * disallows this case.  This reduces the number of needed offset slots
+         * by 1.  */
+        uint32_t window_size = (uint32_t)1 << window_order;
+        uint32_t max_offset = window_size - LZX_MIN_MATCH_LEN - 1;
+        unsigned num_offset_slots = 30;
+        while (max_offset >= (uint32_t)lzx_offset_slot_base[num_offset_slots])
+                num_offset_slots++;
+
+        return LZX_NUM_CHARS + (num_offset_slots * LZX_NUM_LEN_HEADERS);
+}
+
+static void
+do_translate_target(void *target, int32_t input_pos, int32_t e8_file_size)
+{
+        int32_t abs_offset, rel_offset;
+
+        rel_offset = get_unaligned_le32(target);
+        if (rel_offset >= -input_pos && rel_offset < e8_file_size) {
+                if (rel_offset < e8_file_size - input_pos) {
+                        /* "good translation" */
+                        abs_offset = rel_offset + input_pos;
+                } else {
+                        /* "compensating translation" */
+                        abs_offset = rel_offset - e8_file_size;
+                }
+                put_unaligned_le32(abs_offset, target);
+        }
+}
+
+static void
+undo_translate_target(void *target, int32_t input_pos, int32_t e8_file_size)
+{
+        int32_t abs_offset, rel_offset;
+
+        abs_offset = get_unaligned_le32(target);
+        if (abs_offset >= 0) {
+                if (abs_offset < e8_file_size) {
+                        /* "good translation" */
+                        rel_offset = abs_offset - input_pos;
+                        put_unaligned_le32(rel_offset, target);
+                }
+        } else {
+                if (abs_offset >= -input_pos) {
+                        /* "compensating translation" */
+                        rel_offset = abs_offset + e8_file_size;
+                        put_unaligned_le32(rel_offset, target);
+                }
+        }
+}
+
+/*
+ * Do or undo the 'E8' preprocessing used in LZX.  Before compression, the
+ * uncompressed data is preprocessed by changing the targets of x86 CALL
+ * instructions from relative offsets to absolute offsets.  After decompression,
+ * the translation is undone by changing the targets of x86 CALL instructions
+ * from absolute offsets to relative offsets.
+ *
+ * Note that despite its intent, E8 preprocessing can be done on any data even
+ * if it is not actually x86 machine code.  In fact, E8 preprocessing appears to
+ * always be used in LZX-compressed resources in WIM files; there is no bit to
+ * indicate whether it is used or not, unlike in the LZX compressed format as
+ * used in cabinet files, where a bit is reserved for that purpose.
+ *
+ * E8 preprocessing is disabled in the last 6 bytes of the uncompressed data,
+ * which really means the 5-byte call instruction cannot start in the last 10
+ * bytes of the uncompressed data.  This is one of the errors in the LZX
+ * documentation.
+ *
+ * E8 preprocessing does not appear to be disabled after the 32768th chunk of a
+ * WIM resource, which apparently is another difference from the LZX compression
+ * used in cabinet files.
+ *
+ * E8 processing is supposed to take the file size as a parameter, as it is used
+ * in calculating the translated jump targets.  But in WIM files, this file size
+ * is always the same (LZX_WIM_MAGIC_FILESIZE == 12000000).
+ */
+static void
+lzx_e8_filter(uint8_t *data, uint32_t size, uint32_t chunk_offset, uint32_t e8_file_size,
+              void (*process_target)(void *, int32_t, int32_t))
+{
+
+#if !defined(__SSE2__) && !defined(__AVX2__)
+        uint8_t *tail;
+        uint8_t *p;
+
+        if (size <= LZX_E8_FILTER_TAIL_SIZE)
+                return;
+
+        tail = &data[size - LZX_E8_FILTER_TAIL_SIZE];
+        p = data;
+        while (p < tail) {
+                if (*p != 0xE8) {
+                        p++;
+                        continue;
+                }
+
+                (*process_target)(p + 1, (int32_t)(p - data + chunk_offset),
+                                  e8_file_size);
+                p += 5;
+        }
+#else
+        /* SSE2 or AVX-2 optimized version for x86_64  */
+
+        uint8_t *p = data;
+        uint64_t valid_mask = ~0;
+
+        if (size <= LZX_E8_FILTER_TAIL_SIZE)
+                return;
+#ifdef __AVX2__
+#  define ALIGNMENT_REQUIRED 32
+#else
+#  define ALIGNMENT_REQUIRED 16
+#endif
+
+        /* Process one byte at a time until the pointer is properly aligned.  */
+        while ((uintptr_t)p % ALIGNMENT_REQUIRED != 0) {
+                if (p >= data + size - LZX_E8_FILTER_TAIL_SIZE)
+                        return;
+                if (*p == 0xE8 && (valid_mask & 1)) {
+                        (*process_target)(p + 1, p - data + chunk_offset,
+                                          e8_file_size);
+                        valid_mask &= ~0x1F;
+                }
+                p++;
+                valid_mask >>= 1;
+                valid_mask |= (uint64_t)1 << 63;
+        }
+
+        if (data + size - p >= 64) {
+
+                /* Vectorized processing  */
+
+                /* Note: we use a "trap" E8 byte to eliminate the need to check
+                 * for end-of-buffer in the inner loop.  This byte is carefully
+                 * positioned so that it will never be changed by a previous
+                 * translation before it is detected.  */
+
+                uint8_t *trap = p + ((data + size - p) & ~31) - 32 + 4;
+                uint8_t saved_byte = *trap;
+                *trap = 0xE8;
+
+                for (;;) {
+                        uint32_t e8_mask;
+                        uint8_t *orig_p = p;
+                #ifdef __AVX2__
+                        const __m256i e8_bytes = _mm256_set1_epi8(0xE8);
+                        for (;;) {
+                                __m256i bytes = *(const __m256i *)p;
+                                __m256i cmpresult = _mm256_cmpeq_epi8(bytes, e8_bytes);
+                                e8_mask = _mm256_movemask_epi8(cmpresult);
+                                if (e8_mask)
+                                        break;
+                                p += 32;
+                        }
+                #else
+                        const __m128i e8_bytes = _mm_set1_epi8(0xE8);
+                        for (;;) {
+                                /* Read the next 32 bytes of data and test them
+                                 * for E8 bytes.  */
+                                __m128i bytes1 = *(const __m128i *)p;
+                                __m128i bytes2 = *(const __m128i *)(p + 16);
+                                __m128i cmpresult1 = _mm_cmpeq_epi8(bytes1, e8_bytes);
+                                __m128i cmpresult2 = _mm_cmpeq_epi8(bytes2, e8_bytes);
+                                uint32_t mask1 = _mm_movemask_epi8(cmpresult1);
+                                uint32_t mask2 = _mm_movemask_epi8(cmpresult2);
+                                /* The masks have a bit set for each E8 byte.
+                                 * We stay in this fast inner loop as long as
+                                 * there are no E8 bytes.  */
+                                if (mask1 | mask2) {
+                                        e8_mask = mask1 | (mask2 << 16);
+                                        break;
+                                }
+                                p += 32;
+                        }
+                #endif
+
+                        /* Did we pass over data with no E8 bytes?  */
+                        if (p != orig_p)
+                                valid_mask = ~0;
+
+                        /* Are we nearing end-of-buffer?  */
+                        if (p == trap - 4)
+                                break;
+
+                        /* Process the E8 bytes.  However, the AND with
+                         * 'valid_mask' ensures we never process an E8 byte that
+                         * was itself part of a translation target.  */
+                        while ((e8_mask &= valid_mask)) {
+                                unsigned bit = bsf32(e8_mask);
+                                (*process_target)(p + bit + 1,
+                                                  p + bit - data + chunk_offset,
+                                                  e8_file_size);
+                                valid_mask &= ~((uint64_t)0x1F << bit);
+                        }
+
+                        valid_mask >>= 32;
+                        valid_mask |= 0xFFFFFFFF00000000;
+                        p += 32;
+                }
+
+                *trap = saved_byte;
+        }
+
+        /* Approaching the end of the buffer; process one byte a time.  */
+        while (p < data + size - LZX_E8_FILTER_TAIL_SIZE) {
+                if (*p == 0xE8 && (valid_mask & 1)) {
+                        (*process_target)(p + 1, p - data + chunk_offset,
+                                          e8_file_size);
+                        valid_mask &= ~0x1F;
+                }
+                p++;
+                valid_mask >>= 1;
+                valid_mask |= (uint64_t)1 << 63;
+        }
+#endif /* __SSE2__ || __AVX2__ */
+}
+
+void
+lzx_preprocess(uint8_t *data, uint32_t size, uint32_t chunk_offset, uint32_t e8_file_size)
+{
+        lzx_e8_filter(data, size, chunk_offset, e8_file_size,
+                      do_translate_target);
+}
+
+void
+lzx_postprocess(uint8_t *data, uint32_t size, uint32_t chunk_offset, uint32_t e8_file_size)
+{
+        lzx_e8_filter(data, size, chunk_offset, e8_file_size,
+                      undo_translate_target);
+}
diff --git a/dlls/cabinet/liblzx_lzx_common.h b/dlls/cabinet/liblzx_lzx_common.h
new file mode 100644
index 00000000000..d6fd20b095c
--- /dev/null
+++ b/dlls/cabinet/liblzx_lzx_common.h
@@ -0,0 +1,29 @@
+/*
+ * lzx_common.h
+ *
+ * Declarations shared between LZX compression and decompression.
+ */
+
+#ifndef _LZX_COMMON_H
+#define _LZX_COMMON_H
+
+#include "liblzx_lzx_constants.h"
+#include "liblzx_types.h"
+
+extern const int32_t lzx_offset_slot_base[LZX_MAX_OFFSET_SLOTS + 1];
+
+extern const uint8_t lzx_extra_offset_bits[LZX_MAX_OFFSET_SLOTS];
+
+unsigned
+lzx_get_window_order(size_t max_bufsize);
+
+unsigned
+lzx_get_num_main_syms(unsigned window_order);
+
+void
+lzx_preprocess(uint8_t *data, uint32_t size, uint32_t chunk_offset, uint32_t e8_file_size);
+
+void
+lzx_postprocess(uint8_t *data, uint32_t size, uint32_t chunk_offset, uint32_t e8_file_size);
+
+#endif /* _LZX_COMMON_H */
diff --git a/dlls/cabinet/liblzx_lzx_compress.c b/dlls/cabinet/liblzx_lzx_compress.c
new file mode 100644
index 00000000000..f26629d5fdb
--- /dev/null
+++ b/dlls/cabinet/liblzx_lzx_compress.c
@@ -0,0 +1,3662 @@
+/*
+ * lzx_compress.c
+ *
+ * A compressor for the LZX compression format, as used in WIM archives.
+ */
+
+/*
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright (C) 2012-2017 Eric Biggers
+ *
+ * This file is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU Lesser General Public License as published by the Free
+ * Software Foundation; either version 2.1 of the License, or (at your option) any
+ * later version.
+ *
+ * This file is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * along with this file; if not, see https://www.gnu.org/licenses/.
+ */
+
+
+/*
+ * This file contains a compressor for the LZX ("Lempel-Ziv eXtended")
+ * compression format, as used in the WIM (Windows IMaging) file format.
+ *
+ * Two different LZX-compatible algorithms are implemented: "near-optimal" and
+ * "lazy".  "Near-optimal" is significantly slower than "lazy", but results in a
+ * better compression ratio.  The "near-optimal" algorithm is used at the
+ * default compression level.
+ *
+ * This file may need some slight modifications to be used outside of the WIM
+ * format.  In particular, in other situations the LZX block header might be
+ * slightly different, and sliding window support might be required.
+ *
+ * LZX is a compression format derived from DEFLATE, the format used by zlib and
+ * gzip.  Both LZX and DEFLATE use LZ77 matching and Huffman coding.  Certain
+ * details are quite similar, such as the method for storing Huffman codes.
+ * However, the main differences are:
+ *
+ * - LZX preprocesses the data to attempt to make x86 machine code slightly more
+ *   compressible before attempting to compress it further.
+ *
+ * - LZX uses a "main" alphabet which combines literals and matches, with the
+ *   match symbols containing a "length header" (giving all or part of the match
+ *   length) and an "offset slot" (giving, roughly speaking, the order of
+ *   magnitude of the match offset).
+ *
+ * - LZX does not have static Huffman blocks (that is, the kind with preset
+ *   Huffman codes); however it does have two types of dynamic Huffman blocks
+ *   ("verbatim" and "aligned").
+ *
+ * - LZX has a minimum match length of 2 rather than 3.  Length 2 matches can be
+ *   useful, but generally only if the compressor is smart about choosing them.
+ *
+ * - In LZX, offset slots 0 through 2 actually represent entries in an LRU queue
+ *   of match offsets.  This is very useful for certain types of files, such as
+ *   binary files that have repeating records.
+ */
+
+/******************************************************************************/
+/*                            General parameters                              */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * The compressor uses the faster algorithm at levels <= MAX_FAST_LEVEL.  It
+ * uses the slower algorithm at levels > MAX_FAST_LEVEL.
+ */
+#define MAX_FAST_LEVEL                                34
+
+/*
+ * The compressor-side limits on the codeword lengths (in bits) for each Huffman
+ * code.  To make outputting bits slightly faster, some of these limits are
+ * lower than the limits defined by the LZX format.  This does not significantly
+ * affect the compression ratio.
+ */
+#define MAIN_CODEWORD_LIMIT                        16
+#define LENGTH_CODEWORD_LIMIT                        12
+#define ALIGNED_CODEWORD_LIMIT                        7
+#define PRE_CODEWORD_LIMIT                        7
+
+/******************************************************************************/
+/*                         Block splitting parameters                         */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * The compressor always outputs blocks of at least this size in bytes, except
+ * for the last block which may need to be smaller.
+ */
+#define MIN_BLOCK_SIZE                                6500
+
+/*
+ * The compressor attempts to end a block when it reaches this size in bytes.
+ * The final size might be slightly larger due to matches extending beyond the
+ * end of the block.  Specifically:
+ *
+ *  - The near-optimal compressor may choose a match of up to LZX_MAX_MATCH_LEN
+ *    bytes starting at position 'SOFT_MAX_BLOCK_SIZE - 1'.
+ *
+ *  - The lazy compressor may choose a sequence of literals starting at position
+ *    'SOFT_MAX_BLOCK_SIZE - 1' when it sees a sequence of increasingly better
+ *    matches.  The final match may be up to LZX_MAX_MATCH_LEN bytes.  The
+ *    length of the literal sequence is approximately limited by the "nice match
+ *    length" parameter.
+ */
+#define SOFT_MAX_BLOCK_SIZE                        100000
+
+/*
+ * The number of observed items (matches and literals) that represents
+ * sufficient data for the compressor to decide whether the current block should
+ * be ended or not.
+ */
+#define NUM_OBSERVATIONS_PER_BLOCK_CHECK        400
+
+
+/******************************************************************************/
+/*                      Parameters for slower algorithm                       */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * The log base 2 of the number of entries in the hash table for finding length
+ * 2 matches.  This could be as high as 16, but using a smaller hash table
+ * speeds up compression due to reduced cache pressure.
+ */
+#define BT_MATCHFINDER_HASH2_ORDER                12
+
+/*
+ * The number of lz_match structures in the match cache, excluding the extra
+ * "overflow" entries.  This value should be high enough so that nearly the
+ * time, all matches found in a given block can fit in the match cache.
+ * However, fallback behavior (immediately terminating the block) on cache
+ * overflow is still required.
+ */
+#define CACHE_LENGTH                                (SOFT_MAX_BLOCK_SIZE * 5)
+
+/*
+ * An upper bound on the number of matches that can ever be saved in the match
+ * cache for a single position.  Since each match we save for a single position
+ * has a distinct length, we can use the number of possible match lengths in LZX
+ * as this bound.  This bound is guaranteed to be valid in all cases, although
+ * if 'nice_match_length < LZX_MAX_MATCH_LEN', then it will never actually be
+ * reached.
+ */
+#define MAX_MATCHES_PER_POS                        LZX_NUM_LENS
+
+/*
+ * A scaling factor that makes it possible to consider fractional bit costs.  A
+ * single bit has a cost of BIT_COST.
+ *
+ * Note: this is only useful as a statistical trick for when the true costs are
+ * unknown.  Ultimately, each token in LZX requires a whole number of bits to
+ * output.
+ */
+#define BIT_COST_BITS                                6
+#define BIT_COST                                (1 << BIT_COST_BITS)
+
+/*
+ * Should the compressor take into account the costs of aligned offset symbols
+ * instead of assuming that all are equally likely?
+ */
+#define CONSIDER_ALIGNED_COSTS                        1
+
+/*
+ * Should the "minimum" cost path search algorithm consider "gap" matches, where
+ * a normal match is followed by a literal, then by a match with the same
+ * offset?  This is one specific, somewhat common situation in which the true
+ * minimum cost path is often different from the path found by looking only one
+ * edge ahead.
+ */
+#define CONSIDER_GAP_MATCHES                        1
+
+/******************************************************************************/
+/*                                  Includes                                  */
+/*----------------------------------------------------------------------------*/
+
+#ifdef HAVE_CONFIG_H
+#  include "config.h"
+#endif
+
+#include "liblzx_compress_common.h"
+#include "liblzx_error.h"
+#include "liblzx_lzx_common.h"
+#include "liblzx_minmax.h"
+#include "liblzx_unaligned.h"
+#include "liblzx_util.h"
+#include "liblzx.h"
+
+#include <assert.h>
+
+#include <malloc.h>
+
+/* Note: BT_MATCHFINDER_HASH2_ORDER must be defined before including
+ * bt_matchfinder.h. */
+
+/* Matchfinders with 16-bit positions */
+#define mf_pos_t        uint16_t
+#define MF_SUFFIX        _16
+#define MF_INVALID_POS        (0xFFFFu)
+#include "liblzx_bt_matchfinder.h"
+#include "liblzx_hc_matchfinder.h"
+
+/* Matchfinders with 32-bit positions */
+#undef mf_pos_t
+#undef MF_SUFFIX
+#undef MF_INVALID_POS
+#define mf_pos_t        uint32_t
+#define MF_SUFFIX        _32
+#define MF_INVALID_POS        (0xFFFFFFFFu)
+#include "liblzx_bt_matchfinder.h"
+#include "liblzx_hc_matchfinder.h"
+
+#undef mf_pos_t
+#undef MF_SUFFIX
+#undef MF_INVALID_POS
+
+/******************************************************************************/
+/*                            Compressor structure                            */
+/*----------------------------------------------------------------------------*/
+
+/* Codewords for the Huffman codes */
+struct lzx_codewords {
+        uint32_t main[LZX_MAINCODE_MAX_NUM_SYMBOLS];
+        uint32_t len[LZX_LENCODE_NUM_SYMBOLS];
+        uint32_t aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS];
+};
+
+/*
+ * Codeword lengths, in bits, for the Huffman codes.
+ *
+ * A codeword length of 0 means the corresponding codeword has zero frequency.
+ *
+ * The main and length codes each have one extra entry for use as a sentinel.
+ * See lzx_write_compressed_code().
+ */
+struct lzx_lens {
+        uint8_t main[LZX_MAINCODE_MAX_NUM_SYMBOLS + 1];
+        uint8_t len[LZX_LENCODE_NUM_SYMBOLS + 1];
+        uint8_t aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS];
+};
+
+/* Codewords and lengths for the Huffman codes */
+struct lzx_codes {
+        struct lzx_codewords codewords;
+        struct lzx_lens lens;
+};
+
+/* Symbol frequency counters for the Huffman-encoded alphabets */
+struct lzx_freqs {
+        uint32_t main[LZX_MAINCODE_MAX_NUM_SYMBOLS];
+        uint32_t len[LZX_LENCODE_NUM_SYMBOLS];
+        uint32_t aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS];
+};
+
+/* Block split statistics.  See the "Block splitting algorithm" section later in
+ * this file for details. */
+#define NUM_LITERAL_OBSERVATION_TYPES 8
+#define NUM_MATCH_OBSERVATION_TYPES 2
+#define NUM_OBSERVATION_TYPES (NUM_LITERAL_OBSERVATION_TYPES + \
+                               NUM_MATCH_OBSERVATION_TYPES)
+struct lzx_block_split_stats {
+        uint32_t new_observations[NUM_OBSERVATION_TYPES];
+        uint32_t observations[NUM_OBSERVATION_TYPES];
+        uint32_t num_new_observations;
+        uint32_t num_observations;
+};
+
+/*
+ * Represents a run of literals followed by a match or end-of-block.  This
+ * structure is needed to temporarily store items chosen by the compressor,
+ * since items cannot be written until all items for the block have been chosen
+ * and the block's Huffman codes have been computed.
+ */
+struct attrib_aligned(8) lzx_sequence {
+
+        /*
+         * Bits 9..31: the number of literals in this run.  This may be 0 and
+         * can be at most about SOFT_MAX_BLOCK_LENGTH.  The literals are not
+         * stored explicitly in this structure; instead, they are read directly
+         * from the uncompressed data.
+         *
+         * Bits 0..8: the length of the match which follows the literals, or 0
+         * if this literal run was the last in the block, so there is no match
+         * which follows it.  This can be at most LZX_MAX_MATCH_LEN.
+         */
+        uint32_t litrunlen_and_matchlen;
+#define SEQ_MATCHLEN_BITS        9
+#define SEQ_MATCHLEN_MASK        (((uint32_t)1 << SEQ_MATCHLEN_BITS) - 1)
+
+        /*
+         * If 'matchlen' doesn't indicate end-of-block, then this contains:
+         *
+         * Bits 10..31: either the offset plus LZX_OFFSET_ADJUSTMENT or a recent
+         * offset code, depending on the offset slot encoded in the main symbol.
+         *
+         * Bits 0..9: the main symbol.
+         */
+        uint32_t adjusted_offset_and_mainsym;
+#define SEQ_MAINSYM_BITS        10
+#define SEQ_MAINSYM_MASK        (((uint32_t)1 << SEQ_MAINSYM_BITS) - 1)
+};
+
+/*
+ * This structure represents a byte position in the input buffer and a node in
+ * the graph of possible match/literal choices.
+ *
+ * Logically, each incoming edge to this node is labeled with a literal or a
+ * match that can be taken to reach this position from an earlier position; and
+ * each outgoing edge from this node is labeled with a literal or a match that
+ * can be taken to advance from this position to a later position.
+ */
+struct attrib_aligned(8) lzx_optimum_node {
+
+        /* The cost, in bits, of the lowest-cost path that has been found to
+         * reach this position.  This can change as progressively lower cost
+         * paths are found to reach this position.  */
+        uint32_t cost;
+
+        /*
+         * The best arrival to this node, i.e. the match or literal that was
+         * used to arrive to this position at the given 'cost'.  This can change
+         * as progressively lower cost paths are found to reach this position.
+         *
+         * For non-gap matches, this variable is divided into two bitfields
+         * whose meanings depend on the item type:
+         *
+         * Literals:
+         *        Low bits are 0, high bits are the literal.
+         *
+         * Explicit offset matches:
+         *        Low bits are the match length, high bits are the offset plus
+         *        LZX_OFFSET_ADJUSTMENT.
+         *
+         * Repeat offset matches:
+         *        Low bits are the match length, high bits are the queue index.
+         *
+         * For gap matches, identified by OPTIMUM_GAP_MATCH set, special
+         * behavior applies --- see the code.
+         */
+        uint32_t item;
+#define OPTIMUM_OFFSET_SHIFT        SEQ_MATCHLEN_BITS
+#define OPTIMUM_LEN_MASK        SEQ_MATCHLEN_MASK
+#if CONSIDER_GAP_MATCHES
+#  define OPTIMUM_GAP_MATCH 0x80000000
+#endif
+
+};
+
+/* The cost model for near-optimal parsing */
+struct lzx_costs {
+
+        /*
+         * 'match_cost[offset_slot][len - LZX_MIN_MATCH_LEN]' is the cost of a
+         * length 'len' match which has an offset belonging to 'offset_slot'.
+         * The cost includes the main symbol, the length symbol if required, and
+         * the extra offset bits if any, excluding any entropy-coded bits
+         * (aligned offset bits).  It does *not* include the cost of the aligned
+         * offset symbol which may be required.
+         */
+        uint16_t match_cost[LZX_MAX_OFFSET_SLOTS][LZX_NUM_LENS];
+
+        /* Cost of each symbol in the main code */
+        uint32_t main[LZX_MAINCODE_MAX_NUM_SYMBOLS];
+
+        /* Cost of each symbol in the length code */
+        uint32_t len[LZX_LENCODE_NUM_SYMBOLS];
+
+#if CONSIDER_ALIGNED_COSTS
+        /* Cost of each symbol in the aligned offset code */
+        uint32_t aligned[LZX_ALIGNEDCODE_NUM_SYMBOLS];
+#endif
+};
+
+struct lzx_output_bitstream;
+
+/* The main LZX compressor structure */
+struct liblzx_compressor {
+
+        /* The LZX variant to use */
+        enum liblzx_variant variant;
+
+        /* Output chunk */
+        liblzx_output_chunk_t out_chunk;
+
+        /* True if a flush was requested*/
+        bool flushing;
+
+        /* Memory allocation function */
+        liblzx_alloc_func_t alloc_func;
+
+        /* Memory free function */
+        liblzx_free_func_t free_func;
+
+        /* Memory allocation userdata */
+        void *alloc_userdata;
+
+        /* True if the compressor is outputting the first block */
+        bool first_block;
+
+        /* E8 preprocessor file size */
+        uint32_t e8_file_size;
+
+        /* E8 preprocessor chunk offset */
+        uint32_t e8_chunk_offset;
+
+        /* The buffer for preprocessed input data, if not using destructive
+         * compression */
+        void *in_buffer;
+
+        /* The buffer for output data */
+        void *out_buffer;
+
+        /* Capacity of in_buffer */
+        uint32_t in_buffer_capacity;
+
+        /* Capacity of out_buffer */
+        uint32_t out_buffer_capacity;
+
+        /* Number of prefix bytes currently in in_buffer */
+        uint32_t in_prefix_size;
+
+        /* Number of bytes currently in in_buffer */
+        uint32_t in_used;
+
+        /* Maximum size of a chunk */
+        uint32_t chunk_size;
+
+        /* Pointer to the reset() implementation chosen at allocation time */
+        void (*reset)(struct liblzx_compressor *);
+
+        /* Pointer to the compress() implementation chosen at allocation time */
+        void (*impl)(struct liblzx_compressor *, const uint8_t *, size_t, size_t,
+                     struct lzx_output_bitstream *);
+
+        /* Pointer to the cul() implementation chosen at allocation time */
+        void (*cull)(struct liblzx_compressor *, size_t);
+
+        /* The window size. */
+        uint32_t window_size;
+
+        /* The log base 2 of the window size for match offset encoding purposes.
+         * This will be >= LZX_MIN_WINDOW_ORDER and <= LZX_MAX_WINDOW_ORDER. */
+        unsigned window_order;
+
+        /* The number of symbols in the main alphabet.  This depends on the
+         * window order, since the window order determines the maximum possible
+         * match offset. */
+        unsigned num_main_syms;
+
+        /* The "nice" match length: if a match of this length is found, then it
+         * is chosen immediately without further consideration. */
+        unsigned nice_match_length;
+
+        /* The maximum search depth: at most this many potential matches are
+         * considered at each position. */
+        unsigned max_search_depth;
+
+        /* The number of optimization passes per block */
+        unsigned num_optim_passes;
+
+        /* The symbol frequency counters for the current block */
+        struct lzx_freqs freqs;
+
+        /* Block split statistics for the current block */
+        struct lzx_block_split_stats split_stats;
+
+        /* The Huffman codes for the current and previous blocks.  The one with
+         * index 'codes_index' is for the current block, and the other one is
+         * for the previous block. */
+        struct lzx_codes codes[2];
+        unsigned codes_index;
+
+        /* The matches and literals that the compressor has chosen for the
+         * current block.  The required length of this array is limited by the
+         * maximum number of matches that can ever be chosen for a single block,
+         * plus one for the special entry at the end. */
+        struct lzx_sequence chosen_sequences[
+                       DIV_ROUND_UP(SOFT_MAX_BLOCK_SIZE, LZX_MIN_MATCH_LEN) + 1];
+
+        /* Least-recently-used match queue */
+        uint32_t lru_queue[LZX_NUM_RECENT_OFFSETS];
+
+        /* Next hashes */
+        uint32_t next_hashes[2];
+
+        /* Tables for mapping adjusted offsets to offset slots */
+        uint8_t offset_slot_tab_1[32768]; /* offset slots [0, 29] */
+        uint8_t offset_slot_tab_2[128]; /* offset slots [30, 49] */
+
+        union {
+                /* Data for lzx_compress_lazy() */
+                struct {
+                        /* Hash chains matchfinder (MUST BE LAST!!!) */
+                        union {
+                                struct hc_matchfinder_16 hc_mf_16;
+                                struct hc_matchfinder_32 hc_mf_32;
+                        };
+                };
+
+                /* Data for lzx_compress_near_optimal() */
+                struct {
+                        /*
+                         * Array of nodes, one per position, for running the
+                         * minimum-cost path algorithm.
+                         *
+                         * This array must be large enough to accommodate the
+                         * worst-case number of nodes, which occurs if the
+                         * compressor finds a match of length LZX_MAX_MATCH_LEN
+                         * at position 'SOFT_MAX_BLOCK_SIZE - 1', producing a
+                         * block of size 'SOFT_MAX_BLOCK_SIZE - 1 +
+                         * LZX_MAX_MATCH_LEN'.  Add one for the end-of-block
+                         * node.
+                         */
+                        struct lzx_optimum_node optimum_nodes[
+                                                    SOFT_MAX_BLOCK_SIZE - 1 +
+                                                    LZX_MAX_MATCH_LEN + 1];
+
+                        /* The cost model for the current optimization pass */
+                        struct lzx_costs costs;
+
+                        /*
+                         * Cached matches for the current block.  This array
+                         * contains the matches that were found at each position
+                         * in the block.  Specifically, for each position, there
+                         * is a special 'struct lz_match' whose 'length' field
+                         * contains the number of matches that were found at
+                         * that position; this is followed by the matches
+                         * themselves, if any, sorted by strictly increasing
+                         * length.
+                         *
+                         * Note: in rare cases, there will be a very high number
+                         * of matches in the block and this array will overflow.
+                         * If this happens, we force the end of the current
+                         * block.  CACHE_LENGTH is the length at which we
+                         * actually check for overflow.  The extra slots beyond
+                         * this are enough to absorb the worst case overflow,
+                         * which occurs if starting at &match_cache[CACHE_LENGTH
+                         * - 1], we write the match count header, then write
+                         * MAX_MATCHES_PER_POS matches, then skip searching for
+                         * matches at 'LZX_MAX_MATCH_LEN - 1' positions and
+                         * write the match count header for each.
+                         */
+                        struct lz_match match_cache[CACHE_LENGTH +
+                                                    MAX_MATCHES_PER_POS +
+                                                    LZX_MAX_MATCH_LEN - 1];
+
+                        /* Binary trees matchfinder (MUST BE LAST!!!) */
+                        union {
+                                struct bt_matchfinder_16 bt_mf_16;
+                                struct bt_matchfinder_32 bt_mf_32;
+                        };
+                };
+        };
+};
+
+/******************************************************************************/
+/*                            Matchfinder utilities                           */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * Will a matchfinder using 16-bit positions be sufficient for compressing
+ * buffers of up to the specified size?  The limit could be 65536 bytes, but we
+ * also want to optimize out the use of offset_slot_tab_2 in the 16-bit case.
+ * This requires that the limit be no more than the length of offset_slot_tab_1
+ * (currently 32768).
+ */
+static attrib_forceinline bool
+lzx_is_16_bit(size_t max_bufsize)
+{
+        STATIC_ASSERT(ARRAY_LEN(((struct liblzx_compressor *)0)->offset_slot_tab_1) == 32768);
+        return max_bufsize <= 32768;
+}
+
+/*
+ * Return the offset slot for the specified adjusted match offset.
+ */
+static attrib_forceinline unsigned
+lzx_get_offset_slot(struct liblzx_compressor *c, uint32_t adjusted_offset,
+                    bool is_16_bit)
+{
+        if (__builtin_constant_p(adjusted_offset) &&
+            adjusted_offset < LZX_NUM_RECENT_OFFSETS)
+                return adjusted_offset;
+        if (is_16_bit || adjusted_offset < ARRAY_LEN(c->offset_slot_tab_1))
+                return c->offset_slot_tab_1[adjusted_offset];
+
+        assert((adjusted_offset >> 14) < ARRAY_LEN(c->offset_slot_tab_2));
+
+        return c->offset_slot_tab_2[adjusted_offset >> 14];
+}
+
+/*
+ * For a match that has the specified length and adjusted offset, tally its main
+ * symbol, and if needed its length symbol; then return its main symbol.
+ */
+static attrib_forceinline unsigned
+lzx_tally_main_and_lensyms(struct liblzx_compressor *c, unsigned length,
+                           uint32_t adjusted_offset, bool is_16_bit)
+{
+        unsigned mainsym;
+
+        if (length >= LZX_MIN_SECONDARY_LEN) {
+                /* Length symbol needed */
+                c->freqs.len[length - LZX_MIN_SECONDARY_LEN]++;
+                mainsym = LZX_NUM_CHARS + LZX_NUM_PRIMARY_LENS;
+        } else {
+                /* No length symbol needed */
+                mainsym = LZX_NUM_CHARS + length - LZX_MIN_MATCH_LEN;
+        }
+
+        mainsym += LZX_NUM_LEN_HEADERS *
+                   lzx_get_offset_slot(c, adjusted_offset, is_16_bit);
+        c->freqs.main[mainsym]++;
+        return mainsym;
+}
+
+/*
+ * The following macros call either the 16-bit or the 32-bit version of a
+ * matchfinder function based on the value of 'is_16_bit', which will be known
+ * at compilation time.
+ */
+
+#define CALL_HC_MF(is_16_bit, c, funcname, ...)                                      \
+        ((is_16_bit) ? CONCAT(funcname, _16)(&(c)->hc_mf_16, ##__VA_ARGS__) : \
+                       CONCAT(funcname, _32)(&(c)->hc_mf_32, ##__VA_ARGS__));
+
+#define CALL_BT_MF(is_16_bit, c, funcname, ...)                                      \
+        ((is_16_bit) ? CONCAT(funcname, _16)(&(c)->bt_mf_16, ##__VA_ARGS__) : \
+                       CONCAT(funcname, _32)(&(c)->bt_mf_32, ##__VA_ARGS__));
+
+/******************************************************************************/
+/*                             Output bitstream                               */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * The LZX bitstream is encoded as a sequence of little endian 16-bit coding
+ * units.  Bits are ordered from most significant to least significant within
+ * each coding unit.
+ */
+
+/*
+ * Structure to keep track of the current state of sending bits to the
+ * compressed output buffer.
+ */
+struct lzx_output_bitstream {
+
+        /* Bits that haven't yet been written to the output buffer */
+        machine_word_t bitbuf;
+
+        /* Number of bits currently held in @bitbuf */
+        machine_word_t bitcount;
+
+        /* Pointer to the start of the output buffer */
+        uint8_t *start;
+
+        /* Pointer to the position in the output buffer at which the next coding
+         * unit should be written */
+        uint8_t *next;
+
+        /* Pointer to just past the end of the output buffer, rounded down by
+         * one byte if needed to make 'end - start' a multiple of 2 */
+        uint8_t *end;
+};
+
+/* Can the specified number of bits always be added to 'bitbuf' after all
+ * pending 16-bit coding units have been flushed?  */
+#define CAN_BUFFER(n)        ((n) <= WORDBITS - 15)
+
+/* Initialize the output bitstream to write to the specified buffer. */
+static void
+lzx_init_output(struct lzx_output_bitstream *os, void *buffer, size_t size)
+{
+        os->bitbuf = 0;
+        os->bitcount = 0;
+        os->start = buffer;
+        os->next = buffer;
+        os->end = (uint8_t *)buffer + (size & ~1);
+}
+
+/*
+ * Add some bits to the bitbuffer variable of the output bitstream.  The caller
+ * must make sure there is enough room.
+ */
+static attrib_forceinline void
+lzx_add_bits(struct lzx_output_bitstream *os, uint32_t bits, unsigned num_bits)
+{
+        os->bitbuf = (os->bitbuf << num_bits) | bits;
+        os->bitcount += num_bits;
+}
+
+/*
+ * Flush bits from the bitbuffer variable to the output buffer.  'max_num_bits'
+ * specifies the maximum number of bits that may have been added since the last
+ * flush.
+ */
+static attrib_forceinline void
+lzx_flush_bits(struct lzx_output_bitstream *os, unsigned max_num_bits)
+{
+        /* Masking the number of bits to shift is only needed to avoid undefined
+         * behavior; we don't actually care about the results of bad shifts.  On
+         * x86, the explicit masking generates no extra code.  */
+        const uint32_t shift_mask = WORDBITS - 1;
+
+        if (os->end - os->next < 6)
+                return;
+        put_unaligned_le16(os->bitbuf >> ((os->bitcount - 16) &
+                                            shift_mask), os->next + 0);
+        if (max_num_bits > 16)
+                put_unaligned_le16(os->bitbuf >> ((os->bitcount - 32) &
+                                                shift_mask), os->next + 2);
+        if (max_num_bits > 32)
+                put_unaligned_le16(os->bitbuf >> ((os->bitcount - 48) &
+                                                shift_mask), os->next + 4);
+        os->next += (os->bitcount >> 4) << 1;
+        os->bitcount &= 15;
+}
+
+/* Add at most 16 bits to the bitbuffer and flush it.  */
+static attrib_forceinline void
+lzx_write_bits(struct lzx_output_bitstream *os, uint32_t bits, unsigned num_bits)
+{
+        lzx_add_bits(os, bits, num_bits);
+        lzx_flush_bits(os, 16);
+}
+
+/*
+ * Flush the last coding unit to the output buffer if needed.  Return the total
+ * number of bytes written to the output buffer, or 0 if an overflow occurred.
+ */
+static size_t
+lzx_flush_output(struct lzx_output_bitstream *os)
+{
+        if (os->end - os->next < 6)
+                return 0;
+
+        if (os->bitcount != 0) {
+                put_unaligned_le16(os->bitbuf << (16 - os->bitcount), os->next);
+                os->next += 2;
+        }
+
+        return os->next - os->start;
+}
+
+/******************************************************************************/
+/*                           Preparing Huffman codes                          */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * Build the Huffman codes.  This takes as input the frequency tables for each
+ * code and produces as output a set of tables that map symbols to codewords and
+ * codeword lengths.
+ */
+static void
+lzx_build_huffman_codes(struct liblzx_compressor *c)
+{
+        const struct lzx_freqs *freqs = &c->freqs;
+        struct lzx_codes *codes = &c->codes[c->codes_index];
+
+        STATIC_ASSERT_STMT(MAIN_CODEWORD_LIMIT >= 9 &&
+                      MAIN_CODEWORD_LIMIT <= LZX_MAX_MAIN_CODEWORD_LEN);
+        make_canonical_huffman_code(c->num_main_syms,
+                                    MAIN_CODEWORD_LIMIT,
+                                    freqs->main,
+                                    codes->lens.main,
+                                    codes->codewords.main);
+
+        STATIC_ASSERT_STMT(LENGTH_CODEWORD_LIMIT >= 8 &&
+                      LENGTH_CODEWORD_LIMIT <= LZX_MAX_LEN_CODEWORD_LEN);
+        make_canonical_huffman_code(LZX_LENCODE_NUM_SYMBOLS,
+                                    LENGTH_CODEWORD_LIMIT,
+                                    freqs->len,
+                                    codes->lens.len,
+                                    codes->codewords.len);
+
+        STATIC_ASSERT_STMT(
+            ALIGNED_CODEWORD_LIMIT >= LZX_NUM_ALIGNED_OFFSET_BITS &&
+                      ALIGNED_CODEWORD_LIMIT <= LZX_MAX_ALIGNED_CODEWORD_LEN);
+        make_canonical_huffman_code(LZX_ALIGNEDCODE_NUM_SYMBOLS,
+                                    ALIGNED_CODEWORD_LIMIT,
+                                    freqs->aligned,
+                                    codes->lens.aligned,
+                                    codes->codewords.aligned);
+}
+
+/* Reset the symbol frequencies for the current block. */
+static void
+lzx_reset_symbol_frequencies(struct liblzx_compressor *c)
+{
+        memset(&c->freqs, 0, sizeof(c->freqs));
+}
+
+static unsigned
+lzx_compute_precode_items(const uint8_t * restrict lens,
+                          const uint8_t * restrict prev_lens,
+                          uint32_t * restrict precode_freqs,
+                          unsigned * restrict precode_items)
+{
+        unsigned *itemptr;
+        unsigned run_start;
+        unsigned run_end;
+        unsigned extra_bits;
+        int delta;
+        uint8_t len;
+
+        itemptr = precode_items;
+        run_start = 0;
+
+        while (!((len = lens[run_start]) & 0x80)) {
+
+                /* len = the length being repeated  */
+
+                /* Find the next run of codeword lengths.  */
+
+                run_end = run_start + 1;
+
+                /* Fast case for a single length.  */
+                if (likely(len != lens[run_end])) {
+                        delta = prev_lens[run_start] - len;
+                        if (delta < 0)
+                                delta += 17;
+                        precode_freqs[delta]++;
+                        *itemptr++ = delta;
+                        run_start++;
+                        continue;
+                }
+
+                /* Extend the run.  */
+                do {
+                        run_end++;
+                } while (len == lens[run_end]);
+
+                if (len == 0) {
+                        /* Run of zeroes.  */
+
+                        /* Symbol 18: RLE 20 to 51 zeroes at a time.  */
+                        while ((run_end - run_start) >= 20) {
+                                extra_bits =
+                                    min_uint((run_end - run_start) - 20, 0x1F);
+                                precode_freqs[18]++;
+                                *itemptr++ = 18 | (extra_bits << 5);
+                                run_start += 20 + extra_bits;
+                        }
+
+                        /* Symbol 17: RLE 4 to 19 zeroes at a time.  */
+                        if ((run_end - run_start) >= 4) {
+                                extra_bits =
+                                    min_uint((run_end - run_start) - 4, 0xF);
+                                precode_freqs[17]++;
+                                *itemptr++ = 17 | (extra_bits << 5);
+                                run_start += 4 + extra_bits;
+                        }
+                } else {
+
+                        /* A run of nonzero lengths. */
+
+                        /* Symbol 19: RLE 4 to 5 of any length at a time.  */
+                        while ((run_end - run_start) >= 4) {
+                                extra_bits = (run_end - run_start) > 4;
+                                delta = prev_lens[run_start] - len;
+                                if (delta < 0)
+                                        delta += 17;
+                                precode_freqs[19]++;
+                                precode_freqs[delta]++;
+                                *itemptr++ = 19 | (extra_bits << 5) | (delta << 6);
+                                run_start += 4 + extra_bits;
+                        }
+                }
+
+                /* Output any remaining lengths without RLE.  */
+                while (run_start != run_end) {
+                        delta = prev_lens[run_start] - len;
+                        if (delta < 0)
+                                delta += 17;
+                        precode_freqs[delta]++;
+                        *itemptr++ = delta;
+                        run_start++;
+                }
+        }
+
+        return itemptr - precode_items;
+}
+
+/******************************************************************************/
+/*                          Outputting compressed data                        */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * Output a Huffman code in the compressed form used in LZX.
+ *
+ * The Huffman code is represented in the output as a logical series of codeword
+ * lengths from which the Huffman code, which must be in canonical form, can be
+ * reconstructed.
+ *
+ * The codeword lengths are themselves compressed using a separate Huffman code,
+ * the "precode", which contains a symbol for each possible codeword length in
+ * the larger code as well as several special symbols to represent repeated
+ * codeword lengths (a form of run-length encoding).  The precode is itself
+ * constructed in canonical form, and its codeword lengths are represented
+ * literally in 20 4-bit fields that immediately precede the compressed codeword
+ * lengths of the larger code.
+ *
+ * Furthermore, the codeword lengths of the larger code are actually represented
+ * as deltas from the codeword lengths of the corresponding code in the previous
+ * block.
+ *
+ * @os:
+ *        Bitstream to which to write the compressed Huffman code.
+ * @lens:
+ *        The codeword lengths, indexed by symbol, in the Huffman code.
+ * @prev_lens:
+ *        The codeword lengths, indexed by symbol, in the corresponding Huffman
+ *        code in the previous block, or all zeroes if this is the first block.
+ * @num_lens:
+ *        The number of symbols in the Huffman code.
+ */
+static void
+lzx_write_compressed_code(struct lzx_output_bitstream *os,
+                          const uint8_t * restrict lens,
+                          const uint8_t * restrict prev_lens,
+                          unsigned num_lens)
+{
+        uint32_t precode_freqs[LZX_PRECODE_NUM_SYMBOLS];
+        uint8_t precode_lens[LZX_PRECODE_NUM_SYMBOLS];
+        uint32_t precode_codewords[LZX_PRECODE_NUM_SYMBOLS];
+        unsigned *precode_items = (unsigned *)alloca(sizeof(unsigned) * num_lens);
+        unsigned num_precode_items;
+        unsigned precode_item;
+        unsigned precode_sym;
+        unsigned i;
+        uint8_t saved = lens[num_lens];
+        *(uint8_t *)(lens + num_lens) = 0x80;
+
+        for (i = 0; i < LZX_PRECODE_NUM_SYMBOLS; i++)
+                precode_freqs[i] = 0;
+
+        /* Compute the "items" (RLE / literal tokens and extra bits) with which
+         * the codeword lengths in the larger code will be output.  */
+        num_precode_items = lzx_compute_precode_items(lens,
+                                                      prev_lens,
+                                                      precode_freqs,
+                                                      precode_items);
+
+        /* Build the precode.  */
+        STATIC_ASSERT_STMT(PRE_CODEWORD_LIMIT >= 5 &&
+                      PRE_CODEWORD_LIMIT <= LZX_MAX_PRE_CODEWORD_LEN);
+        make_canonical_huffman_code(LZX_PRECODE_NUM_SYMBOLS, PRE_CODEWORD_LIMIT,
+                                    precode_freqs, precode_lens,
+                                    precode_codewords);
+
+        /* Output the lengths of the codewords in the precode.  */
+        for (i = 0; i < LZX_PRECODE_NUM_SYMBOLS; i++)
+                lzx_write_bits(os, precode_lens[i], LZX_PRECODE_ELEMENT_SIZE);
+
+        /* Output the encoded lengths of the codewords in the larger code.  */
+        for (i = 0; i < num_precode_items; i++) {
+                precode_item = precode_items[i];
+                precode_sym = precode_item & 0x1F;
+                lzx_add_bits(os, precode_codewords[precode_sym],
+                             precode_lens[precode_sym]);
+                if (precode_sym >= 17) {
+                        if (precode_sym == 17) {
+                                lzx_add_bits(os, precode_item >> 5, 4);
+                        } else if (precode_sym == 18) {
+                                lzx_add_bits(os, precode_item >> 5, 5);
+                        } else {
+                                lzx_add_bits(os, (precode_item >> 5) & 1, 1);
+                                precode_sym = precode_item >> 6;
+                                lzx_add_bits(os, precode_codewords[precode_sym],
+                                             precode_lens[precode_sym]);
+                        }
+                }
+                STATIC_ASSERT_STMT(CAN_BUFFER(2 * PRE_CODEWORD_LIMIT + 1));
+                lzx_flush_bits(os, 2 * PRE_CODEWORD_LIMIT + 1);
+        }
+
+        *(uint8_t *)(lens + num_lens) = saved;
+}
+
+/*
+ * Write all matches and literal bytes (which were precomputed) in an LZX
+ * compressed block to the output bitstream in the final compressed
+ * representation.
+ *
+ * @os
+ *        The output bitstream.
+ * @block_type
+ *        The chosen type of the LZX compressed block (LZX_BLOCKTYPE_ALIGNED or
+ *        LZX_BLOCKTYPE_VERBATIM).
+ * @block_data
+ *        The uncompressed data of the block.
+ * @sequences
+ *        The matches and literals to output, given as a series of sequences.
+ * @codes
+ *        The main, length, and aligned offset Huffman codes for the block.
+ */
+static void
+lzx_write_sequences(struct lzx_output_bitstream *os, int block_type,
+                    const uint8_t *block_data, const struct lzx_sequence sequences[],
+                    const struct lzx_codes *codes)
+{
+        const struct lzx_sequence *seq = sequences;
+        unsigned min_aligned_offset_slot;
+
+        if (block_type == LZX_BLOCKTYPE_ALIGNED)
+                min_aligned_offset_slot = LZX_MIN_ALIGNED_OFFSET_SLOT;
+        else
+                min_aligned_offset_slot = LZX_MAX_OFFSET_SLOTS;
+
+        for (;;) {
+                /* Output the next sequence.  */
+
+                uint32_t litrunlen = seq->litrunlen_and_matchlen >> SEQ_MATCHLEN_BITS;
+                unsigned matchlen = seq->litrunlen_and_matchlen & SEQ_MATCHLEN_MASK;
+                STATIC_ASSERT((uint32_t)~SEQ_MATCHLEN_MASK >> SEQ_MATCHLEN_BITS >=
+                              SOFT_MAX_BLOCK_SIZE);
+                uint32_t adjusted_offset;
+                unsigned main_symbol;
+                unsigned offset_slot;
+                unsigned num_extra_bits;
+                uint32_t extra_bits;
+
+                /* Output the literal run of the sequence.  */
+
+                if (litrunlen) {  /* Is the literal run nonempty?  */
+
+                        /* Verify optimization is enabled on 64-bit  */
+                        STATIC_ASSERT(WORDBITS < 64 ||
+                                      CAN_BUFFER(3 * MAIN_CODEWORD_LIMIT));
+
+                        if (CAN_BUFFER(3 * MAIN_CODEWORD_LIMIT)) {
+
+                                /* 64-bit: write 3 literals at a time.  */
+                                while (litrunlen >= 3) {
+                                        unsigned lit0 = block_data[0];
+                                        unsigned lit1 = block_data[1];
+                                        unsigned lit2 = block_data[2];
+                                        lzx_add_bits(os, codes->codewords.main[lit0],
+                                                     codes->lens.main[lit0]);
+                                        lzx_add_bits(os, codes->codewords.main[lit1],
+                                                     codes->lens.main[lit1]);
+                                        lzx_add_bits(os, codes->codewords.main[lit2],
+                                                     codes->lens.main[lit2]);
+                                        lzx_flush_bits(os, 3 * MAIN_CODEWORD_LIMIT);
+                                        block_data += 3;
+                                        litrunlen -= 3;
+                                }
+                                if (litrunlen--) {
+                                        unsigned lit = *block_data++;
+                                        lzx_add_bits(os, codes->codewords.main[lit],
+                                                     codes->lens.main[lit]);
+                                        if (litrunlen--) {
+                                                unsigned lit = *block_data++;
+                                                lzx_add_bits(os, codes->codewords.main[lit],
+                                                             codes->lens.main[lit]);
+                                                lzx_flush_bits(os, 2 * MAIN_CODEWORD_LIMIT);
+                                        } else {
+                                                lzx_flush_bits(os, 1 * MAIN_CODEWORD_LIMIT);
+                                        }
+                                }
+                        } else {
+                                /* 32-bit: write 1 literal at a time.  */
+                                do {
+                                        unsigned lit = *block_data++;
+                                        lzx_add_bits(os, codes->codewords.main[lit],
+                                                     codes->lens.main[lit]);
+                                        lzx_flush_bits(os, MAIN_CODEWORD_LIMIT);
+                                } while (--litrunlen);
+                        }
+                }
+
+                /* Was this the last literal run?  */
+                if (matchlen == 0)
+                        return;
+
+                /* Nope; output the match.  */
+
+                block_data += matchlen;
+
+                adjusted_offset = seq->adjusted_offset_and_mainsym >> SEQ_MAINSYM_BITS;
+                main_symbol = seq->adjusted_offset_and_mainsym & SEQ_MAINSYM_MASK;
+
+                offset_slot = (main_symbol - LZX_NUM_CHARS) / LZX_NUM_LEN_HEADERS;
+                num_extra_bits = lzx_extra_offset_bits[offset_slot];
+                extra_bits = adjusted_offset - (lzx_offset_slot_base[offset_slot] +
+                                                LZX_OFFSET_ADJUSTMENT);
+
+        #define MAX_MATCH_BITS (MAIN_CODEWORD_LIMIT +                \
+                                LENGTH_CODEWORD_LIMIT +                \
+                                LZX_MAX_NUM_EXTRA_BITS -        \
+                                LZX_NUM_ALIGNED_OFFSET_BITS +        \
+                                ALIGNED_CODEWORD_LIMIT)
+
+                /* Verify optimization is enabled on 64-bit  */
+                STATIC_ASSERT_STMT(WORDBITS < 64 || CAN_BUFFER(MAX_MATCH_BITS));
+
+                /* Output the main symbol for the match.  */
+
+                lzx_add_bits(os, codes->codewords.main[main_symbol],
+                             codes->lens.main[main_symbol]);
+                if (!CAN_BUFFER(MAX_MATCH_BITS))
+                        lzx_flush_bits(os, MAIN_CODEWORD_LIMIT);
+
+                /* If needed, output the length symbol for the match.  */
+
+                if (matchlen >= LZX_MIN_SECONDARY_LEN) {
+                        lzx_add_bits(os, codes->codewords.len[matchlen -
+                                                              LZX_MIN_SECONDARY_LEN],
+                                     codes->lens.len[matchlen -
+                                                     LZX_MIN_SECONDARY_LEN]);
+                        if (!CAN_BUFFER(MAX_MATCH_BITS))
+                                lzx_flush_bits(os, LENGTH_CODEWORD_LIMIT);
+                }
+
+                /* Output the extra offset bits for the match.  In aligned
+                 * offset blocks, the lowest 3 bits of the adjusted offset are
+                 * Huffman-encoded using the aligned offset code, provided that
+                 * there are at least extra 3 offset bits required.  All other
+                 * extra offset bits are output verbatim.  */
+
+                if (offset_slot >= min_aligned_offset_slot) {
+
+                        lzx_add_bits(os, extra_bits >> LZX_NUM_ALIGNED_OFFSET_BITS,
+                                     num_extra_bits - LZX_NUM_ALIGNED_OFFSET_BITS);
+                        if (!CAN_BUFFER(MAX_MATCH_BITS))
+                                lzx_flush_bits(os, LZX_MAX_NUM_EXTRA_BITS -
+                                                   LZX_NUM_ALIGNED_OFFSET_BITS);
+
+                        lzx_add_bits(os, codes->codewords.aligned[adjusted_offset &
+                                                                  LZX_ALIGNED_OFFSET_BITMASK],
+                                     codes->lens.aligned[adjusted_offset &
+                                                         LZX_ALIGNED_OFFSET_BITMASK]);
+                        if (!CAN_BUFFER(MAX_MATCH_BITS))
+                                lzx_flush_bits(os, ALIGNED_CODEWORD_LIMIT);
+                } else {
+                        STATIC_ASSERT(CAN_BUFFER(LZX_MAX_NUM_EXTRA_BITS));
+
+                        lzx_add_bits(os, extra_bits, num_extra_bits);
+                        if (!CAN_BUFFER(MAX_MATCH_BITS))
+                                lzx_flush_bits(os, LZX_MAX_NUM_EXTRA_BITS);
+                }
+
+                if (CAN_BUFFER(MAX_MATCH_BITS))
+                        lzx_flush_bits(os, MAX_MATCH_BITS);
+
+                /* Advance to the next sequence.  */
+                seq++;
+        }
+}
+
+static void
+lzx_write_header(uint32_t e8_file_size, struct lzx_output_bitstream *os)
+{
+        if (e8_file_size == 0) {
+                lzx_write_bits(os, 0, 1);
+        } else {
+                lzx_write_bits(os, 1, 1);
+                lzx_write_bits(os, (e8_file_size >> 16) & 0xffffu, 16);
+                lzx_write_bits(os, e8_file_size & 0xffffu, 16);
+        }
+}
+
+static void
+lzx_write_compressed_block(const uint8_t *block_begin,
+                           int block_type,
+                           uint32_t block_size,
+                           enum liblzx_variant variant,
+                           unsigned window_order,
+                           unsigned num_main_syms,
+                           const struct lzx_sequence sequences[],
+                           const struct lzx_codes * codes,
+                           const struct lzx_lens * prev_lens,
+                           struct lzx_output_bitstream * os)
+{
+        /* The first three bits indicate the type of block and are one of the
+         * LZX_BLOCKTYPE_* constants.  */
+        lzx_write_bits(os, block_type, 3);
+
+        /*
+         * Output the block size.
+         *
+         * The original LZX format encoded the block size in 24 bits.  However,
+         * the LZX format used in WIM archives uses 1 bit to specify whether the
+         * block has the default size of 32768 bytes, then optionally 16 bits to
+         * specify a non-default size.  This works fine for Microsoft's WIM
+         * software (WIMGAPI), which never compresses more than 32768 bytes at a
+         * time with LZX.  However, as an extension, our LZX compressor supports
+         * compressing up to 2097152 bytes, with a corresponding increase in
+         * window size.  It is possible for blocks in these larger buffers to
+         * exceed 65535 bytes; such blocks cannot have their size represented in
+         * 16 bits.
+         *
+         * The chosen solution was to use 24 bits for the block size when
+         * possibly required --- specifically, when the compressor has been
+         * allocated to be capable of compressing more than 32768 bytes at once
+         * (which also causes the number of main symbols to be increased).
+         */
+        if (variant == LIBLZX_VARIANT_WIM) {
+                if (block_size == LZX_DEFAULT_BLOCK_SIZE) {
+                        lzx_write_bits(os, 1, 1);
+                } else {
+                        lzx_write_bits(os, 0, 1);
+
+                        if (window_order >= 16)
+                                lzx_write_bits(os, block_size >> 16, 8);
+
+                        lzx_write_bits(os, block_size & 0xFFFF, 16);
+                }
+        } else {
+                lzx_write_bits(os, block_size >> 16, 8);
+                lzx_write_bits(os, block_size & 0xFFFF, 16);
+        }
+
+        /* If it's an aligned offset block, output the aligned offset code.  */
+        if (block_type == LZX_BLOCKTYPE_ALIGNED) {
+                for (int i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) {
+                        lzx_write_bits(os, codes->lens.aligned[i],
+                                       LZX_ALIGNEDCODE_ELEMENT_SIZE);
+                }
+        }
+
+        /* Output the main code (two parts).  */
+        lzx_write_compressed_code(os, codes->lens.main,
+                                  prev_lens->main,
+                                  LZX_NUM_CHARS);
+        lzx_write_compressed_code(os, codes->lens.main + LZX_NUM_CHARS,
+                                  prev_lens->main + LZX_NUM_CHARS,
+                                  num_main_syms - LZX_NUM_CHARS);
+
+        /* Output the length code.  */
+        lzx_write_compressed_code(os, codes->lens.len,
+                                  prev_lens->len,
+                                  LZX_LENCODE_NUM_SYMBOLS);
+
+        /* Output the compressed matches and literals.  */
+        lzx_write_sequences(os, block_type, block_begin, sequences, codes);
+}
+
+/*
+ * Given the frequencies of symbols in an LZX-compressed block and the
+ * corresponding Huffman codes, return LZX_BLOCKTYPE_ALIGNED or
+ * LZX_BLOCKTYPE_VERBATIM if an aligned offset or verbatim block, respectively,
+ * will take fewer bits to output.
+ */
+static int
+lzx_choose_verbatim_or_aligned(const struct lzx_freqs * freqs,
+                               const struct lzx_codes * codes)
+{
+        uint32_t verbatim_cost = 0;
+        uint32_t aligned_cost = 0;
+
+        /* A verbatim block requires 3 bits in each place that an aligned offset
+         * symbol would be used in an aligned offset block.  */
+        for (unsigned i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) {
+                verbatim_cost += LZX_NUM_ALIGNED_OFFSET_BITS * freqs->aligned[i];
+                aligned_cost += codes->lens.aligned[i] * freqs->aligned[i];
+        }
+
+        /* Account for the cost of sending the codeword lengths of the aligned
+         * offset code.  */
+        aligned_cost += LZX_ALIGNEDCODE_ELEMENT_SIZE *
+                        LZX_ALIGNEDCODE_NUM_SYMBOLS;
+
+        if (aligned_cost < verbatim_cost)
+                return LZX_BLOCKTYPE_ALIGNED;
+        else
+                return LZX_BLOCKTYPE_VERBATIM;
+}
+
+/*
+ * Flush an LZX block:
+ *
+ * 1. Build the Huffman codes.
+ * 2. Decide whether to output the block as VERBATIM or ALIGNED.
+ * 3. Write the block.
+ * 4. Swap the indices of the current and previous Huffman codes.
+ *
+ * Note: we never output UNCOMPRESSED blocks.  This probably should be
+ * implemented sometime, but it doesn't make much difference.
+ */
+static void
+lzx_flush_block(struct liblzx_compressor *c, struct lzx_output_bitstream *os,
+                const uint8_t *block_begin, uint32_t block_size, uint32_t seq_idx)
+{
+        int block_type;
+
+        lzx_build_huffman_codes(c);
+
+        block_type = lzx_choose_verbatim_or_aligned(&c->freqs,
+                                                    &c->codes[c->codes_index]);
+
+        if (c->variant != LIBLZX_VARIANT_WIM) {
+                if (c->first_block) {
+                        lzx_write_header(c->e8_file_size, os);
+                        c->first_block = false;
+                }
+        }
+
+        lzx_write_compressed_block(block_begin,
+                                   block_type,
+                                   block_size,
+                                   c->variant,
+                                   c->window_order,
+                                   c->num_main_syms,
+                                   &c->chosen_sequences[seq_idx],
+                                   &c->codes[c->codes_index],
+                                   &c->codes[c->codes_index ^ 1].lens,
+                                   os);
+        c->codes_index ^= 1;
+}
+
+/******************************************************************************/
+/*                          Block splitting algorithm                         */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * The problem of block splitting is to decide when it is worthwhile to start a
+ * new block with new entropy codes.  There is a theoretically optimal solution:
+ * recursively consider every possible block split, considering the exact cost
+ * of each block, and choose the minimum cost approach.  But this is far too
+ * slow.  Instead, as an approximation, we can count symbols and after every N
+ * symbols, compare the expected distribution of symbols based on the previous
+ * data with the actual distribution.  If they differ "by enough", then start a
+ * new block.
+ *
+ * As an optimization and heuristic, we don't distinguish between every symbol
+ * but rather we combine many symbols into a single "observation type".  For
+ * literals we only look at the high bits and low bits, and for matches we only
+ * look at whether the match is long or not.  The assumption is that for typical
+ * "real" data, places that are good block boundaries will tend to be noticeable
+ * based only on changes in these aggregate frequencies, without looking for
+ * subtle differences in individual symbols.  For example, a change from ASCII
+ * bytes to non-ASCII bytes, or from few matches (generally less compressible)
+ * to many matches (generally more compressible), would be easily noticed based
+ * on the aggregates.
+ *
+ * For determining whether the frequency distributions are "different enough" to
+ * start a new block, the simply heuristic of splitting when the sum of absolute
+ * differences exceeds a constant seems to be good enough.
+ *
+ * Finally, for an approximation, it is not strictly necessary that the exact
+ * symbols being used are considered.  With "near-optimal parsing", for example,
+ * the actual symbols that will be used are unknown until after the block
+ * boundary is chosen and the block has been optimized.  Since the final choices
+ * cannot be used, we can use preliminary "greedy" choices instead.
+ */
+
+/* Initialize the block split statistics when starting a new block. */
+static void
+lzx_init_block_split_stats(struct lzx_block_split_stats *stats)
+{
+        memset(stats, 0, sizeof(*stats));
+}
+
+/* Literal observation.  Heuristic: use the top 2 bits and low 1 bits of the
+ * literal, for 8 possible literal observation types.  */
+static attrib_forceinline void
+lzx_observe_literal(struct lzx_block_split_stats *stats, uint8_t lit)
+{
+        stats->new_observations[((lit >> 5) & 0x6) | (lit & 1)]++;
+        stats->num_new_observations++;
+}
+
+/* Match observation.  Heuristic: use one observation type for "short match" and
+ * one observation type for "long match".  */
+static attrib_forceinline void
+lzx_observe_match(struct lzx_block_split_stats *stats, unsigned length)
+{
+        stats->new_observations[NUM_LITERAL_OBSERVATION_TYPES + (length >= 5)]++;
+        stats->num_new_observations++;
+}
+
+static bool
+lzx_should_end_block(struct lzx_block_split_stats *stats)
+{
+        if (stats->num_observations > 0) {
+
+                /* Note: to avoid slow divisions, we do not divide by
+                 * 'num_observations', but rather do all math with the numbers
+                 * multiplied by 'num_observations'. */
+                uint32_t total_delta = 0;
+                for (int i = 0; i < NUM_OBSERVATION_TYPES; i++) {
+                        uint32_t expected = stats->observations[i] *
+                                       stats->num_new_observations;
+                        uint32_t actual = stats->new_observations[i] *
+                                     stats->num_observations;
+                        uint32_t delta = (actual > expected) ? actual - expected :
+                                                          expected - actual;
+                        total_delta += delta;
+                }
+
+                /* Ready to end the block? */
+                if (total_delta >=
+                    stats->num_new_observations * 7 / 8 * stats->num_observations)
+                        return true;
+        }
+
+        for (int i = 0; i < NUM_OBSERVATION_TYPES; i++) {
+                stats->num_observations += stats->new_observations[i];
+                stats->observations[i] += stats->new_observations[i];
+                stats->new_observations[i] = 0;
+        }
+        stats->num_new_observations = 0;
+        return false;
+}
+
+/******************************************************************************/
+/*                   Slower ("near-optimal") compression algorithm            */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * Least-recently-used queue for match offsets.
+ *
+ * This is represented as a 64-bit integer for efficiency.  There are three
+ * offsets of 21 bits each.  Bit 64 is garbage.
+ */
+struct attrib_aligned(8) lzx_lru_queue {
+        uint64_t R;
+};
+
+#define LZX_QUEUE_OFFSET_SHIFT        21
+#define LZX_QUEUE_OFFSET_MASK        (((uint64_t)1 << LZX_QUEUE_OFFSET_SHIFT) - 1)
+
+#define LZX_QUEUE_R0_SHIFT (0 * LZX_QUEUE_OFFSET_SHIFT)
+#define LZX_QUEUE_R1_SHIFT (1 * LZX_QUEUE_OFFSET_SHIFT)
+#define LZX_QUEUE_R2_SHIFT (2 * LZX_QUEUE_OFFSET_SHIFT)
+
+#define LZX_QUEUE_R0_MASK (LZX_QUEUE_OFFSET_MASK << LZX_QUEUE_R0_SHIFT)
+#define LZX_QUEUE_R1_MASK (LZX_QUEUE_OFFSET_MASK << LZX_QUEUE_R1_SHIFT)
+#define LZX_QUEUE_R2_MASK (LZX_QUEUE_OFFSET_MASK << LZX_QUEUE_R2_SHIFT)
+
+static attrib_forceinline uint64_t
+lzx_lru_queue_R0(struct lzx_lru_queue queue)
+{
+        return (queue.R >> LZX_QUEUE_R0_SHIFT) & LZX_QUEUE_OFFSET_MASK;
+}
+
+static attrib_forceinline uint64_t
+lzx_lru_queue_R1(struct lzx_lru_queue queue)
+{
+        return (queue.R >> LZX_QUEUE_R1_SHIFT) & LZX_QUEUE_OFFSET_MASK;
+}
+
+static attrib_forceinline uint64_t
+lzx_lru_queue_R2(struct lzx_lru_queue queue)
+{
+        return (queue.R >> LZX_QUEUE_R2_SHIFT) & LZX_QUEUE_OFFSET_MASK;
+}
+
+static attrib_forceinline void
+lzx_lru_queue_save(uint32_t * restrict out_queue,
+                   const struct lzx_lru_queue * restrict in_queue)
+{
+        struct lzx_lru_queue queue = *in_queue;
+        out_queue[0] = lzx_lru_queue_R0(queue);
+        out_queue[1] = lzx_lru_queue_R1(queue);
+        out_queue[2] = lzx_lru_queue_R2(queue);
+}
+
+static attrib_forceinline void
+lzx_lru_queue_load(struct lzx_lru_queue *restrict out_queue,
+                   const uint32_t *restrict in_queue)
+{
+        uint64_t r = 0;
+        r |= (uint64_t)(in_queue[0]) << LZX_QUEUE_R0_SHIFT;
+        r |= (uint64_t)(in_queue[1]) << LZX_QUEUE_R1_SHIFT;
+        r |= (uint64_t)(in_queue[2]) << LZX_QUEUE_R2_SHIFT;
+        out_queue->R = r;
+}
+
+/* Push a match offset onto the front (most recently used) end of the queue.  */
+static attrib_forceinline struct lzx_lru_queue
+lzx_lru_queue_push(struct lzx_lru_queue queue, uint32_t offset)
+{
+        return (struct lzx_lru_queue) {
+                .R = (queue.R << LZX_QUEUE_OFFSET_SHIFT) | offset,
+        };
+}
+
+/* Swap a match offset to the front of the queue.  */
+static attrib_forceinline struct lzx_lru_queue
+lzx_lru_queue_swap(struct lzx_lru_queue queue, unsigned idx)
+{
+        unsigned shift = idx * 21;
+        const uint64_t mask = LZX_QUEUE_R0_MASK;
+        const uint64_t mask_high = mask << shift;
+
+        return (struct lzx_lru_queue) {
+                (queue.R & ~(mask | mask_high)) |
+                ((queue.R & mask_high) >> shift) |
+                ((queue.R & mask) << shift)
+        };
+}
+
+static attrib_forceinline uint32_t
+lzx_walk_item_list(struct liblzx_compressor *c, uint32_t block_size, bool is_16_bit,
+                   bool record)
+{
+        struct lzx_sequence *seq =
+                &c->chosen_sequences[ARRAY_LEN(c->chosen_sequences) - 1];
+        uint32_t node_idx = block_size;
+        uint32_t litrun_end; /* if record=true: end of the current literal run */
+
+        if (record) {
+                /* The last sequence has matchlen 0 */
+                seq->litrunlen_and_matchlen = 0;
+                litrun_end = node_idx;
+        }
+
+        for (;;) {
+                uint32_t item;
+                unsigned matchlen;
+                uint32_t adjusted_offset;
+                unsigned mainsym;
+
+                /* Tally literals until either a match or the beginning of the
+                 * block is reached.  Note: the item in the node at the
+                 * beginning of the block (c->optimum_nodes[0]) has all bits
+                 * set, causing this loop to end when it is reached. */
+                for (;;) {
+                        item = c->optimum_nodes[node_idx].item;
+                        if (item & OPTIMUM_LEN_MASK)
+                                break;
+                        c->freqs.main[item >> OPTIMUM_OFFSET_SHIFT]++;
+                        node_idx--;
+                }
+
+        #if CONSIDER_GAP_MATCHES
+                if (item & OPTIMUM_GAP_MATCH) {
+                        if (node_idx == 0)
+                                break;
+                        /* Tally/record the rep0 match after the gap. */
+                        matchlen = item & OPTIMUM_LEN_MASK;
+                        mainsym = lzx_tally_main_and_lensyms(c, matchlen, 0,
+                                                             is_16_bit);
+                        if (record) {
+                                seq->litrunlen_and_matchlen |=
+                                        (litrun_end - node_idx) <<
+                                         SEQ_MATCHLEN_BITS;
+                                seq--;
+                                seq->litrunlen_and_matchlen = matchlen;
+                                seq->adjusted_offset_and_mainsym = mainsym;
+                                litrun_end = node_idx - matchlen;
+                        }
+
+                        /* Tally the literal in the gap. */
+                        c->freqs.main[(uint8_t)(item >> OPTIMUM_OFFSET_SHIFT)]++;
+
+                        /* Fall through and tally the match before the gap.
+                         * (It was temporarily saved in the 'cost' field of the
+                         * previous node, which was free to reuse.) */
+                        item = c->optimum_nodes[--node_idx].cost;
+                        node_idx -= matchlen;
+                }
+        #else /* CONSIDER_GAP_MATCHES */
+                if (node_idx == 0)
+                        break;
+        #endif /* !CONSIDER_GAP_MATCHES */
+
+                /* Tally/record a match. */
+                matchlen = item & OPTIMUM_LEN_MASK;
+                adjusted_offset = item >> OPTIMUM_OFFSET_SHIFT;
+                mainsym = lzx_tally_main_and_lensyms(c, matchlen,
+                                                     adjusted_offset,
+                                                     is_16_bit);
+                if (adjusted_offset >= LZX_MIN_ALIGNED_OFFSET +
+                                       LZX_OFFSET_ADJUSTMENT)
+                        c->freqs.aligned[adjusted_offset &
+                                         LZX_ALIGNED_OFFSET_BITMASK]++;
+                if (record) {
+                        seq->litrunlen_and_matchlen |=
+                                (litrun_end - node_idx) << SEQ_MATCHLEN_BITS;
+                        seq--;
+                        seq->litrunlen_and_matchlen = matchlen;
+                        seq->adjusted_offset_and_mainsym =
+                                (adjusted_offset << SEQ_MAINSYM_BITS) | mainsym;
+                        litrun_end = node_idx - matchlen;
+                }
+                node_idx -= matchlen;
+        }
+
+        /* Record the literal run length for the first sequence. */
+        if (record) {
+                seq->litrunlen_and_matchlen |=
+                        (litrun_end - node_idx) << SEQ_MATCHLEN_BITS;
+        }
+
+        /* Return the index in chosen_sequences at which the sequences begin. */
+        return seq - &c->chosen_sequences[0];
+}
+
+/*
+ * Given the minimum-cost path computed through the item graph for the current
+ * block, walk the path and count how many of each symbol in each Huffman-coded
+ * alphabet would be required to output the items (matches and literals) along
+ * the path.
+ *
+ * Note that the path will be walked backwards (from the end of the block to the
+ * beginning of the block), but this doesn't matter because this function only
+ * computes frequencies.
+ */
+static attrib_forceinline void
+lzx_tally_item_list(struct liblzx_compressor *c, uint32_t block_size, bool is_16_bit)
+{
+        lzx_walk_item_list(c, block_size, is_16_bit, false);
+}
+
+/*
+ * Like lzx_tally_item_list(), but this function also generates the list of
+ * lzx_sequences for the minimum-cost path and writes it to c->chosen_sequences,
+ * ready to be output to the bitstream after the Huffman codes are computed.
+ * The lzx_sequences will be written to decreasing memory addresses as the path
+ * is walked backwards, which means they will end up in the expected
+ * first-to-last order.  The return value is the index in c->chosen_sequences at
+ * which the lzx_sequences begin.
+ */
+static attrib_forceinline uint32_t
+lzx_record_item_list(struct liblzx_compressor *c, uint32_t block_size, bool is_16_bit)
+{
+        return lzx_walk_item_list(c, block_size, is_16_bit, true);
+}
+
+/*
+ * Find an inexpensive path through the graph of possible match/literal choices
+ * for the current block.  The nodes of the graph are
+ * c->optimum_nodes[0...block_size].  They correspond directly to the bytes in
+ * the current block, plus one extra node for end-of-block.  The edges of the
+ * graph are matches and literals.  The goal is to find the minimum cost path
+ * from 'c->optimum_nodes[0]' to 'c->optimum_nodes[block_size]', given the cost
+ * model 'c->costs'.
+ *
+ * The algorithm works forwards, starting at 'c->optimum_nodes[0]' and
+ * proceeding forwards one node at a time.  At each node, a selection of matches
+ * (len >= 2), as well as the literal byte (len = 1), is considered.  An item of
+ * length 'len' provides a new path to reach the node 'len' bytes later.  If
+ * such a path is the lowest cost found so far to reach that later node, then
+ * that later node is updated with the new cost and the "arrival" which provided
+ * that cost.
+ *
+ * Note that although this algorithm is based on minimum cost path search, due
+ * to various simplifying assumptions the result is not guaranteed to be the
+ * true minimum cost, or "optimal", path over the graph of all valid LZX
+ * representations of this block.
+ *
+ * Also, note that because of the presence of the recent offsets queue (which is
+ * a type of adaptive state), the algorithm cannot work backwards and compute
+ * "cost to end" instead of "cost to beginning".  Furthermore, the way the
+ * algorithm handles this adaptive state in the "minimum cost" parse is actually
+ * only an approximation.  It's possible for the globally optimal, minimum cost
+ * path to contain a prefix, ending at a position, where that path prefix is
+ * *not* the minimum cost path to that position.  This can happen if such a path
+ * prefix results in a different adaptive state which results in lower costs
+ * later.  The algorithm does not solve this problem in general; it only looks
+ * one step ahead, with the exception of special consideration for "gap
+ * matches".
+ */
+static attrib_forceinline struct lzx_lru_queue
+lzx_find_min_cost_path(struct liblzx_compressor * const restrict c,
+                       const uint8_t * const restrict block_begin,
+                       const uint32_t block_size,
+                       const struct lzx_lru_queue initial_queue,
+                       bool is_16_bit)
+{
+        struct lzx_optimum_node *cur_node = c->optimum_nodes;
+        struct lzx_optimum_node * const end_node = cur_node + block_size;
+        struct lz_match *cache_ptr = c->match_cache;
+        const uint8_t *in_next = block_begin;
+        const uint8_t * const block_end = block_begin + block_size;
+
+        /*
+         * Instead of storing the match offset LRU queues in the
+         * 'lzx_optimum_node' structures, we save memory (and cache lines) by
+         * storing them in a smaller array.  This works because the algorithm
+         * only requires a limited history of the adaptive state.  Once a given
+         * state is more than LZX_MAX_MATCH_LEN bytes behind the current node
+         * (more if gap match consideration is enabled; we just round up to 512
+         * so it's a power of 2), it is no longer needed.
+         *
+         * The QUEUE() macro finds the queue for the given node.  This macro has
+         * been optimized by taking advantage of 'struct lzx_lru_queue' and
+         * 'struct lzx_optimum_node' both being 8 bytes in size and alignment.
+         */
+        struct lzx_lru_queue queues[512];
+        STATIC_ASSERT(ARRAY_LEN(queues) >= LZX_MAX_MATCH_LEN + 1);
+        STATIC_ASSERT(sizeof(c->optimum_nodes[0]) == sizeof(queues[0]));
+#define QUEUE(node) \
+        (*(struct lzx_lru_queue *)((char *)queues + \
+                        ((uintptr_t)(node) % (ARRAY_LEN(queues) * sizeof(queues[0])))))
+        /*(queues[(uintptr_t)(node) / sizeof(*(node)) % ARRAY_LEN(queues)])*/
+
+#if CONSIDER_GAP_MATCHES
+        uint32_t matches_before_gap[ARRAY_LEN(queues)];
+#define MATCH_BEFORE_GAP(node) \
+        (matches_before_gap[(uintptr_t)(node) / sizeof(*(node)) % \
+                            ARRAY_LEN(matches_before_gap)])
+#endif
+
+        /*
+         * Initially, the cost to reach each node is "infinity".
+         *
+         * The first node actually should have cost 0, but "infinity"
+         * (0xFFFFFFFF) works just as well because it immediately overflows.
+         *
+         * The following statement also intentionally sets the 'item' of the
+         * first node, which would otherwise have no meaning, to 0xFFFFFFFF for
+         * use as a sentinel.  See lzx_walk_item_list().
+         */
+        memset(c->optimum_nodes, 0xFF,
+               (block_size + 1) * sizeof(c->optimum_nodes[0]));
+
+        /* Initialize the recent offsets queue for the first node. */
+        QUEUE(cur_node) = initial_queue;
+
+        do { /* For each node in the block in position order... */
+
+                unsigned num_matches;
+                unsigned literal;
+                uint32_t cost;
+
+                /*
+                 * A selection of matches for the block was already saved in
+                 * memory so that we don't have to run the uncompressed data
+                 * through the matchfinder on every optimization pass.  However,
+                 * we still search for repeat offset matches during each
+                 * optimization pass because we cannot predict the state of the
+                 * recent offsets queue.  But as a heuristic, we don't bother
+                 * searching for repeat offset matches if the general-purpose
+                 * matchfinder failed to find any matches.
+                 *
+                 * Note that a match of length n at some offset implies there is
+                 * also a match of length l for LZX_MIN_MATCH_LEN <= l <= n at
+                 * that same offset.  In other words, we don't necessarily need
+                 * to use the full length of a match.  The key heuristic that
+                 * saves a significicant amount of time is that for each
+                 * distinct length, we only consider the smallest offset for
+                 * which that length is available.  This heuristic also applies
+                 * to repeat offsets, which we order specially: R0 < R1 < R2 <
+                 * any explicit offset.  Of course, this heuristic may be
+                 * produce suboptimal results because offset slots in LZX are
+                 * subject to entropy encoding, but in practice this is a useful
+                 * heuristic.
+                 */
+
+                num_matches = cache_ptr->length;
+                cache_ptr++;
+
+                if (num_matches) {
+                        struct lz_match *end_matches = cache_ptr + num_matches;
+                        unsigned next_len = LZX_MIN_MATCH_LEN;
+                        unsigned max_len =
+                            min_uint(block_end - in_next, LZX_MAX_MATCH_LEN);
+                        const uint8_t *matchptr;
+
+                        /* Consider rep0 matches. */
+                        matchptr = in_next - lzx_lru_queue_R0(QUEUE(cur_node));
+                        if (load_u16_unaligned(matchptr) != load_u16_unaligned(in_next))
+                                goto rep0_done;
+                        STATIC_ASSERT_STMT(LZX_MIN_MATCH_LEN == 2);
+                        do {
+                                uint32_t cost = cur_node->cost +
+                                           c->costs.match_cost[0][
+                                                        next_len - LZX_MIN_MATCH_LEN];
+                                if (cost <= (cur_node + next_len)->cost) {
+                                        (cur_node + next_len)->cost = cost;
+                                        (cur_node + next_len)->item =
+                                                (0 << OPTIMUM_OFFSET_SHIFT) | next_len;
+                                }
+                                if (unlikely(++next_len > max_len)) {
+                                        cache_ptr = end_matches;
+                                        goto done_matches;
+                                }
+                        } while (in_next[next_len - 1] == matchptr[next_len - 1]);
+
+                rep0_done:
+
+                        /* Consider rep1 matches. */
+                        matchptr = in_next - lzx_lru_queue_R1(QUEUE(cur_node));
+                        if (load_u16_unaligned(matchptr) != load_u16_unaligned(in_next))
+                                goto rep1_done;
+                        if (matchptr[next_len - 1] != in_next[next_len - 1])
+                                goto rep1_done;
+                        for (unsigned len = 2; len < next_len - 1; len++)
+                                if (matchptr[len] != in_next[len])
+                                        goto rep1_done;
+                        do {
+                                uint32_t cost = cur_node->cost +
+                                           c->costs.match_cost[1][
+                                                        next_len - LZX_MIN_MATCH_LEN];
+                                if (cost <= (cur_node + next_len)->cost) {
+                                        (cur_node + next_len)->cost = cost;
+                                        (cur_node + next_len)->item =
+                                                (1 << OPTIMUM_OFFSET_SHIFT) | next_len;
+                                }
+                                if (unlikely(++next_len > max_len)) {
+                                        cache_ptr = end_matches;
+                                        goto done_matches;
+                                }
+                        } while (in_next[next_len - 1] == matchptr[next_len - 1]);
+
+                rep1_done:
+
+                        /* Consider rep2 matches. */
+                        matchptr = in_next - lzx_lru_queue_R2(QUEUE(cur_node));
+                        if (load_u16_unaligned(matchptr) != load_u16_unaligned(in_next))
+                                goto rep2_done;
+                        if (matchptr[next_len - 1] != in_next[next_len - 1])
+                                goto rep2_done;
+                        for (unsigned len = 2; len < next_len - 1; len++)
+                                if (matchptr[len] != in_next[len])
+                                        goto rep2_done;
+                        do {
+                                uint32_t cost = cur_node->cost +
+                                           c->costs.match_cost[2][
+                                                        next_len - LZX_MIN_MATCH_LEN];
+                                if (cost <= (cur_node + next_len)->cost) {
+                                        (cur_node + next_len)->cost = cost;
+                                        (cur_node + next_len)->item =
+                                                (2 << OPTIMUM_OFFSET_SHIFT) | next_len;
+                                }
+                                if (unlikely(++next_len > max_len)) {
+                                        cache_ptr = end_matches;
+                                        goto done_matches;
+                                }
+                        } while (in_next[next_len - 1] == matchptr[next_len - 1]);
+
+                rep2_done:
+
+                        while (next_len > cache_ptr->length)
+                                if (++cache_ptr == end_matches)
+                                        goto done_matches;
+
+                        /* Consider explicit offset matches. */
+                        for (;;) {
+                                uint32_t offset = cache_ptr->offset;
+                                uint32_t adjusted_offset = offset + LZX_OFFSET_ADJUSTMENT;
+                                unsigned offset_slot = lzx_get_offset_slot(c, adjusted_offset, is_16_bit);
+                                uint32_t base_cost = cur_node->cost;
+                                uint32_t cost;
+
+                        #if CONSIDER_ALIGNED_COSTS
+                                if (offset >= LZX_MIN_ALIGNED_OFFSET)
+                                        base_cost += c->costs.aligned[adjusted_offset &
+                                                                      LZX_ALIGNED_OFFSET_BITMASK];
+                        #endif
+                                do {
+                                        cost = base_cost +
+                                               c->costs.match_cost[offset_slot][
+                                                                next_len - LZX_MIN_MATCH_LEN];
+                                        if (cost < (cur_node + next_len)->cost) {
+                                                (cur_node + next_len)->cost = cost;
+                                                (cur_node + next_len)->item =
+                                                        (adjusted_offset << OPTIMUM_OFFSET_SHIFT) | next_len;
+                                        }
+                                } while (++next_len <= cache_ptr->length);
+
+                                if (++cache_ptr == end_matches) {
+                                #if CONSIDER_GAP_MATCHES
+                                        /* Also consider the longest explicit
+                                         * offset match as a "gap match": match
+                                         * + lit + rep0. */
+                                        int32_t remaining = (block_end - in_next) - (int32_t)next_len;
+                                        if (likely(remaining >= 2)) {
+                                                const uint8_t *strptr = in_next + next_len;
+                                                const uint8_t *matchptr = strptr - offset;
+                                                if (load_u16_unaligned(strptr) == load_u16_unaligned(matchptr)) {
+                                                        STATIC_ASSERT(ARRAY_LEN(queues) - LZX_MAX_MATCH_LEN - 2 >= 250);
+                                                        STATIC_ASSERT(ARRAY_LEN(queues) == ARRAY_LEN(matches_before_gap));
+                                                        unsigned limit = min_uint(remaining,
+                                                                             min_uint(ARRAY_LEN(queues) - LZX_MAX_MATCH_LEN - 2,
+                                                                                 LZX_MAX_MATCH_LEN));
+                                                        unsigned rep0_len = lz_extend(strptr, matchptr, 2, limit);
+                                                        uint8_t lit = strptr[-1];
+                                                        unsigned total_len = next_len + rep0_len;
+                                                        cost += c->costs.main[lit] +
+                                                                c->costs.match_cost[0][rep0_len - LZX_MIN_MATCH_LEN];
+                                                        if (cost < (cur_node + total_len)->cost) {
+                                                                (cur_node + total_len)->cost = cost;
+                                                                (cur_node + total_len)->item =
+                                                                        OPTIMUM_GAP_MATCH |
+                                                                        ((uint32_t)lit << OPTIMUM_OFFSET_SHIFT) |
+                                                                        rep0_len;
+                                                                MATCH_BEFORE_GAP(cur_node + total_len) =
+                                                                        (adjusted_offset << OPTIMUM_OFFSET_SHIFT) |
+                                                                        (next_len - 1);
+                                                        }
+                                                }
+                                        }
+                                #endif /* CONSIDER_GAP_MATCHES */
+                                        break;
+                                }
+                        }
+                }
+
+        done_matches:
+
+                /* Consider coding a literal.
+
+                 * To avoid an extra branch, actually checking the preferability
+                 * of coding the literal is integrated into the queue update
+                 * code below. */
+                literal = *in_next++;
+                cost = cur_node->cost + c->costs.main[literal];
+
+                /* Advance to the next position. */
+                cur_node++;
+
+                /* The lowest-cost path to the current position is now known.
+                 * Finalize the recent offsets queue that results from taking
+                 * this lowest-cost path. */
+
+                if (cost <= cur_node->cost) {
+                        /* Literal: queue remains unchanged. */
+                        cur_node->cost = cost;
+                        cur_node->item = (uint32_t)literal << OPTIMUM_OFFSET_SHIFT;
+                        QUEUE(cur_node) = QUEUE(cur_node - 1);
+                } else {
+                        /* Match: queue update is needed. */
+                        unsigned len = cur_node->item & OPTIMUM_LEN_MASK;
+                #if CONSIDER_GAP_MATCHES
+                        int32_t adjusted_offset = (int32_t)cur_node->item >> OPTIMUM_OFFSET_SHIFT;
+                        STATIC_ASSERT(OPTIMUM_GAP_MATCH == 0x80000000); /* assuming sign extension */
+                #else
+                        uint32_t adjusted_offset = cur_node->item >> OPTIMUM_OFFSET_SHIFT;
+                #endif
+
+                        if (adjusted_offset >= LZX_NUM_RECENT_OFFSETS) {
+                                /* Explicit offset match: insert offset at front. */
+                                QUEUE(cur_node) =
+                                        lzx_lru_queue_push(QUEUE(cur_node - len),
+                                                           adjusted_offset - LZX_OFFSET_ADJUSTMENT);
+                        }
+                #if CONSIDER_GAP_MATCHES
+                        else if (adjusted_offset < 0) {
+                                /* "Gap match": Explicit offset match, then a
+                                 * literal, then rep0 match.  Save the explicit
+                                 * offset match information in the cost field of
+                                 * the previous node, which isn't needed
+                                 * anymore.  Then insert the offset at the front
+                                 * of the queue. */
+                                uint32_t match_before_gap = MATCH_BEFORE_GAP(cur_node);
+                                (cur_node - 1)->cost = match_before_gap;
+                                QUEUE(cur_node) =
+                                        lzx_lru_queue_push(QUEUE(cur_node - len - 1 -
+                                                                 (match_before_gap & OPTIMUM_LEN_MASK)),
+                                                           (match_before_gap >> OPTIMUM_OFFSET_SHIFT) -
+                                                           LZX_OFFSET_ADJUSTMENT);
+                        }
+                #endif
+                        else {
+                                /* Repeat offset match: swap offset to front. */
+                                QUEUE(cur_node) =
+                                        lzx_lru_queue_swap(QUEUE(cur_node - len),
+                                                           adjusted_offset);
+                        }
+                }
+        } while (cur_node != end_node);
+
+        /* Return the recent offsets queue at the end of the path. */
+        return QUEUE(cur_node);
+}
+
+/*
+ * Given the costs for the main and length codewords (c->costs.main and
+ * c->costs.len), initialize the match cost array (c->costs.match_cost) which
+ * directly provides the cost of every possible (length, offset slot) pair.
+ */
+static void
+lzx_compute_match_costs(struct liblzx_compressor *c)
+{
+        unsigned num_offset_slots = (c->num_main_syms - LZX_NUM_CHARS) /
+                                        LZX_NUM_LEN_HEADERS;
+        struct lzx_costs *costs = &c->costs;
+        unsigned main_symbol = LZX_NUM_CHARS;
+
+        for (unsigned offset_slot = 0; offset_slot < num_offset_slots;
+             offset_slot++)
+        {
+                uint32_t extra_cost = lzx_extra_offset_bits[offset_slot] * BIT_COST;
+                unsigned i;
+
+        #if CONSIDER_ALIGNED_COSTS
+                if (offset_slot >= LZX_MIN_ALIGNED_OFFSET_SLOT)
+                        extra_cost -= LZX_NUM_ALIGNED_OFFSET_BITS * BIT_COST;
+        #endif
+
+                for (i = 0; i < LZX_NUM_PRIMARY_LENS; i++) {
+                        costs->match_cost[offset_slot][i] =
+                                costs->main[main_symbol++] + extra_cost;
+                }
+
+                extra_cost += costs->main[main_symbol++];
+
+                for (; i < LZX_NUM_LENS; i++) {
+                        costs->match_cost[offset_slot][i] =
+                                costs->len[i - LZX_NUM_PRIMARY_LENS] +
+                                extra_cost;
+                }
+        }
+}
+
+typedef struct fixed32frac_s {
+        uint32_t value;
+} fixed32frac;
+
+typedef struct fixed32_s {
+        uint64_t value;
+} fixed32;
+
+/*
+ * Fast approximation for log2f(x).  This is not as accurate as the standard C
+ * version.  It does not need to be perfectly accurate because it is only used
+ * for estimating symbol costs, which is very approximate anyway.
+ */
+struct log2_fixed_table_pair
+{
+        uint32_t multiplier;
+        uint32_t log_add;
+};
+
+static const struct log2_fixed_table_pair log2_fixed_table_0[255] = {
+        { 0xff01fc08u, 0x16f2dcfu },  { 0xfe05ee36u, 0x2dceffeu },
+        { 0xfd0bd0beu, 0x4494959u },  { 0xfc139debu, 0x5b43ca6u },
+        { 0xfb1d5020u, 0x71dcca2u },  { 0xfa28e1d4u, 0x885fc02u },
+        { 0xf9364d94u, 0x9eccd73u },  { 0xf8458e02u, 0xb52439au },
+        { 0xf7569dd6u, 0xcb66115u },  { 0xf66977dau, 0xe19287au },
+        { 0xf57e16edu, 0xf7a9c58u },  { 0xf4947602u, 0x10dabf37u },
+        { 0xf3ac901fu, 0x12399395u }, { 0xf2c6605bu, 0x13971beeu },
+        { 0xf1e1e1e2u, 0x14f35ab3u }, { 0xf0ff0ff1u, 0x164e524fu },
+        { 0xf01de5d7u, 0x17a80527u }, { 0xef3e5ef4u, 0x19007598u },
+        { 0xee6076bau, 0x1a57a5f9u }, { 0xed8428aau, 0x1bad989cu },
+        { 0xeca97059u, 0x1d024fc9u }, { 0xebd04968u, 0x1e55cdc7u },
+        { 0xeaf8af8bu, 0x1fa814d2u }, { 0xea229e85u, 0x20f92722u },
+        { 0xe94e1228u, 0x224906e7u }, { 0xe87b0655u, 0x2397b64fu },
+        { 0xe7a976fdu, 0x24e5377du }, { 0xe6d9601du, 0x26318c93u },
+        { 0xe60abdc3u, 0x277cb7acu }, { 0xe53d8c0bu, 0x28c6bad9u },
+        { 0xe471c71du, 0x2a0f982cu }, { 0xe3a76b2fu, 0x2b5751aeu },
+        { 0xe2de7486u, 0x2c9de963u }, { 0xe216df74u, 0x2de36147u },
+        { 0xe150a854u, 0x2f27bb59u }, { 0xe08bcb94u, 0x306af989u },
+        { 0xdfc845a9u, 0x31ad1dc8u }, { 0xdf061318u, 0x32ee29feu },
+        { 0xde45306fu, 0x342e2014u }, { 0xdd859a4au, 0x356d01e8u },
+        { 0xdcc74d51u, 0x36aad154u }, { 0xdc0a4635u, 0x37e79032u },
+        { 0xdb4e81b5u, 0x39234051u }, { 0xda93fc99u, 0x3a5de380u },
+        { 0xd9dab3b6u, 0x3b977b86u }, { 0xd922a3e9u, 0x3cd00a2au },
+        { 0xd86bca1bu, 0x3e07912bu }, { 0xd7b62341u, 0x3f3e1241u },
+        { 0xd701ac57u, 0x40738f27u }, { 0xd64e6266u, 0x41a8098eu },
+        { 0xd59c427fu, 0x42db8323u }, { 0xd4eb49bcu, 0x440dfd94u },
+        { 0xd43b7544u, 0x453f7a82u }, { 0xd38cc244u, 0x466ffb93u },
+        { 0xd2df2df3u, 0x479f8265u }, { 0xd232b593u, 0x48ce108eu },
+        { 0xd187566cu, 0x49fba7a9u }, { 0xd0dd0dd1u, 0x4b284946u },
+        { 0xd033d91du, 0x4c53f6f4u }, { 0xcf8bb5b4u, 0x4d7eb23bu },
+        { 0xcee4a102u, 0x4ea87ca4u }, { 0xce3e987au, 0x4fd157b4u },
+        { 0xcd99999au, 0x50f944e7u }, { 0xccf5a1e5u, 0x522045bcu },
+        { 0xcc52aee8u, 0x53465baau }, { 0xcbb0be38u, 0x546b8825u },
+        { 0xcb0fcd6fu, 0x558fcca0u }, { 0xca6fda31u, 0x56b32a89u },
+        { 0xc9d0e229u, 0x57d5a34au }, { 0xc932e309u, 0x58f7384au },
+        { 0xc895da89u, 0x5a17eaf0u }, { 0xc7f9c66bu, 0x5b37bc99u },
+        { 0xc75ea476u, 0x5c56aea2u }, { 0xc6c47277u, 0x5d74c26au },
+        { 0xc62b2e44u, 0x5e91f945u }, { 0xc592d5b8u, 0x5fae5488u },
+        { 0xc4fb66b5u, 0x60c9d584u }, { 0xc464df24u, 0x61e47d87u },
+        { 0xc3cf3cf4u, 0x62fe4dddu }, { 0xc33a7e1au, 0x641747cdu },
+        { 0xc2a6a091u, 0x652f6c9eu }, { 0xc213a25cu, 0x6646bd8fu },
+        { 0xc1818182u, 0x675d3be2u }, { 0xc0f03c0fu, 0x6872e8d5u },
+        { 0xc05fd018u, 0x6987c59fu }, { 0xbfd03bb5u, 0x6a9bd379u },
+        { 0xbf417d06u, 0x6baf1395u }, { 0xbeb3922eu, 0x6cc18729u },
+        { 0xbe267957u, 0x6dd32f61u }, { 0xbd9a30b1u, 0x6ee40d69u },
+        { 0xbd0eb670u, 0x6ff4226du }, { 0xbc8408cdu, 0x71036f95u },
+        { 0xbbfa2609u, 0x7211f601u }, { 0xbb710c66u, 0x731fb6d9u },
+        { 0xbae8ba2fu, 0x742cb339u }, { 0xba612db0u, 0x7538ec41u },
+        { 0xb9da653eu, 0x7644630au }, { 0xb9545f30u, 0x774f18adu },
+        { 0xb8cf19e3u, 0x78590e41u }, { 0xb84a93b8u, 0x796244d9u },
+        { 0xb7c6cb15u, 0x7a6abd86u }, { 0xb743be65u, 0x7b727958u },
+        { 0xb6c16c17u, 0x7c79795bu }, { 0xb63fd29du, 0x7d7fbe9eu },
+        { 0xb5bef071u, 0x7e854a23u }, { 0xb53ec40eu, 0x7f8a1cf4u },
+        { 0xb4bf4bf5u, 0x808e3815u }, { 0xb44086aau, 0x81919c88u },
+        { 0xb3c272b6u, 0x82944b4cu }, { 0xb3450ea6u, 0x83964560u },
+        { 0xb2c8590bu, 0x84978bc0u }, { 0xb24c507au, 0x85981f63u },
+        { 0xb1d0f38cu, 0x86980143u }, { 0xb15640ddu, 0x87973255u },
+        { 0xb0dc370eu, 0x8895b38du }, { 0xb062d4c3u, 0x899385ddu },
+        { 0xafea18a4u, 0x8a90aa35u }, { 0xaf72015du, 0x8b8d2181u },
+        { 0xaefa8d9eu, 0x8c88ecadu }, { 0xae83bc18u, 0x8d840ca6u },
+        { 0xae0d8b83u, 0x8e7e8252u }, { 0xad97fa99u, 0x8f784e96u },
+        { 0xad230816u, 0x90717259u }, { 0xacaeb2bbu, 0x9169ee7eu },
+        { 0xac3af94cu, 0x9261c3e6u }, { 0xabc7da92u, 0x9358f36cu },
+        { 0xab555555u, 0x944f7df3u }, { 0xaae36865u, 0x95456453u },
+        { 0xaa721292u, 0x963aa768u }, { 0xaa0152b0u, 0x972f4809u },
+        { 0xa9912796u, 0x9823470fu }, { 0xa9219020u, 0x9916a54au },
+        { 0xa8b28b29u, 0x9a096393u }, { 0xa8441792u, 0x9afb82bau },
+        { 0xa7d6343fu, 0x9bed038du }, { 0xa768e015u, 0x9cdde6ddu },
+        { 0xa6fc19fdu, 0x9dce2d77u }, { 0xa68fe0e4u, 0x9ebdd823u },
+        { 0xa62433b8u, 0x9face7adu }, { 0xa5b91169u, 0xa09b5cdfu },
+        { 0xa54e78edu, 0xa189387du }, { 0xa4e46939u, 0xa2767b4fu },
+        { 0xa47ae148u, 0xa3632616u }, { 0xa411e014u, 0xa44f3999u },
+        { 0xa3a9649eu, 0xa53ab692u }, { 0xa3416de5u, 0xa6259dc7u },
+        { 0xa2d9faeeu, 0xa70feff2u }, { 0xa2730abfu, 0xa7f9add0u },
+        { 0xa20c9c60u, 0xa8e2d81du }, { 0xa1a6aedcu, 0xa9cb6f95u },
+        { 0xa1414141u, 0xaab374eeu }, { 0xa0dc529fu, 0xab9ae8dfu },
+        { 0xa077e207u, 0xac81cc1fu }, { 0xa013ee8fu, 0xad681f62u },
+        { 0x9fb0774du, 0xae4de35au }, { 0x9f4d7b5au, 0xaf3318bau },
+        { 0x9eeaf9d1u, 0xb017c033u }, { 0x9e88f1d0u, 0xb0fbda74u },
+        { 0x9e276276u, 0xb1df682bu }, { 0x9dc64ae5u, 0xb2c26a06u },
+        { 0x9d65aa42u, 0xb3a4e0acu }, { 0x9d057fb2u, 0xb486cccbu },
+        { 0x9ca5ca5du, 0xb5682f0cu }, { 0x9c46896du, 0xb6490816u },
+        { 0x9be7bc0eu, 0xb7295893u }, { 0x9b896170u, 0xb8092121u },
+        { 0x9b2b78c1u, 0xb8e8626cu }, { 0x9ace0134u, 0xb9c71d13u },
+        { 0x9a70f9fdu, 0xbaa551b9u }, { 0x9a146253u, 0xbb8300fdu },
+        { 0x99b8396cu, 0xbc602b82u }, { 0x995c7e82u, 0xbd3cd1e6u },
+        { 0x990130d1u, 0xbe18f4c6u }, { 0x98a64f97u, 0xbef494bdu },
+        { 0x984bda13u, 0xbfcfb267u }, { 0x97f1cf85u, 0xc0aa4e5fu },
+        { 0x97982f30u, 0xc184693fu }, { 0x973ef859u, 0xc25e039eu },
+        { 0x96e62a46u, 0xc3371e12u }, { 0x968dc43fu, 0xc40fb932u },
+        { 0x9635c58du, 0xc4e7d594u }, { 0x95de2d7cu, 0xc5bf73cau },
+        { 0x9586fb58u, 0xc696946au }, { 0x95302e70u, 0xc76d3803u },
+        { 0x94d9c615u, 0xc8435f25u }, { 0x9483c197u, 0xc9190a64u },
+        { 0x942e204au, 0xc9ee3a4eu }, { 0x93d8e182u, 0xcac2ef71u },
+        { 0x93840497u, 0xcb972a58u }, { 0x932f88e0u, 0xcc6aeb90u },
+        { 0x92db6db7u, 0xcd3e33a3u }, { 0x9287b275u, 0xce110320u },
+        { 0x92345678u, 0xcee35a8du }, { 0x91e1591eu, 0xcfb53a70u },
+        { 0x918eb9c5u, 0xd086a355u }, { 0x913c77ceu, 0xd15795c2u },
+        { 0x90ea929bu, 0xd228123cu }, { 0x90990990u, 0xd2f81946u },
+        { 0x9047dc12u, 0xd3c7ab65u }, { 0x8ff70986u, 0xd496c91du },
+        { 0x8fa69154u, 0xd56572f1u }, { 0x8f5672e4u, 0xd633a963u },
+        { 0x8f06ada2u, 0xd7016cf0u }, { 0x8eb740f9u, 0xd7cebe18u },
+        { 0x8e682c54u, 0xd89b9d5fu }, { 0x8e196f23u, 0xd9680b3du },
+        { 0x8dcb08d4u, 0xda340833u }, { 0x8d7cf8d8u, 0xdaff94bcu },
+        { 0x8d2f3ea0u, 0xdbcab157u }, { 0x8ce1d9a0u, 0xdc955e7au },
+        { 0x8c94c94cu, 0xdd5f9ca0u }, { 0x8c480d19u, 0xde296c45u },
+        { 0x8bfba47eu, 0xdef2cddeu }, { 0x8baf8ef2u, 0xdfbbc1e5u },
+        { 0x8b63cbeeu, 0xe08448d1u }, { 0x8b185aedu, 0xe14c6316u },
+        { 0x8acd3b69u, 0xe214112du }, { 0x8a826cdeu, 0xe2db5389u },
+        { 0x8a37eecau, 0xe3a22a9fu }, { 0x89edc0acu, 0xe46896deu },
+        { 0x89a3e202u, 0xe52e98bfu }, { 0x895a524eu, 0xe5f430b0u },
+        { 0x89111111u, 0xe6b95f22u }, { 0x88c81dceu, 0xe77e2486u },
+        { 0x887f7808u, 0xe842814eu }, { 0x88371f45u, 0xe90675e5u },
+        { 0x87ef130au, 0xe9ca02bcu }, { 0x87a752dfu, 0xea8d283du },
+        { 0x875fde4au, 0xeb4fe6dau }, { 0x8718b4d4u, 0xec123effu },
+        { 0x86d1d608u, 0xecd43114u }, { 0x868b4170u, 0xed95bd86u },
+        { 0x8644f698u, 0xee56e4beu }, { 0x85fef50du, 0xef17a726u },
+        { 0x85b93c5bu, 0xefd8052bu }, { 0x8573cc12u, 0xf097ff30u },
+        { 0x852ea3c2u, 0xf157959cu }, { 0x84e9c2f9u, 0xf216c8ddu },
+        { 0x84a5294au, 0xf2d59955u }, { 0x8460d647u, 0xf3940768u },
+        { 0x841cc982u, 0xf4521381u }, { 0x83d90290u, 0xf50fbdffu },
+        { 0x83958106u, 0xf5cd0747u }, { 0x83524478u, 0xf689efc0u },
+        { 0x830f4c7eu, 0xf74677c9u }, { 0x82cc98afu, 0xf8029fc5u },
+        { 0x828a28a2u, 0xf8be6819u }, { 0x8247fbf2u, 0xf979d120u },
+        { 0x82061236u, 0xfa34db42u }, { 0x81c46b0bu, 0xfaef86d9u },
+        { 0x8183060cu, 0xfba9d445u }, { 0x8141e2d4u, 0xfc63c3e8u },
+        { 0x81010101u, 0xfd1d561du }, { 0x80c06030u, 0xfdd68b44u },
+        { 0x80800000u, 0xfe8f63b9u },
+};
+
+static const uint32_t log2_fixed_table_1[255] = {
+        0x17152u,   0x2e2a3u,        0x453f2u,   0x5c540u,        0x7368du,   0x8a7d8u,
+        0xa1921u,   0xb8a69u,        0xcfbb0u,   0xe6cf6u,        0xfde39u,   0x114f7cu,
+        0x12c0bdu,  0x1431fcu,        0x15a33au,  0x171477u,        0x1885b2u,  0x19f6ecu,
+        0x1b6824u,  0x1cd95bu,        0x1e4a91u,  0x1fbbc5u,        0x212cf7u,  0x229e28u,
+        0x240f58u,  0x258086u,        0x26f1b3u,  0x2862deu,        0x29d408u,  0x2b4530u,
+        0x2cb657u,  0x2e277du,        0x2f98a1u,  0x3109c4u,        0x327ae5u,  0x33ec05u,
+        0x355d23u,  0x36ce40u,        0x383f5bu,  0x39b077u,        0x3b218fu,  0x3c92a6u,
+        0x3e03bcu,  0x3f74d0u,        0x40e5e3u,  0x4256f4u,        0x43c804u,  0x453912u,
+        0x46aa1fu,  0x481b2bu,        0x498c36u,  0x4afd3fu,        0x4c6e46u,  0x4ddf4cu,
+        0x4f5050u,  0x50c153u,        0x523254u,  0x53a355u,        0x551454u,  0x568551u,
+        0x57f64cu,  0x596747u,        0x5ad83fu,  0x5c4938u,        0x5dba2eu,  0x5f2b22u,
+        0x609c15u,  0x620d06u,        0x637df8u,  0x64eee6u,        0x665fd3u,  0x67d0bfu,
+        0x6941abu,  0x6ab293u,        0x6c237bu,  0x6d9461u,        0x6f0546u,  0x707629u,
+        0x71e70bu,  0x7357ecu,        0x74c8cbu,  0x7639a8u,        0x77aa84u,  0x791b5fu,
+        0x7a8c38u,  0x7bfd10u,        0x7d6de7u,  0x7edebbu,        0x804f8eu,  0x81c061u,
+        0x833131u,  0x84a202u,        0x8612cfu,  0x87839au,        0x88f466u,  0x8a652fu,
+        0x8bd5f8u,  0x8d46beu,        0x8eb784u,  0x902847u,        0x91990au,  0x9309cau,
+        0x947a89u,  0x95eb47u,        0x975c03u,  0x98ccbfu,        0x9a3d79u,  0x9bae31u,
+        0x9d1ee8u,  0x9e8f9cu,        0xa00051u,  0xa17103u,        0xa2e1b4u,  0xa45263u,
+        0xa5c312u,  0xa733bfu,        0xa8a469u,  0xaa1513u,        0xab85bcu,  0xacf662u,
+        0xae6708u,  0xafd7adu,        0xb1484eu,  0xb2b8f0u,        0xb42990u,  0xb59a2du,
+        0xb70acbu,  0xb87b67u,        0xb9ec01u,  0xbb5c98u,        0xbccd30u,  0xbe3dc6u,
+        0xbfae5au,  0xc11eedu,        0xc28f7fu,  0xc4000eu,        0xc5709cu,  0xc6e12au,
+        0xc851b5u,  0xc9c240u,        0xcb32c9u,  0xcca350u,        0xce13d6u,  0xcf845bu,
+        0xd0f4deu,  0xd2655fu,        0xd3d5dfu,  0xd5465eu,        0xd6b6dbu,  0xd82757u,
+        0xd997d2u,  0xdb084au,        0xdc78c2u,  0xdde938u,        0xdf59acu,  0xe0ca1fu,
+        0xe23a91u,  0xe3ab01u,        0xe51b70u,  0xe68bdfu,        0xe7fc4au,  0xe96cb5u,
+        0xeadd1du,  0xec4d85u,        0xedbdecu,  0xef2e51u,        0xf09eb4u,  0xf20f17u,
+        0xf37f77u,  0xf4efd6u,        0xf66033u,  0xf7d090u,        0xf940eau,  0xfab144u,
+        0xfc219cu,  0xfd91f2u,        0xff0248u,  0x100729bu, 0x101e2eeu, 0x103533eu,
+        0x104c38eu, 0x10633dbu, 0x107a428u, 0x1091472u, 0x10a84bdu, 0x10bf504u,
+        0x10d654bu, 0x10ed591u, 0x11045d4u, 0x111b617u, 0x1132658u, 0x1149697u,
+        0x11606d6u, 0x1177713u, 0x118e74du, 0x11a5787u, 0x11bc7c0u, 0x11d37f7u,
+        0x11ea82du, 0x1201860u, 0x1218892u, 0x122f8c4u, 0x12468f4u, 0x125d922u,
+        0x127494fu, 0x128b97bu, 0x12a29a5u, 0x12b99ceu, 0x12d09f5u, 0x12e7a1bu,
+        0x12fea3fu, 0x1315a62u, 0x132ca83u, 0x1343aa3u, 0x135aac1u, 0x1371adeu,
+        0x1388afau, 0x139fb15u, 0x13b6b2eu, 0x13cdb45u, 0x13e4b5bu, 0x13fbb6fu,
+        0x1412b83u, 0x1429b94u, 0x1440ba4u, 0x1457bb4u, 0x146ebc1u, 0x1485bccu,
+        0x149cbd8u, 0x14b3be0u, 0x14cabe8u, 0x14e1beeu, 0x14f8bf2u, 0x150fbf6u,
+        0x1526bf9u, 0x153dbf9u, 0x1554bf8u, 0x156bbf5u, 0x1582bf2u, 0x1599becu,
+        0x15b0be6u, 0x15c7bdeu, 0x15debd3u, 0x15f5bc9u, 0x160cbbdu, 0x1623bafu,
+        0x163ab9fu, 0x1651b8eu, 0x1668b7du, 0x167fb69u, 0x1696b54u, 0x16adb3eu,
+        0x16c4b26u, 0x16dbb0du, 0x16f2af3u,
+};
+
+static int32_t
+log2_fixed_fast_normalized(const fixed32 *value, int multiplier_bits)
+{
+        int32_t base_pos = 0;
+        uint32_t mantissa = 0;
+        uint32_t mantissa_log2 = 0;
+        uint8_t mantissa_byte = 0;
+        int64_t final_log = 0;
+
+        if (value->value == 0)
+                return 0;
+
+        base_pos = (int)bsr64(value->value) - 32;
+        if (base_pos > 0) {
+                mantissa = (value->value >> base_pos) & 0xffffffffu;
+        } else {
+                mantissa = (value->value << -base_pos) & 0xffffffffu;
+        }
+
+        /* Get the first byte */
+        mantissa_byte = (mantissa >> 24) & 0xffu;
+
+        if (mantissa_byte != 0) {
+                const struct log2_fixed_table_pair *pair = log2_fixed_table_0 + (mantissa_byte - 1);
+                mantissa = (uint32_t)((mantissa * (uint64_t)pair->multiplier) >> 32) + pair->multiplier;
+
+                mantissa_log2 += pair->log_add;
+        }
+
+        mantissa_byte = (mantissa >> 16) & 0xffu;
+
+        if (mantissa_byte != 0) {
+                mantissa_log2 += log2_fixed_table_1[mantissa_byte - 1];
+        }
+
+        final_log = (int64_t)mantissa_log2 + ((int64_t)base_pos << 32);
+        final_log /= ((int64_t)1 << (32 - multiplier_bits));
+
+        return (int32_t)final_log;
+}
+
+/*
+ * Return the estimated cost of a symbol which has been estimated to have the
+ * given probability.
+ */
+static uint32_t
+lzx_cost_for_probability(const fixed32* prob)
+{
+        /*
+         * The basic formula is:
+         *
+         *        entropy = -log2(probability)
+         *
+         * Use this to get the cost in fractional bits.  Then multiply by our
+         * scaling factor of BIT_COST and convert to an integer.
+         *
+         * In addition, the minimum cost is BIT_COST (one bit) because the
+         * entropy coding method will be Huffman codes.
+         *
+         * Careful: even though 'prob' should be <= 1.0, 'log2f_fast(prob)' may
+         * be positive due to inaccuracy in our log2 approximation.  Therefore,
+         * we cannot, in general, assume the computed cost is non-negative, and
+         * we should make sure negative costs get rounded up correctly.
+         */
+        int32_t cost = -log2_fixed_fast_normalized(prob, BIT_COST_BITS);
+        return max_u32(cost, BIT_COST);
+}
+
+/*
+ * Mapping: number of used literals => heuristic probability of a literal times
+ * 6870.  Generated by running this R command:
+ *
+ *        cat(paste(round(6870*2^-((304+(0:256))/64)), collapse=", "))
+ */
+static const uint8_t literal_scaled_probs[257] = {
+        255, 253, 250, 247, 244, 242, 239, 237, 234, 232, 229, 227, 224, 222,
+        219, 217, 215, 212, 210, 208, 206, 203, 201, 199, 197, 195, 193, 191,
+        189, 186, 184, 182, 181, 179, 177, 175, 173, 171, 169, 167, 166, 164,
+        162, 160, 159, 157, 155, 153, 152, 150, 149, 147, 145, 144, 142, 141,
+        139, 138, 136, 135, 133, 132, 130, 129, 128, 126, 125, 124, 122, 121,
+        120, 118, 117, 116, 115, 113, 112, 111, 110, 109, 107, 106, 105, 104,
+        103, 102, 101, 100, 98, 97, 96, 95, 94, 93, 92, 91, 90, 89, 88, 87, 86,
+        86, 85, 84, 83, 82, 81, 80, 79, 78, 78, 77, 76, 75, 74, 73, 73, 72, 71,
+        70, 70, 69, 68, 67, 67, 66, 65, 65, 64, 63, 62, 62, 61, 60, 60, 59, 59,
+        58, 57, 57, 56, 55, 55, 54, 54, 53, 53, 52, 51, 51, 50, 50, 49, 49, 48,
+        48, 47, 47, 46, 46, 45, 45, 44, 44, 43, 43, 42, 42, 41, 41, 40, 40, 40,
+        39, 39, 38, 38, 38, 37, 37, 36, 36, 36, 35, 35, 34, 34, 34, 33, 33, 33,
+        32, 32, 32, 31, 31, 31, 30, 30, 30, 29, 29, 29, 28, 28, 28, 27, 27, 27,
+        27, 26, 26, 26, 25, 25, 25, 25, 24, 24, 24, 24, 23, 23, 23, 23, 22, 22,
+        22, 22, 21, 21, 21, 21, 20, 20, 20, 20, 20, 19, 19, 19, 19, 19, 18, 18,
+        18, 18, 18, 17, 17, 17, 17, 17, 16, 16, 16, 16
+};
+
+/*
+ * Mapping: length symbol => default cost of that symbol.  This is derived from
+ * sample data but has been slightly edited to add more bias towards the
+ * shortest lengths, which are the most common.
+ */
+static const uint16_t lzx_default_len_costs[LZX_LENCODE_NUM_SYMBOLS] = {
+        300, 310, 320, 330, 360, 396, 399, 416, 451, 448, 463, 466, 505, 492,
+        503, 514, 547, 531, 566, 561, 589, 563, 592, 586, 623, 602, 639, 627,
+        659, 643, 657, 650, 685, 662, 661, 672, 685, 686, 696, 680, 657, 682,
+        666, 699, 674, 699, 679, 709, 688, 712, 692, 714, 694, 716, 698, 712,
+        706, 727, 714, 727, 713, 723, 712, 718, 719, 719, 720, 735, 725, 735,
+        728, 740, 727, 739, 727, 742, 716, 733, 733, 740, 738, 746, 737, 747,
+        738, 745, 736, 748, 742, 749, 745, 749, 743, 748, 741, 752, 745, 752,
+        747, 750, 747, 752, 748, 753, 750, 752, 753, 753, 749, 744, 752, 755,
+        753, 756, 745, 748, 746, 745, 723, 757, 755, 758, 755, 758, 752, 757,
+        754, 757, 755, 759, 755, 758, 753, 755, 755, 758, 757, 761, 755, 750,
+        758, 759, 759, 760, 758, 751, 757, 757, 759, 759, 758, 759, 758, 761,
+        750, 761, 758, 760, 759, 761, 758, 761, 760, 752, 759, 760, 759, 759,
+        757, 762, 760, 761, 761, 748, 761, 760, 762, 763, 752, 762, 762, 763,
+        762, 762, 763, 763, 762, 763, 762, 763, 762, 763, 763, 764, 763, 762,
+        763, 762, 762, 762, 764, 764, 763, 764, 763, 763, 763, 762, 763, 763,
+        762, 764, 764, 763, 762, 763, 763, 763, 763, 762, 764, 763, 762, 764,
+        764, 763, 763, 765, 764, 764, 762, 763, 764, 765, 763, 764, 763, 764,
+        762, 764, 764, 754, 763, 764, 763, 763, 762, 763, 584,
+};
+
+static void
+fixed_rcp_approx(fixed32frac *result, uint32_t i)
+{
+        result->value = ((uint64_t)0x100000000ull) / i;
+}
+
+static void
+fixed_set(fixed32 *result, uint32_t i)
+{
+        result->value = ((uint64_t)i << 32);
+}
+
+static void
+fixed_set_fraction(fixed32frac *result, uint32_t num, uint32_t denom)
+{
+        result->value = (((uint64_t)num) << 32) / denom;
+}
+
+static void
+fixed_mul_uint_frac(fixed32 *result, uint32_t a, const fixed32frac *b)
+{
+        result->value = ((uint64_t)a) * (uint64_t)b->value;
+}
+
+static void
+fixed_mul_uint_frac_to_frac(fixed32frac *result, uint32_t a, const fixed32frac *b)
+{
+        fixed32 fixed;
+        fixed_mul_uint_frac(&fixed, a, b);
+        result->value = (uint32_t)fixed.value;
+}
+
+static void
+fixed_add_frac(fixed32 *result, const fixed32 *a, const fixed32frac *b)
+{
+        result->value = a->value + b->value;
+}
+
+static void
+fixed_sub(fixed32 *result, const fixed32 *a, const fixed32 *b)
+{
+        result->value = a->value + b->value;
+}
+
+static void
+fixed_max_frac(fixed32 *result, const fixed32 *a, const fixed32frac *b)
+{
+        result->value = max_u64(a->value, b->value);
+}
+
+static void
+fixed_div_uint(fixed32 *result, const fixed32 *a, uint32_t b)
+{
+        result->value = a->value / b;
+}
+
+/* Set default costs to bootstrap the iterative optimization algorithm. */
+static void
+lzx_set_default_costs(struct liblzx_compressor *c)
+{
+        unsigned i;
+        uint32_t num_literals = 0;
+        uint32_t num_used_literals = 0;
+        fixed32frac inv_num_matches;
+        fixed32frac half_inv_num_items;
+        fixed32frac half_inv_6870;
+        fixed32 prob_match;
+        fixed32frac frac_15_100;
+        uint32_t match_cost;
+        fixed32frac half_base_literal_prob;
+        fixed32 temp_fixed;
+
+        fixed_rcp_approx(&inv_num_matches, c->freqs.main[LZX_NUM_CHARS]);
+        fixed_rcp_approx(&half_inv_6870, 6870 * 2);
+        fixed_set(&prob_match, 1);
+        fixed_set_fraction(&frac_15_100, 15, 100);
+
+        /* Some numbers here have been hardcoded to assume a bit cost of 64. */
+        STATIC_ASSERT_STMT(BIT_COST == 64);
+
+        /* Estimate the number of literals that will used.  'num_literals' is
+         * the total number, whereas 'num_used_literals' is the number of
+         * distinct symbols. */
+        for (i = 0; i < LZX_NUM_CHARS; i++) {
+                num_literals += c->freqs.main[i];
+                num_used_literals += (c->freqs.main[i] != 0);
+        }
+
+        /* Note: all match headers were tallied as symbol 'LZX_NUM_CHARS'.  We
+         * don't attempt to estimate which ones will be used. */
+
+        fixed_rcp_approx(&half_inv_num_items,
+                         (num_literals + c->freqs.main[LZX_NUM_CHARS]) * 2);
+        fixed_mul_uint_frac_to_frac(&half_base_literal_prob,
+                            literal_scaled_probs[num_used_literals],
+                            &half_inv_6870);
+
+        /* Literal costs.  We use two different methods to compute the
+         * probability of each literal and mix together their results. */
+        for (i = 0; i < LZX_NUM_CHARS; i++) {
+                uint32_t freq = c->freqs.main[i];
+                if (freq != 0) {
+                        fixed32 prob;
+                        fixed_mul_uint_frac(&prob, freq, &half_inv_num_items);
+                        fixed_add_frac(&prob, &prob, &half_base_literal_prob);
+
+                        c->costs.main[i] = lzx_cost_for_probability(&prob);
+                        fixed_sub(&prob_match, &prob_match, &prob);
+                } else {
+                        c->costs.main[i] = 11 * BIT_COST;
+                }
+        }
+
+        /* Match header costs.  We just assume that all match headers are
+         * equally probable, but we do take into account the relative cost of a
+         * match header vs. a literal depending on how common matches are
+         * expected to be vs. literals. */
+        fixed_max_frac(&prob_match, &prob_match, &frac_15_100);
+        fixed_div_uint(&temp_fixed, &prob_match,
+                       (c->num_main_syms - LZX_NUM_CHARS));
+        match_cost = lzx_cost_for_probability(&temp_fixed);
+        for (; i < c->num_main_syms; i++)
+                c->costs.main[i] = match_cost;
+
+        /* Length symbol costs.  These are just set to fixed values which
+         * reflect the fact the smallest lengths are typically the most common,
+         * and therefore are typically the cheapest. */
+        for (i = 0; i < LZX_LENCODE_NUM_SYMBOLS; i++)
+                c->costs.len[i] = lzx_default_len_costs[i];
+
+#if CONSIDER_ALIGNED_COSTS
+        /* Aligned offset symbol costs.  These are derived from the estimated
+         * probability of each aligned offset symbol. */
+        for (i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) {
+                /* We intentionally tallied the frequencies in the wrong slots,
+                 * not accounting for LZX_OFFSET_ADJUSTMENT, since doing the
+                 * fixup here is faster: a constant 8 subtractions here vs. one
+                 * addition for every match. */
+                unsigned j = (i - LZX_OFFSET_ADJUSTMENT) & LZX_ALIGNED_OFFSET_BITMASK;
+                if (c->freqs.aligned[j] != 0) {
+                        fixed32 prob;
+                        fixed_mul_uint_frac(&prob, c->freqs.aligned[j],
+                                            &inv_num_matches);
+                        c->costs.aligned[i] = lzx_cost_for_probability(&prob);
+                } else {
+                        c->costs.aligned[i] =
+                                (2 * LZX_NUM_ALIGNED_OFFSET_BITS) * BIT_COST;
+                }
+        }
+#endif
+}
+
+/* Update the current cost model to reflect the computed Huffman codes.  */
+static void
+lzx_set_costs_from_codes(struct liblzx_compressor *c)
+{
+        unsigned i;
+        const struct lzx_lens *lens = &c->codes[c->codes_index].lens;
+
+        for (i = 0; i < c->num_main_syms; i++) {
+                c->costs.main[i] = (lens->main[i] ? lens->main[i] :
+                                    MAIN_CODEWORD_LIMIT) * BIT_COST;
+        }
+
+        for (i = 0; i < LZX_LENCODE_NUM_SYMBOLS; i++) {
+                c->costs.len[i] = (lens->len[i] ? lens->len[i] :
+                                   LENGTH_CODEWORD_LIMIT) * BIT_COST;
+        }
+
+#if CONSIDER_ALIGNED_COSTS
+        for (i = 0; i < LZX_ALIGNEDCODE_NUM_SYMBOLS; i++) {
+                c->costs.aligned[i] = (lens->aligned[i] ? lens->aligned[i] :
+                                       ALIGNED_CODEWORD_LIMIT) * BIT_COST;
+        }
+#endif
+}
+
+/*
+ * Choose a "near-optimal" literal/match sequence to use for the current block,
+ * then flush the block.  Because the cost of each Huffman symbol is unknown
+ * until the Huffman codes have been built and the Huffman codes themselves
+ * depend on the symbol frequencies, this uses an iterative optimization
+ * algorithm to approximate an optimal solution.  The first optimization pass
+ * for the block uses default costs; additional passes use costs derived from
+ * the Huffman codes computed in the previous pass.
+ */
+static attrib_forceinline struct lzx_lru_queue
+lzx_optimize_and_flush_block(struct liblzx_compressor * const restrict c,
+                             struct lzx_output_bitstream * const restrict os,
+                             const uint8_t * const restrict block_begin,
+                             const uint32_t block_size,
+                             const struct lzx_lru_queue initial_queue,
+                             bool is_16_bit)
+{
+        unsigned num_passes_remaining = c->num_optim_passes;
+        struct lzx_lru_queue new_queue;
+        uint32_t seq_idx;
+
+        lzx_set_default_costs(c);
+
+        for (;;) {
+                lzx_compute_match_costs(c);
+                new_queue = lzx_find_min_cost_path(c, block_begin, block_size,
+                                                   initial_queue, is_16_bit);
+
+                if (--num_passes_remaining == 0)
+                        break;
+
+                /* At least one optimization pass remains.  Update the costs. */
+                lzx_reset_symbol_frequencies(c);
+                lzx_tally_item_list(c, block_size, is_16_bit);
+                lzx_build_huffman_codes(c);
+                lzx_set_costs_from_codes(c);
+        }
+
+        /* Done optimizing.  Generate the sequence list and flush the block. */
+        lzx_reset_symbol_frequencies(c);
+        seq_idx = lzx_record_item_list(c, block_size, is_16_bit);
+        lzx_flush_block(c, os, block_begin, block_size, seq_idx);
+        return new_queue;
+}
+
+/*
+ * This is the "near-optimal" LZX compressor.
+ *
+ * For each block, it performs a relatively thorough graph search to find an
+ * inexpensive (in terms of compressed size) way to output the block.
+ *
+ * Note: there are actually many things this algorithm leaves on the table in
+ * terms of compression ratio.  So although it may be "near-optimal", it is
+ * certainly not "optimal".  The goal is not to produce the optimal compression
+ * ratio, which for LZX is probably impossible within any practical amount of
+ * time, but rather to produce a compression ratio significantly better than a
+ * simpler "greedy" or "lazy" parse while still being relatively fast.
+ */
+static attrib_forceinline void
+lzx_reset_near_optimal(struct liblzx_compressor *c, bool is_16_bit)
+{
+        /* Initialize the matchfinder. */
+        CALL_BT_MF(is_16_bit, c, bt_matchfinder_init);
+}
+
+static void
+lzx_reset_near_optimal_16(struct liblzx_compressor *c)
+{
+        lzx_reset_near_optimal(c, true);
+}
+
+static void
+lzx_reset_near_optimal_32(struct liblzx_compressor *c)
+{
+        lzx_reset_near_optimal(c, false);
+}
+
+static attrib_forceinline void
+lzx_compress_near_optimal(struct liblzx_compressor * restrict c,
+                          const uint8_t *restrict in_begin,
+                          size_t in_nchunk, size_t in_ndata,
+                          struct lzx_output_bitstream * restrict os,
+                          bool is_16_bit)
+{
+        uint32_t max_offset                 = c->window_size;
+        const uint8_t *         in_next = in_begin;
+        const uint8_t * const in_chunk_end = in_begin + in_nchunk;
+        const uint8_t * const in_data_end = in_begin + in_ndata;
+        uint32_t max_find_len = LZX_MAX_MATCH_LEN;
+        uint32_t max_produce_len = LZX_MAX_MATCH_LEN;
+        uint32_t nice_len = min_u32(c->nice_match_length, max_find_len);
+        uint32_t next_hashes[2] = {0, 0};
+        struct lzx_lru_queue queue;
+
+        if (max_offset >= LZX_MAX_WINDOW_SIZE) {
+                /* Slightly shrink window to avoid offset values that are
+                 * greater than 21 bits. */
+                max_offset = LZX_MAX_WINDOW_SIZE - 1 - LZX_OFFSET_ADJUSTMENT;
+        }
+
+        in_begin -= c->in_prefix_size;
+
+        if (c->variant == LIBLZX_VARIANT_WIM) {
+                CALL_BT_MF(is_16_bit, c, bt_matchfinder_init);
+        } else {
+                /* Load the LRU queue */
+                lzx_lru_queue_load(&queue, c->lru_queue);
+        }
+
+        do {
+                /* Starting a new block */
+
+                const uint8_t * const in_block_begin = in_next;
+                const uint8_t * const in_max_block_end =
+                        in_next + min_size(SOFT_MAX_BLOCK_SIZE, in_chunk_end - in_next);
+                struct lz_match *cache_ptr = c->match_cache;
+                const uint8_t *next_search_pos = in_next;
+                const uint8_t *next_observation = in_next;
+                const uint8_t *next_pause_point =
+                        min_constptr(in_next + min_size(MIN_BLOCK_SIZE,
+                                          in_max_block_end - in_next),
+                    in_max_block_end - min_size(LZX_MAX_MATCH_LEN - 1,
+                                                   in_max_block_end - in_next));
+
+                lzx_init_block_split_stats(&c->split_stats);
+                lzx_reset_symbol_frequencies(c);
+
+                if (in_next >= next_pause_point)
+                        goto pause;
+
+                /*
+                 * Run the input buffer through the matchfinder, caching the
+                 * matches, until we decide to end the block.
+                 *
+                 * For a tighter matchfinding loop, we compute a "pause point",
+                 * which is the next position at which we may need to check
+                 * whether to end the block or to decrease max_len.  We then
+                 * only do these extra checks upon reaching the pause point.
+                 */
+        resume_matchfinding:
+                do {
+                        size_t min_match_pos = in_next - in_begin;
+                        min_match_pos -= min_size(min_match_pos, max_offset);
+
+                        if (in_next >= next_search_pos &&
+                                likely(nice_len >= LZX_MIN_MATCH_LEN)) {
+                                /* Search for matches at this position. */
+                                struct lz_match *lz_matchptr;
+                                uint32_t best_len;
+
+                                lz_matchptr = CALL_BT_MF(is_16_bit, c,
+                                                         bt_matchfinder_get_matches,
+                                                         in_begin,
+                                                         min_match_pos,
+                                                         in_next - in_begin,
+                                                         max_find_len,
+                                                         max_produce_len,
+                                                         nice_len,
+                                                         c->max_search_depth,
+                                                         next_hashes,
+                                                         &best_len,
+                                                         cache_ptr + 1);
+                                cache_ptr->length = lz_matchptr - (cache_ptr + 1);
+                                cache_ptr = lz_matchptr;
+
+                                /* Accumulate literal/match statistics for block
+                                 * splitting and for generating the initial cost
+                                 * model. */
+                                if (in_next >= next_observation) {
+                                        best_len = cache_ptr[-1].length;
+                                        if (best_len >= 3) {
+                                                /* Match (len >= 3) */
+
+                                                /*
+                                                 * Note: for performance reasons this has
+                                                 * been simplified significantly:
+                                                 *
+                                                 * - We wait until later to account for
+                                                 *   LZX_OFFSET_ADJUSTMENT.
+                                                 * - We don't account for repeat offsets.
+                                                 * - We don't account for different match headers.
+                                                 */
+                                                c->freqs.aligned[cache_ptr[-1].offset &
+                                                        LZX_ALIGNED_OFFSET_BITMASK]++;
+                                                c->freqs.main[LZX_NUM_CHARS]++;
+
+                                                lzx_observe_match(&c->split_stats, best_len);
+                                                next_observation = in_next + best_len;
+                                        } else {
+                                                /* Literal */
+                                                c->freqs.main[*in_next]++;
+                                                lzx_observe_literal(&c->split_stats, *in_next);
+                                                next_observation = in_next + 1;
+                                        }
+                                }
+
+                                /*
+                                 * If there was a very long match found, then
+                                 * don't cache any matches for the bytes covered
+                                 * by that match.  This avoids degenerate
+                                 * behavior when compressing highly redundant
+                                 * data, where the number of matches can be very
+                                 * large.
+                                 *
+                                 * This heuristic doesn't actually hurt the
+                                 * compression ratio *too* much.  If there's a
+                                 * long match, then the data must be highly
+                                 * compressible, so it doesn't matter as much
+                                 * what we do.
+                                 */
+                                if (best_len >= nice_len)
+                                        next_search_pos = in_next + best_len;
+                        } else {
+                                /* Don't search for matches at this position. */
+                                CALL_BT_MF(is_16_bit, c,
+                                           bt_matchfinder_skip_byte,
+                                           in_begin,
+                                           min_match_pos,
+                                           in_next - in_begin,
+                                           nice_len,
+                                           c->max_search_depth,
+                                           next_hashes);
+                                cache_ptr->length = 0;
+                                cache_ptr++;
+                        }
+                } while (++in_next < next_pause_point &&
+                         likely(cache_ptr < &c->match_cache[CACHE_LENGTH]));
+
+        pause:
+
+                /* Adjust max_len and nice_len if we're nearing the end of the
+                 * input buffer.  In addition, if we are so close to the end of
+                 * the input buffer that there cannot be any more matches, then
+                 * just advance through the last few positions and record no
+                 * matches. */
+                if (unlikely(max_produce_len > in_data_end - in_next)) {
+                        max_produce_len = in_chunk_end - in_next;
+                        max_find_len = in_data_end - in_next;
+                        nice_len = min_u32(max_produce_len, nice_len);
+                        if (max_find_len < BT_MATCHFINDER_REQUIRED_NBYTES) {
+                                while (in_next != in_chunk_end) {
+                                        cache_ptr->length = 0;
+                                        cache_ptr++;
+                                        in_next++;
+                                }
+                        }
+                }
+
+                /* End the block if the match cache may overflow. */
+                if (unlikely(cache_ptr >= &c->match_cache[CACHE_LENGTH]))
+                        goto end_block;
+
+                /* End the block if the soft maximum size has been reached. */
+                if (in_next >= in_max_block_end)
+                        goto end_block;
+
+                /* End the block if the block splitting algorithm thinks this is
+                 * a good place to do so. */
+                if (c->split_stats.num_new_observations >=
+                                NUM_OBSERVATIONS_PER_BLOCK_CHECK &&
+                    in_max_block_end - in_next >= MIN_BLOCK_SIZE &&
+                    lzx_should_end_block(&c->split_stats))
+                        goto end_block;
+
+                /* It's not time to end the block yet.  Compute the next pause
+                 * point and resume matchfinding. */
+                next_pause_point =
+                        min_constptr(in_next + min_size(NUM_OBSERVATIONS_PER_BLOCK_CHECK * 2 -
+                                            c->split_stats.num_new_observations,
+                                          in_max_block_end - in_next),
+                            in_max_block_end - min_size(LZX_MAX_MATCH_LEN - 1,
+                                                   in_max_block_end - in_next));
+                goto resume_matchfinding;
+
+        end_block:
+                /* We've decided on a block boundary and cached matches.  Now
+                 * choose a match/literal sequence and flush the block. */
+                queue = lzx_optimize_and_flush_block(c, os, in_block_begin,
+                                                     in_next - in_block_begin,
+                                                     queue, is_16_bit);
+        } while (in_next != in_chunk_end);
+
+        /* Save the LRU queue and next hashes */
+        lzx_lru_queue_save(c->lru_queue, &queue);
+}
+
+static void
+lzx_compress_near_optimal_16(struct liblzx_compressor *c, const uint8_t *in,
+                             size_t in_nchunk, size_t in_ndata,
+                             struct lzx_output_bitstream *os)
+{
+        lzx_compress_near_optimal(c, in, in_nchunk, in_ndata, os, true);
+}
+
+static void
+lzx_compress_near_optimal_32(struct liblzx_compressor *c, const uint8_t *in,
+                             size_t in_nchunk, size_t in_ndata,
+                             struct lzx_output_bitstream *os)
+{
+        lzx_compress_near_optimal(c, in, in_nchunk, in_ndata, os, false);
+}
+
+static attrib_forceinline void
+lzx_cull_near_optimal(struct liblzx_compressor *c, size_t nbytes, const bool is_16_bit)
+{
+        CALL_BT_MF(is_16_bit, c, bt_matchfinder_cull, nbytes, c->window_size);
+}
+
+static void
+lzx_cull_near_optimal_16(struct liblzx_compressor *c, size_t nbytes)
+{
+        lzx_cull_near_optimal(c, nbytes, true);
+}
+
+static void
+lzx_cull_near_optimal_32(struct liblzx_compressor *c, size_t nbytes)
+{
+        lzx_cull_near_optimal(c, nbytes, false);
+}
+
+/******************************************************************************/
+/*                     Faster ("lazy") compression algorithm                  */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * Called when the compressor chooses to use a literal.  This tallies the
+ * Huffman symbol for the literal, increments the current literal run length,
+ * and "observes" the literal for the block split statistics.
+ */
+static attrib_forceinline void
+lzx_choose_literal(struct liblzx_compressor *c, unsigned literal, uint32_t *litrunlen_p)
+{
+        lzx_observe_literal(&c->split_stats, literal);
+        c->freqs.main[literal]++;
+        ++*litrunlen_p;
+}
+
+/*
+ * Called when the compressor chooses to use a match.  This tallies the Huffman
+ * symbol(s) for a match, saves the match data and the length of the preceding
+ * literal run, updates the recent offsets queue, and "observes" the match for
+ * the block split statistics.
+ */
+static attrib_forceinline void
+lzx_choose_match(struct liblzx_compressor *c, unsigned length, uint32_t adjusted_offset,
+                 uint32_t recent_offsets[LZX_NUM_RECENT_OFFSETS], bool is_16_bit,
+                 uint32_t *litrunlen_p, struct lzx_sequence **next_seq_p)
+{
+        struct lzx_sequence *next_seq = *next_seq_p;
+        unsigned mainsym;
+
+        lzx_observe_match(&c->split_stats, length);
+
+        mainsym = lzx_tally_main_and_lensyms(c, length, adjusted_offset,
+                                             is_16_bit);
+        next_seq->litrunlen_and_matchlen =
+                (*litrunlen_p << SEQ_MATCHLEN_BITS) | length;
+        next_seq->adjusted_offset_and_mainsym =
+                (adjusted_offset << SEQ_MAINSYM_BITS) | mainsym;
+
+        /* Update the recent offsets queue. */
+        if (adjusted_offset < LZX_NUM_RECENT_OFFSETS) {
+                /* Repeat offset match. */
+                uint32_t temp = recent_offsets[adjusted_offset];
+                recent_offsets[adjusted_offset] = recent_offsets[0];
+                recent_offsets[0] = temp;
+        } else {
+                /* Explicit offset match. */
+
+                /* Tally the aligned offset symbol if needed. */
+                if (adjusted_offset >= LZX_MIN_ALIGNED_OFFSET + LZX_OFFSET_ADJUSTMENT)
+                        c->freqs.aligned[adjusted_offset & LZX_ALIGNED_OFFSET_BITMASK]++;
+
+                recent_offsets[2] = recent_offsets[1];
+                recent_offsets[1] = recent_offsets[0];
+                recent_offsets[0] = adjusted_offset - LZX_OFFSET_ADJUSTMENT;
+        }
+
+        /* Reset the literal run length and advance to the next sequence. */
+        *next_seq_p = next_seq + 1;
+        *litrunlen_p = 0;
+}
+
+/*
+ * Called when the compressor ends a block.  This finshes the last lzx_sequence,
+ * which is just a literal run with no following match.  This literal run might
+ * be empty.
+ */
+static attrib_forceinline void
+lzx_finish_sequence(struct lzx_sequence *last_seq, uint32_t litrunlen)
+{
+        last_seq->litrunlen_and_matchlen = litrunlen << SEQ_MATCHLEN_BITS;
+}
+
+/*
+ * Find the longest repeat offset match with the current position.  If a match
+ * is found, return its length and set *best_rep_idx_ret to the index of its
+ * offset in @recent_offsets.  Otherwise, return 0.
+ *
+ * Don't bother with length 2 matches; consider matches of length >= 3 only.
+ * Also assume that max_len >= 3.
+ */
+static unsigned
+lzx_find_longest_repeat_offset_match(const uint8_t * const in_next,
+                                     const uint32_t recent_offsets[],
+                                     const unsigned max_len,
+                                     unsigned *best_rep_idx_ret)
+{
+        STATIC_ASSERT(LZX_NUM_RECENT_OFFSETS == 3); /* loop is unrolled */
+
+        const uint32_t seq3 = load_u24_unaligned(in_next);
+        const uint8_t *matchptr;
+        unsigned best_rep_len = 0;
+        unsigned best_rep_idx = 0;
+        unsigned rep_len;
+
+        /* Check for rep0 match (most recent offset) */
+        matchptr = in_next - recent_offsets[0];
+        if (load_u24_unaligned(matchptr) == seq3)
+                best_rep_len = lz_extend(in_next, matchptr, 3, max_len);
+
+        /* Check for rep1 match (second most recent offset) */
+        matchptr = in_next - recent_offsets[1];
+        if (load_u24_unaligned(matchptr) == seq3) {
+                rep_len = lz_extend(in_next, matchptr, 3, max_len);
+                if (rep_len > best_rep_len) {
+                        best_rep_len = rep_len;
+                        best_rep_idx = 1;
+                }
+        }
+
+        /* Check for rep2 match (third most recent offset) */
+        matchptr = in_next - recent_offsets[2];
+        if (load_u24_unaligned(matchptr) == seq3) {
+                rep_len = lz_extend(in_next, matchptr, 3, max_len);
+                if (rep_len > best_rep_len) {
+                        best_rep_len = rep_len;
+                        best_rep_idx = 2;
+                }
+        }
+
+        *best_rep_idx_ret = best_rep_idx;
+        return best_rep_len;
+}
+
+/*
+ * Fast heuristic scoring for lazy parsing: how "good" is this match?
+ * This is mainly determined by the length: longer matches are better.
+ * However, we also give a bonus to close (small offset) matches and to repeat
+ * offset matches, since those require fewer bits to encode.
+ */
+
+static attrib_forceinline unsigned
+lzx_explicit_offset_match_score(unsigned len, uint32_t adjusted_offset)
+{
+        unsigned score = len;
+
+        if (adjusted_offset < 4096)
+                score++;
+        if (adjusted_offset < 256)
+                score++;
+
+        return score;
+}
+
+static attrib_forceinline unsigned
+lzx_repeat_offset_match_score(unsigned rep_len, unsigned rep_idx)
+{
+        return rep_len + 3;
+}
+
+/*
+ * This is the "lazy" LZX compressor.  The basic idea is that before it chooses
+ * a match, it checks to see if there's a longer match at the next position.  If
+ * yes, it chooses a literal and continues to the next position.  If no, it
+ * chooses the match.
+ *
+ * Some additional heuristics are used as well.  Repeat offset matches are
+ * considered favorably and sometimes are chosen immediately.  In addition, long
+ * matches (at least "nice_len" bytes) are chosen immediately as well.  Finally,
+ * when we decide whether a match is "better" than another, we take the offset
+ * into consideration as well as the length.
+ */
+static attrib_forceinline void
+lzx_reset_lazy(struct liblzx_compressor *c, bool is_16_bit)
+{
+        bool streaming = (c->variant != LIBLZX_VARIANT_WIM);
+
+        /* Initialize the matchfinder. */
+        CALL_HC_MF(is_16_bit, c, hc_matchfinder_init, c->window_size,
+                   streaming);
+}
+
+static void
+lzx_reset_lazy_16(struct liblzx_compressor *c)
+{
+        lzx_reset_lazy(c, true);
+}
+
+static void
+lzx_reset_lazy_32(struct liblzx_compressor *c)
+{
+        lzx_reset_lazy(c, false);
+}
+
+static attrib_forceinline void
+lzx_compress_lazy(struct liblzx_compressor * restrict c,
+                  const uint8_t * restrict in_begin, size_t in_nchunk,
+                  size_t in_ndata, struct lzx_output_bitstream * restrict os,
+                  bool is_16_bit)
+{
+        uint32_t max_offset                 = c->window_size;
+        const uint8_t *         in_next = in_begin;
+        const uint8_t * const in_chunk_end = in_begin + in_nchunk;
+        const uint8_t *const in_data_end = in_begin + in_ndata;
+        unsigned max_find_len = LZX_MAX_MATCH_LEN;
+        unsigned max_produce_len = LZX_MAX_MATCH_LEN;
+        unsigned nice_len = min_uint(c->nice_match_length, max_find_len);
+        STATIC_ASSERT(LZX_NUM_RECENT_OFFSETS == 3);
+        uint32_t recent_offsets[LZX_NUM_RECENT_OFFSETS];
+        uint32_t next_hashes[2];
+
+        if (max_offset >= LZX_MAX_WINDOW_SIZE) {
+                /* Slightly shrink window to avoid offset values that are
+                 * greater than 21 bits. */
+                max_offset = LZX_MAX_WINDOW_SIZE - 1 - LZX_OFFSET_ADJUSTMENT;
+        }
+
+        in_begin -= c->in_prefix_size;
+
+        /* Load the LRU queue and next hashes. */
+        {
+                int i;
+                for (i = 0; i < LZX_NUM_RECENT_OFFSETS; i++) {
+                        recent_offsets[i] = c->lru_queue[i];
+                }
+
+                next_hashes[0] = c->next_hashes[0];
+                next_hashes[1] = c->next_hashes[1];
+        }
+
+        do {
+                /* Starting a new block */
+
+                const uint8_t * const in_block_begin = in_next;
+                const uint8_t * const in_max_block_end =
+                        in_next + min_size(SOFT_MAX_BLOCK_SIZE, in_chunk_end - in_next);
+                struct lzx_sequence *next_seq = c->chosen_sequences;
+                uint32_t litrunlen = 0;
+                unsigned cur_len;
+                uint32_t cur_offset;
+                uint32_t cur_adjusted_offset;
+                unsigned cur_score;
+                unsigned next_len;
+                uint32_t next_offset;
+                uint32_t next_adjusted_offset;
+                unsigned next_score;
+                unsigned best_rep_len;
+                unsigned best_rep_idx;
+                unsigned rep_score;
+                unsigned skip_len;
+
+                lzx_reset_symbol_frequencies(c);
+                lzx_init_block_split_stats(&c->split_stats);
+
+                do {
+                        /* Adjust max_len and nice_len if we're nearing the end
+                         * of the input buffer. */
+                        if (unlikely(max_produce_len >
+                                     in_chunk_end - in_next)) {
+                                max_produce_len = in_chunk_end - in_next;
+                                max_find_len = in_data_end - in_next;
+                                nice_len =
+                                    min_uint(max_produce_len, nice_len);
+                        }
+
+                        /* Find the longest match (subject to the
+                         * max_search_depth cutoff parameter) with the current
+                         * position.  Don't bother with length 2 matches; only
+                         * look for matches of length >= 3. */
+                        {
+                                size_t min_match_pos = in_next - in_begin;
+                                min_match_pos -=
+                                    min_size(min_match_pos, max_offset);
+
+                                cur_len = CALL_HC_MF(is_16_bit, c,
+                                                     hc_matchfinder_longest_match,
+                                                     in_begin,
+                                                     min_match_pos,
+                                                     in_next,
+                                                     2,
+                                                     max_find_len,
+                                                     max_produce_len,
+                                                     nice_len,
+                                                     c->max_search_depth,
+                                                     next_hashes,
+                                                     &cur_offset);
+                        }
+
+                        /* If there was no match found, or the only match found
+                         * was a distant short match, then choose a literal. */
+                        if (cur_len < 3 ||
+                            (cur_len == 3 &&
+                             cur_offset >= 8192 - LZX_OFFSET_ADJUSTMENT &&
+                             cur_offset != recent_offsets[0] &&
+                             cur_offset != recent_offsets[1] &&
+                             cur_offset != recent_offsets[2]))
+                        {
+                                lzx_choose_literal(c, *in_next, &litrunlen);
+                                in_next++;
+                                continue;
+                        }
+
+                        /* Heuristic: if this match has the most recent offset,
+                         * then go ahead and choose it as a rep0 match. */
+                        if (cur_offset == recent_offsets[0]) {
+                                in_next++;
+                                skip_len = cur_len - 1;
+                                cur_adjusted_offset = 0;
+                                goto choose_cur_match;
+                        }
+
+                        /* Compute the longest match's score as an explicit
+                         * offset match. */
+                        cur_adjusted_offset = cur_offset + LZX_OFFSET_ADJUSTMENT;
+                        cur_score = lzx_explicit_offset_match_score(cur_len, cur_adjusted_offset);
+
+                        /* Find the longest repeat offset match at this
+                         * position.  If we find one and it's "better" than the
+                         * explicit offset match we found, then go ahead and
+                         * choose the repeat offset match immediately. */
+                        best_rep_len = lzx_find_longest_repeat_offset_match(in_next,
+                                                                            recent_offsets,
+                                                                            max_produce_len,
+                                                                            &best_rep_idx);
+                        in_next++;
+
+                        if (best_rep_len != 0 &&
+                            (rep_score = lzx_repeat_offset_match_score(best_rep_len,
+                                                                       best_rep_idx)) >= cur_score)
+                        {
+                                cur_len = best_rep_len;
+                                cur_adjusted_offset = best_rep_idx;
+                                skip_len = best_rep_len - 1;
+                                goto choose_cur_match;
+                        }
+
+                have_cur_match:
+                        /*
+                         * We have a match at the current position.  If the
+                         * match is very long, then choose it immediately.
+                         * Otherwise, see if there's a better match at the next
+                         * position.
+                         */
+
+                        if (cur_len >= nice_len) {
+                                skip_len = cur_len - 1;
+                                goto choose_cur_match;
+                        }
+
+                        if (unlikely(max_produce_len >
+                                     in_chunk_end - in_next)) {
+                                max_produce_len = in_chunk_end - in_next;
+                                max_find_len = in_data_end - in_next;
+                                nice_len =
+                                    min_uint(max_produce_len, nice_len);
+                        }
+
+                        {
+                                size_t min_match_pos = in_next - in_begin;
+                                min_match_pos -=
+                                    min_uint(min_match_pos, max_offset);
+
+                                next_len = CALL_HC_MF(
+                                                      is_16_bit, c,
+                                                      hc_matchfinder_longest_match,
+                                                      in_begin,
+                                                      min_match_pos,
+                                                      in_next,
+                                                      cur_len - 2,
+                                                      max_find_len,
+                                                      max_produce_len,
+                                                      nice_len,
+                                                      c->max_search_depth / 2,
+                                                      next_hashes,
+                                                      &next_offset);
+                        }
+
+                        if (next_len <= cur_len - 2) {
+                                /* No potentially better match was found. */
+                                in_next++;
+                                skip_len = cur_len - 2;
+                                goto choose_cur_match;
+                        }
+
+                        next_adjusted_offset = next_offset + LZX_OFFSET_ADJUSTMENT;
+                        next_score = lzx_explicit_offset_match_score(next_len, next_adjusted_offset);
+
+                        best_rep_len = lzx_find_longest_repeat_offset_match(in_next,
+                                                                            recent_offsets,
+                                                                            max_produce_len,
+                                                                            &best_rep_idx);
+                        in_next++;
+
+                        if (best_rep_len != 0 &&
+                            (rep_score = lzx_repeat_offset_match_score(best_rep_len,
+                                                                       best_rep_idx)) >= next_score)
+                        {
+
+                                if (rep_score > cur_score) {
+                                        /* The next match is better, and it's a
+                                         * repeat offset match. */
+                                        lzx_choose_literal(c, *(in_next - 2),
+                                                           &litrunlen);
+                                        cur_len = best_rep_len;
+                                        cur_adjusted_offset = best_rep_idx;
+                                        skip_len = cur_len - 1;
+                                        goto choose_cur_match;
+                                }
+                        } else {
+                                if (next_score > cur_score) {
+                                        /* The next match is better, and it's an
+                                         * explicit offset match. */
+                                        lzx_choose_literal(c, *(in_next - 2),
+                                                           &litrunlen);
+                                        cur_len = next_len;
+                                        cur_adjusted_offset = next_adjusted_offset;
+                                        cur_score = next_score;
+                                        goto have_cur_match;
+                                }
+                        }
+
+                        /* The original match was better; choose it. */
+                        skip_len = cur_len - 2;
+
+                choose_cur_match:
+                        /* Choose a match and have the matchfinder skip over its
+                         * remaining bytes. */
+                        lzx_choose_match(c, cur_len, cur_adjusted_offset,
+                                         recent_offsets, is_16_bit,
+                                         &litrunlen, &next_seq);
+
+                        CALL_HC_MF(is_16_bit, c,
+                                   hc_matchfinder_skip_bytes,
+                                   in_begin,
+                                   in_next,
+                                   in_chunk_end,
+                                   skip_len,
+                                   next_hashes);
+                        in_next += skip_len;
+
+                        /* Keep going until it's time to end the block. */
+                } while (in_next < in_max_block_end &&
+                         !(c->split_stats.num_new_observations >=
+                                        NUM_OBSERVATIONS_PER_BLOCK_CHECK &&
+                           in_next - in_block_begin >= MIN_BLOCK_SIZE &&
+                           in_chunk_end - in_next >= MIN_BLOCK_SIZE &&
+                           lzx_should_end_block(&c->split_stats)));
+
+                /* Flush the block. */
+                lzx_finish_sequence(next_seq, litrunlen);
+                lzx_flush_block(c, os, in_block_begin, in_next - in_block_begin, 0);
+
+                /* Keep going until we've reached the end of the input buffer. */
+        } while (in_next != in_chunk_end);
+
+        /* Save the LRU queue and next hashes */
+        {
+                int i;
+                for (i = 0; i < LZX_NUM_RECENT_OFFSETS; i++) {
+                        c->lru_queue[i] = recent_offsets[i];
+                }
+                c->next_hashes[0] = next_hashes[0];
+                c->next_hashes[1] = next_hashes[1];
+        }
+}
+
+static void
+lzx_compress_lazy_16(struct liblzx_compressor *c, const uint8_t *in,
+                     size_t in_nchunk, size_t in_navail,
+                     struct lzx_output_bitstream *os)
+{
+        lzx_compress_lazy(c, in, in_nchunk, in_navail, os, true);
+}
+
+static void
+lzx_compress_lazy_32(struct liblzx_compressor *c, const uint8_t *in,
+                     size_t in_nchunk, size_t in_navail,
+                     struct lzx_output_bitstream *os)
+{
+        lzx_compress_lazy(c, in, in_nchunk, in_navail, os, false);
+}
+
+static void
+lzx_cull_lazy_16(struct liblzx_compressor *c, size_t nbytes)
+{
+        CALL_HC_MF(true, c, hc_matchfinder_cull, nbytes, c->window_size);
+}
+
+static void
+lzx_cull_lazy_32(struct liblzx_compressor *c, size_t nbytes)
+{
+        CALL_HC_MF(false, c, hc_matchfinder_cull, nbytes, c->window_size);
+}
+
+/******************************************************************************/
+/*                          Compressor operations                             */
+/*----------------------------------------------------------------------------*/
+
+/*
+ * Generate tables for mapping match offsets (actually, "adjusted" match
+ * offsets) to offset slots.
+ */
+static void
+lzx_init_offset_slot_tabs(struct liblzx_compressor *c)
+{
+        uint32_t adjusted_offset = 0;
+        unsigned slot = 0;
+
+        /* slots [0, 29] */
+        for (; adjusted_offset < ARRAY_LEN(c->offset_slot_tab_1);
+             adjusted_offset++)
+        {
+                if (adjusted_offset >= lzx_offset_slot_base[slot + 1] +
+                                       LZX_OFFSET_ADJUSTMENT)
+                        slot++;
+                c->offset_slot_tab_1[adjusted_offset] = slot;
+        }
+
+        /* slots [30, 49] */
+        for (; adjusted_offset < LZX_MAX_WINDOW_SIZE;
+             adjusted_offset += (uint32_t)1 << 14)
+        {
+                if (adjusted_offset >= lzx_offset_slot_base[slot + 1] +
+                                       LZX_OFFSET_ADJUSTMENT)
+                        slot++;
+                c->offset_slot_tab_2[adjusted_offset >> 14] = slot;
+        }
+}
+
+static size_t
+lzx_bt_max_search_depth(unsigned compression_level)
+{
+        return (24 * compression_level) / 50;
+}
+
+static size_t
+lzx_get_compressor_size(size_t window_size, unsigned compression_level,
+        bool streaming)
+{
+
+        if (compression_level <= MAX_FAST_LEVEL) {
+                if (lzx_is_16_bit(window_size))
+                        return offsetof(struct liblzx_compressor, hc_mf_16) +
+                               hc_matchfinder_size_16(window_size, streaming);
+                else
+                        return offsetof(struct liblzx_compressor, hc_mf_32) +
+                               hc_matchfinder_size_32(window_size, streaming);
+        } else {
+                if (lzx_is_16_bit(window_size))
+                        return offsetof(struct liblzx_compressor, bt_mf_16) +
+                               bt_matchfinder_size_16(window_size, streaming);
+                else
+                        return offsetof(struct liblzx_compressor, bt_mf_32) +
+                               bt_matchfinder_size_32(window_size, streaming);
+        }
+}
+
+/* Compress a buffer of data. */
+static void
+lzx_reset(struct liblzx_compressor *c)
+{
+        /* Initially, the previous Huffman codeword lengths are all zeroes. */
+        c->codes_index = 0;
+        memset(&c->codes[1].lens, 0, sizeof(struct lzx_lens));
+
+        /* Reset the E8 preprocessor offset */
+        c->e8_chunk_offset = 0;
+
+        /* Reset the streaming prefix */
+        c->in_prefix_size = 0;
+
+        /* Reset the LRU queue */
+        {
+                int i;
+                for (i = 0; i < LZX_NUM_RECENT_OFFSETS; i++) {
+                        c->lru_queue[i] = 1;
+                }
+        }
+
+        /* Reset next hashes */
+        c->next_hashes[0] = 0;
+        c->next_hashes[1] = 0;
+
+        c->reset(c);
+}
+
+/* Allocate an LZX compressor. */
+liblzx_compressor_t *
+liblzx_compress_create(const struct liblzx_compress_properties *props)
+{
+        unsigned window_order;
+        struct liblzx_compressor *c;
+        bool streaming = (props->lzx_variant != LIBLZX_VARIANT_WIM);
+
+        /* Validate the maximum buffer size and get the window order from it. */
+        window_order = lzx_get_window_order(props->window_size);
+        if (window_order == 0)
+                return NULL;
+
+        /* Allocate the compressor. */
+        c = props->alloc_func(props->userdata,
+            lzx_get_compressor_size(props->window_size, props->compression_level, streaming));
+        if (!c)
+                goto oom0;
+
+        c->alloc_func = props->alloc_func;
+        c->free_func = props->free_func;
+        c->alloc_userdata = props->userdata;
+        c->window_size = props->window_size;
+        c->window_order = window_order;
+        c->num_main_syms = lzx_get_num_main_syms(window_order);
+        c->variant = props->lzx_variant;
+        c->first_block = true;
+        c->out_chunk.data = NULL;
+        c->out_chunk.size = 0;
+        c->flushing = false;
+        c->e8_chunk_offset = 0;
+        c->e8_file_size = props->e8_file_size;
+        c->in_buffer_capacity = c->window_size;
+        c->in_prefix_size = 0;
+        c->in_used = 0;
+        c->chunk_size = props->chunk_granularity;
+
+        /* Allocate the buffer for preprocessed data if needed. */
+        if (streaming) {
+                /* Pad out to include past blocks and extra
+                 * matchfinding space  */
+                c->in_buffer_capacity *= 2;
+                c->in_buffer_capacity +=
+                    LZX_MAX_MATCH_LEN + LZX_E8_FILTER_TAIL_SIZE;
+        }
+
+        if (c->variant == LIBLZX_VARIANT_WIM)
+                c->e8_file_size = LZX_WIM_MAGIC_FILESIZE;
+
+        c->in_buffer =
+            props->alloc_func(props->userdata, c->in_buffer_capacity);
+
+        if (!c->in_buffer)
+                goto oom1;
+
+        c->out_buffer_capacity = c->chunk_size;
+        if (c->variant != LIBLZX_VARIANT_WIM)
+                c->out_buffer_capacity += 6144;
+
+        c->out_chunk.data = c->out_buffer =
+            props->alloc_func(props->userdata, c->out_buffer_capacity);
+
+        if (!c->out_buffer)
+                goto oom2;
+
+        if (props->compression_level <= MAX_FAST_LEVEL) {
+
+                /* Fast compression: Use lazy parsing. */
+                if (lzx_is_16_bit(props->window_size)) {
+                        c->reset = lzx_reset_lazy_16;
+                        c->impl = lzx_compress_lazy_16;
+                        c->cull = lzx_cull_lazy_16;
+                } else {
+                        c->reset = lzx_reset_lazy_32;
+                        c->impl = lzx_compress_lazy_32;
+                        c->cull = lzx_cull_lazy_32;
+                }
+
+                /* Scale max_search_depth and nice_match_length with the
+                 * compression level. */
+                c->max_search_depth = (60 * props->compression_level) / 20;
+                c->nice_match_length = (80 * props->compression_level) / 20;
+
+                /* lzx_compress_lazy() needs max_search_depth >= 2 because it
+                 * halves the max_search_depth when attempting a lazy match, and
+                 * max_search_depth must be at least 1. */
+                c->max_search_depth = max_uint(c->max_search_depth, 2);
+        } else {
+
+                /* Normal / high compression: Use near-optimal parsing. */
+                if (lzx_is_16_bit(c->window_size)) {
+                        c->reset = lzx_reset_near_optimal_16;
+                        c->impl = lzx_compress_near_optimal_16;
+                        c->cull = lzx_cull_near_optimal_16;
+                } else {
+                        c->reset = lzx_reset_near_optimal_32;
+                        c->impl = lzx_compress_near_optimal_32;
+                        c->cull = lzx_cull_near_optimal_32;
+                }
+
+                /* Scale max_search_depth and nice_match_length with the
+                 * compression level. */
+                c->max_search_depth = lzx_bt_max_search_depth(props->compression_level);
+                c->nice_match_length = (48 * props->compression_level) / 50;
+
+                /* Also scale num_optim_passes with the compression level.  But
+                 * the more passes there are, the less they help --- so don't
+                 * add them linearly.  */
+                c->num_optim_passes = 1;
+                c->num_optim_passes += (props->compression_level >= 45);
+                c->num_optim_passes += (props->compression_level >= 70);
+                c->num_optim_passes += (props->compression_level >= 100);
+                c->num_optim_passes += (props->compression_level >= 150);
+                c->num_optim_passes += (props->compression_level >= 200);
+                c->num_optim_passes += (props->compression_level >= 300);
+
+                /* max_search_depth must be at least 1. */
+                c->max_search_depth = max_uint(c->max_search_depth, 1);
+        }
+
+        /* Prepare the offset => offset slot mapping. */
+        lzx_init_offset_slot_tabs(c);
+
+        lzx_reset(c);
+
+        return c;
+
+oom2:
+        props->free_func(props->userdata, c->in_buffer);
+oom1:
+        props->free_func(props->userdata, c);
+oom0:
+        return NULL;
+}
+
+/* Compress a buffer of data. */
+static size_t
+lzx_compress_chunk(struct liblzx_compressor *c)
+{
+        struct lzx_output_bitstream os;
+        size_t result;
+        bool e8_preprocess_enabled = (c->e8_chunk_offset < 0x40000000);
+        bool next_e8_preprocess_enabled =
+            (c->e8_chunk_offset + c->chunk_size < 0x40000000);
+        uint32_t chunk_size = min_u32(c->chunk_size, c->in_used);
+        uint32_t next_chunk_preprocess_size = 0;
+
+        uint8_t *in = (uint8_t *)c->in_buffer + c->in_prefix_size;
+
+        /* Preprocess the input data. */
+        if (e8_preprocess_enabled) {
+                lzx_preprocess(in, chunk_size, c->e8_chunk_offset,
+                               c->e8_file_size);
+        }
+
+        if (c->in_used > c->chunk_size && next_e8_preprocess_enabled) {
+                next_chunk_preprocess_size =
+                    min_u32(LZX_MAX_MATCH_LEN + LZX_E8_FILTER_TAIL_SIZE,
+                            c->in_used - c->chunk_size);
+        }
+
+        /* Preprocess enough of the next block input data for the
+           matchfinder */
+        if (next_chunk_preprocess_size > 0) {
+                lzx_preprocess(in + c->chunk_size, next_chunk_preprocess_size,
+                               c->e8_chunk_offset + c->chunk_size,
+                               c->e8_file_size);
+        }
+
+        /* Initialize the output bitstream. */
+        lzx_init_output(&os, c->out_buffer, c->out_buffer_capacity);
+
+        /* Call the compression level-specific compress() function. */
+        (*c->impl)(c, in, chunk_size, c->in_used, &os);
+
+        /* Undo next block preprocessing */
+        if (next_chunk_preprocess_size > 0) {
+                lzx_postprocess(in + c->chunk_size, next_chunk_preprocess_size,
+                               c->e8_chunk_offset + c->chunk_size,
+                               c->e8_file_size);
+        }
+
+        /* Flush the output bitstream. */
+        result = lzx_flush_output(&os);
+
+        /* Update the E8 chunk offset. */
+        c->e8_chunk_offset += (uint32_t)chunk_size;
+
+        /* Update the prefix and used amounts. */
+        c->in_prefix_size += (uint32_t)chunk_size;
+        c->in_used -= chunk_size;
+
+        if (c->in_prefix_size >= c->window_size * 2) {
+                uint32_t cull_amount = (c->in_prefix_size - c->window_size);
+
+                in = (uint8_t *)c->in_buffer + c->in_prefix_size;
+
+                memmove(c->in_buffer, in - c->window_size,
+                        c->in_used + c->window_size);
+                c->in_prefix_size = c->window_size;
+
+                (*c->cull)(c, cull_amount);
+        }
+
+        /* Return the number of compressed bytes, or 0 if the input did not
+         * compress to less than its original size. */
+        return result;
+}
+
+void
+liblzx_compress_destroy(liblzx_compressor_t *c)
+{
+        c->free_func(c->alloc_userdata, c->out_buffer);
+        c->free_func(c->alloc_userdata, c->in_buffer);
+        c->free_func(c->alloc_userdata, c);
+}
+
+size_t
+liblzx_compress_add_input(liblzx_compressor_t *c, const void *in_data,
+                          size_t in_data_size)
+{
+        uint32_t max_used = 0;
+        size_t fill_amount = 0;
+
+        if (c->out_chunk.size > 0 || c->flushing)
+                return 0;
+
+        max_used = min_uint(c->in_buffer_capacity - c->in_prefix_size,
+                            c->chunk_size + LZX_MAX_MATCH_LEN +
+                                LZX_E8_FILTER_TAIL_SIZE);
+        fill_amount = min_size(in_data_size, max_used - c->in_used);
+
+        memcpy(((uint8_t *)c->in_buffer) + c->in_prefix_size + c->in_used, in_data,
+               fill_amount);
+
+        c->in_used += fill_amount;
+
+        if (c->in_used == max_used) {
+                c->out_chunk.size = lzx_compress_chunk(c);
+        }
+
+        return fill_amount;
+}
+
+const liblzx_output_chunk_t *
+liblzx_compress_get_next_chunk(const liblzx_compressor_t *c)
+{
+        if (c->out_chunk.size > 0)
+                return &c->out_chunk;
+        else
+                return NULL;
+}
+
+void
+liblzx_compress_release_next_chunk(liblzx_compressor_t *c)
+{
+        c->out_chunk.size = 0;
+        if (c->flushing && c->in_used > 0) {
+                c->out_chunk.size = lzx_compress_chunk(c);
+        }
+}
+
+void
+liblzx_compress_end_input(liblzx_compressor_t *c)
+{
+        if (!c->flushing) {
+                c->flushing = true;
+                if (c->in_used > 0 && c->out_chunk.size == 0) {
+                        c->out_chunk.size = lzx_compress_chunk(c);
+                }
+        }
+}
diff --git a/dlls/cabinet/liblzx_lzx_constants.h b/dlls/cabinet/liblzx_lzx_constants.h
new file mode 100644
index 00000000000..f11ce407873
--- /dev/null
+++ b/dlls/cabinet/liblzx_lzx_constants.h
@@ -0,0 +1,108 @@
+/*
+ * lzx_constants.h
+ *
+ * Constants for the LZX compression format.
+ */
+
+#ifndef _LZX_CONSTANTS_H
+#define _LZX_CONSTANTS_H
+
+/* Number of literal byte values.  */
+#define LZX_NUM_CHARS        256
+
+/* The smallest and largest allowed match lengths.  */
+#define LZX_MIN_MATCH_LEN        2
+#define LZX_MAX_MATCH_LEN        257
+
+/* Number of distinct match lengths that can be represented.  */
+#define LZX_NUM_LENS                    (LZX_MAX_MATCH_LEN - LZX_MIN_MATCH_LEN + 1)
+
+/* Number of match lengths for which no length symbol is required.  */
+#define LZX_NUM_PRIMARY_LENS            7
+#define LZX_NUM_LEN_HEADERS             (LZX_NUM_PRIMARY_LENS + 1)
+
+/* The first length which requires a length symbol.  */
+#define LZX_MIN_SECONDARY_LEN           (LZX_MIN_MATCH_LEN + LZX_NUM_PRIMARY_LENS)
+
+/* Valid values of the 3-bit block type field.  */
+#define LZX_BLOCKTYPE_VERBATIM          1
+#define LZX_BLOCKTYPE_ALIGNED           2
+#define LZX_BLOCKTYPE_UNCOMPRESSED      3
+
+/* 'LZX_MIN_WINDOW_SIZE' and 'LZX_MAX_WINDOW_SIZE' are the minimum and maximum
+ * sizes of the sliding window.  */
+#define LZX_MIN_WINDOW_ORDER            15
+#define LZX_MAX_WINDOW_ORDER            21
+#define LZX_MIN_WINDOW_SIZE             (1UL << LZX_MIN_WINDOW_ORDER)  /* 32768   */
+#define LZX_MAX_WINDOW_SIZE             (1UL << LZX_MAX_WINDOW_ORDER)  /* 2097152 */
+
+/* Maximum number of offset slots.  (The actual number of offset slots depends
+ * on the window size.)  */
+#define LZX_MAX_OFFSET_SLOTS            50
+
+/* Maximum number of symbols in the main code.  (The actual number of symbols in
+ * the main code depends on the window size.)  */
+#define LZX_MAINCODE_MAX_NUM_SYMBOLS        \
+        (LZX_NUM_CHARS + (LZX_MAX_OFFSET_SLOTS * LZX_NUM_LEN_HEADERS))
+
+/* Number of symbols in the length code.  */
+#define LZX_LENCODE_NUM_SYMBOLS         (LZX_NUM_LENS - LZX_NUM_PRIMARY_LENS)
+
+/* Number of symbols in the pre-code.  */
+#define LZX_PRECODE_NUM_SYMBOLS         20
+
+/* Number of bits in which each pre-code codeword length is represented.  */
+#define LZX_PRECODE_ELEMENT_SIZE        4
+
+/* Number of low-order bits of each match offset that are entropy-encoded in
+ * aligned offset blocks.  */
+#define LZX_NUM_ALIGNED_OFFSET_BITS     3
+
+/* Number of symbols in the aligned offset code.  */
+#define LZX_ALIGNEDCODE_NUM_SYMBOLS     (1 << LZX_NUM_ALIGNED_OFFSET_BITS)
+
+/* Mask for the match offset bits that are entropy-encoded in aligned offset
+ * blocks.  */
+#define LZX_ALIGNED_OFFSET_BITMASK      ((1 << LZX_NUM_ALIGNED_OFFSET_BITS) - 1)
+
+/* Number of bits in which each aligned offset codeword length is represented.  */
+#define LZX_ALIGNEDCODE_ELEMENT_SIZE    3
+
+/* The first offset slot which requires an aligned offset symbol in aligned
+ * offset blocks.  */
+#define LZX_MIN_ALIGNED_OFFSET_SLOT     8
+
+/* The offset slot base for LZX_MIN_ALIGNED_OFFSET_SLOT.  */
+#define LZX_MIN_ALIGNED_OFFSET          14
+
+/* The maximum number of extra offset bits in verbatim blocks.  (One would need
+ * to subtract LZX_NUM_ALIGNED_OFFSET_BITS to get the number of extra offset
+ * bits in *aligned* blocks.)  */
+#define LZX_MAX_NUM_EXTRA_BITS          17
+
+/* Maximum lengths (in bits) for length-limited Huffman code construction.  */
+#define LZX_MAX_MAIN_CODEWORD_LEN       16
+#define LZX_MAX_LEN_CODEWORD_LEN        16
+#define LZX_MAX_PRE_CODEWORD_LEN        ((1 << LZX_PRECODE_ELEMENT_SIZE) - 1)
+#define LZX_MAX_ALIGNED_CODEWORD_LEN    ((1 << LZX_ALIGNEDCODE_ELEMENT_SIZE) - 1)
+
+/* For LZX-compressed blocks in WIM resources, this value is always used as the
+ * filesize parameter for the call instruction (0xe8 byte) preprocessing, even
+ * though the blocks themselves are not this size, and the size of the actual
+ * file resource in the WIM file is very likely to be something entirely
+ * different as well.  */
+#define LZX_WIM_MAGIC_FILESIZE          12000000
+
+/* Assumed LZX block size when the encoded block size begins with a 0 bit.
+ * This is probably WIM-specific.  */
+#define LZX_DEFAULT_BLOCK_SIZE          32768
+
+#define LZX_E8_FILTER_TAIL_SIZE         10
+
+/* Number of offsets in the recent (or "repeat") offsets queue.  */
+#define LZX_NUM_RECENT_OFFSETS          3
+
+/* An offset of n bytes is actually encoded as (n + LZX_OFFSET_ADJUSTMENT).  */
+#define LZX_OFFSET_ADJUSTMENT        (LZX_NUM_RECENT_OFFSETS - 1)
+
+#endif /* _LZX_CONSTANTS_H */
diff --git a/dlls/cabinet/liblzx_matchfinder_common.h b/dlls/cabinet/liblzx_matchfinder_common.h
new file mode 100644
index 00000000000..37f16fc3fa1
--- /dev/null
+++ b/dlls/cabinet/liblzx_matchfinder_common.h
@@ -0,0 +1,131 @@
+/*
+ * matchfinder_common.h - common code for Lempel-Ziv matchfinding
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _LIBLZX_MATCHFINDER_COMMON_H
+#define _LIBLZX_MATCHFINDER_COMMON_H
+
+#include "liblzx_bitops.h"
+#include "liblzx_unaligned.h"
+
+/*
+ * Given a 32-bit value that was loaded with the platform's native endianness,
+ * return a 32-bit value whose high-order 8 bits are 0 and whose low-order 24
+ * bits contain the first 3 bytes, arranged in octets in a platform-dependent
+ * order, at the memory location from which the input 32-bit value was loaded.
+ */
+static attrib_forceinline uint32_t
+loaded_u32_to_u24(uint32_t v)
+{
+        if (CPU_IS_LITTLE_ENDIAN())
+                return v & 0xFFFFFF;
+        else
+                return v >> 8;
+}
+
+/*
+ * Load the next 3 bytes from @p into the 24 low-order bits of a 32-bit value.
+ * The order in which the 3 bytes will be arranged as octets in the 24 bits is
+ * platform-dependent.  At least 4 bytes (not 3) must be available at @p.
+ */
+static attrib_forceinline uint32_t
+load_u24_unaligned(const uint8_t *p)
+{
+#if UNALIGNED_ACCESS_IS_FAST
+        return loaded_u32_to_u24(load_u32_unaligned(p));
+#else
+        if (CPU_IS_LITTLE_ENDIAN())
+                return ((uint32_t)p[0] << 0) | ((uint32_t)p[1] << 8) | ((uint32_t)p[2] << 16);
+        else
+                return ((uint32_t)p[2] << 0) | ((uint32_t)p[1] << 8) | ((uint32_t)p[0] << 16);
+#endif
+}
+
+/*
+ * The hash function: given a sequence prefix held in the low-order bits of a
+ * 32-bit value, multiply by a carefully-chosen large constant.  Discard any
+ * bits of the product that don't fit in a 32-bit value, but take the
+ * next-highest @num_bits bits of the product as the hash value, as those have
+ * the most randomness.
+ */
+static attrib_forceinline uint32_t
+lz_hash(uint32_t seq, unsigned num_bits)
+{
+        return (uint32_t)(seq * 0x1E35A7BD) >> (32 - num_bits);
+}
+
+/*
+ * Return the number of bytes at @matchptr that match the bytes at @strptr, up
+ * to a maximum of @max_len.  Initially, @start_len bytes are matched.
+ */
+static attrib_forceinline unsigned
+lz_extend(const uint8_t * const strptr, const uint8_t * const matchptr,
+          const unsigned start_len, const unsigned max_len)
+{
+        unsigned len = start_len;
+        machine_word_t v_word;
+
+        if (UNALIGNED_ACCESS_IS_FAST) {
+
+                if (likely(max_len - len >= 4 * WORDBYTES)) {
+
+                #define COMPARE_WORD_STEP                               \
+                        v_word = load_word_unaligned(&matchptr[len]) ^  \
+                                 load_word_unaligned(&strptr[len]);     \
+                        if (v_word != 0)                                \
+                                goto word_differs;                      \
+                        len += WORDBYTES;                               \
+
+                        COMPARE_WORD_STEP
+                        COMPARE_WORD_STEP
+                        COMPARE_WORD_STEP
+                        COMPARE_WORD_STEP
+                #undef COMPARE_WORD_STEP
+                }
+
+                while (len + WORDBYTES <= max_len) {
+                        v_word = load_word_unaligned(&matchptr[len]) ^
+                                 load_word_unaligned(&strptr[len]);
+                        if (v_word != 0)
+                                goto word_differs;
+                        len += WORDBYTES;
+                }
+        }
+
+        while (len < max_len && matchptr[len] == strptr[len])
+                len++;
+        return len;
+
+word_differs:
+        if (CPU_IS_LITTLE_ENDIAN())
+                len += (bsfw(v_word) >> 3);
+        else
+                len += (WORDBITS - 1 - bsrw(v_word)) >> 3;
+        return len;
+}
+
+#endif /* _LIBLZX_MATCHFINDER_COMMON_H */
diff --git a/dlls/cabinet/liblzx_minmax.h b/dlls/cabinet/liblzx_minmax.h
new file mode 100644
index 00000000000..d4d76130caa
--- /dev/null
+++ b/dlls/cabinet/liblzx_minmax.h
@@ -0,0 +1,122 @@
+/*
+ * compiler.h
+ *
+ * Compiler-specific definitions.
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#ifndef _LIBLZX_MINMAX_H
+#define _LIBLZX_MINMAX_H
+
+#include "liblzx_compiler.h"
+#include "liblzx_types.h"
+
+/* Get the minimum of two variables, without multiple evaluation.  */
+static attrib_forceinline double
+min_double(double a, double b)
+{
+        return (a < b) ? a : b;
+}
+
+static attrib_forceinline float
+min_float(float a, float b)
+{
+        return (a < b) ? a : b;
+}
+
+static attrib_forceinline unsigned
+min_uint(unsigned a, unsigned b)
+{
+        return (a < b) ? a : b;
+}
+
+static attrib_forceinline unsigned
+min_u32(unsigned a, unsigned b)
+{
+        return (a < b) ? a : b;
+}
+
+static attrib_forceinline size_t
+min_size(size_t a, size_t b)
+{
+        return (a < b) ? a : b;
+}
+
+static attrib_forceinline intmax_t
+min_int(int a, int b)
+{
+        return (a < b) ? a : b;
+}
+
+static attrib_forceinline void *
+min_ptr(void *a, void *b)
+{
+        return (a < b) ? a : b;
+}
+
+static attrib_forceinline const void *
+min_constptr(const void *a, const void *b)
+{
+        return (a < b) ? a : b;
+}
+
+/* Get the maximum of two variables, without multiple evaluation.  */
+static attrib_forceinline double
+max_float(double a, double b)
+{
+        return (a > b) ? a : b;
+}
+
+static attrib_forceinline unsigned
+max_uint(unsigned a, unsigned b)
+{
+        return (a > b) ? a : b;
+}
+
+static attrib_forceinline uint32_t
+max_u32(uint32_t a, uint32_t b)
+{
+        return (a > b) ? a : b;
+}
+
+static attrib_forceinline uint64_t
+max_u64(uint64_t a, uint64_t b)
+{
+        return (a > b) ? a : b;
+}
+
+static attrib_forceinline void *
+max_ptr(void *a, void *b)
+{
+        return (a > b) ? a : b;
+}
+
+static attrib_forceinline const void *
+max_constptr(const void *a, const void *b)
+{
+        return (a > b) ? a : b;
+}
+
+#endif
diff --git a/dlls/cabinet/liblzx_types.h b/dlls/cabinet/liblzx_types.h
new file mode 100644
index 00000000000..b63e4b31fc1
--- /dev/null
+++ b/dlls/cabinet/liblzx_types.h
@@ -0,0 +1,33 @@
+#ifndef _LIBLZX_TYPES_H
+#define _LIBLZX_TYPES_H
+
+#include <inttypes.h>
+#include <stdbool.h>
+#include <stddef.h>
+
+#include "liblzx_compiler.h"
+
+/* Unsigned little endian types of exact size */
+typedef uint16_t _bitwise_attr le16_t;
+typedef uint32_t _bitwise_attr le32_t;
+typedef uint64_t _bitwise_attr le64_t;
+
+/* Unsigned big endian types of exact size */
+typedef uint16_t _bitwise_attr be16_t;
+typedef uint32_t _bitwise_attr be32_t;
+typedef uint64_t _bitwise_attr be64_t;
+
+/* A pointer to 'utf16lechar' indicates a UTF-16LE encoded string */
+typedef le16_t utf16lechar;
+
+/*
+ * Type of a machine word.  'unsigned long' would be logical, but that is only
+ * 32 bits on x86_64 Windows.  The same applies to 'uint_fast32_t'.  So the best
+ * we can do without a bunch of #ifdefs appears to be 'size_t'.
+ */
+typedef size_t machine_word_t;
+
+#define WORDBYTES        sizeof(machine_word_t)
+#define WORDBITS         (8 * WORDBYTES)
+
+#endif /* _LIBLZX_TYPES_H */
diff --git a/dlls/cabinet/liblzx_unaligned.h b/dlls/cabinet/liblzx_unaligned.h
new file mode 100644
index 00000000000..23ee938fb62
--- /dev/null
+++ b/dlls/cabinet/liblzx_unaligned.h
@@ -0,0 +1,134 @@
+/*
+ * unaligned.h - inline functions for unaligned memory accesses
+ *
+ * Copyright (C) 2025 Eric Lasota
+ * Based on wimlib.  Copyright 2022 Eric Biggers
+ *
+ * Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _LIBLZX_UNALIGNED_H
+#define _LIBLZX_UNALIGNED_H
+
+#include <string.h>
+
+#include "liblzx_compiler.h"
+#include "liblzx_endianness.h"
+#include "liblzx_types.h"
+
+#define DEFINE_UNALIGNED_TYPE(name, type)                        \
+static attrib_forceinline type                                   \
+load_##name##_unaligned(const void *p)                           \
+{                                                                \
+        type v;                                                  \
+        memcpy(&v, p, sizeof(v));                                \
+        return v;                                                \
+}                                                                \
+                                                                 \
+static attrib_forceinline void                                   \
+store_##name##_unaligned(type v, void *p)                        \
+{                                                                \
+        memcpy(p, &v, sizeof(v));                                \
+}
+
+DEFINE_UNALIGNED_TYPE(u16, uint16_t);
+DEFINE_UNALIGNED_TYPE(u32, uint32_t);
+DEFINE_UNALIGNED_TYPE(u64, uint64_t);
+DEFINE_UNALIGNED_TYPE(le16, le16_t);
+DEFINE_UNALIGNED_TYPE(le32, le32_t);
+DEFINE_UNALIGNED_TYPE(le64, le64_t);
+DEFINE_UNALIGNED_TYPE(be16, be16_t);
+DEFINE_UNALIGNED_TYPE(be32, be32_t);
+DEFINE_UNALIGNED_TYPE(be64, be64_t);
+DEFINE_UNALIGNED_TYPE(size_t, size_t);
+DEFINE_UNALIGNED_TYPE(machine_word_t, machine_word_t);
+
+#define load_word_unaligned        load_machine_word_t_unaligned
+#define store_word_unaligned        store_machine_word_t_unaligned
+
+static attrib_forceinline uint16_t
+get_unaligned_le16(const uint8_t *p)
+{
+        if (UNALIGNED_ACCESS_IS_FAST)
+                return le16_to_cpu(load_le16_unaligned(p));
+        else
+                return ((uint16_t)p[1] << 8) | p[0];
+}
+
+static attrib_forceinline uint32_t
+get_unaligned_le32(const uint8_t *p)
+{
+        if (UNALIGNED_ACCESS_IS_FAST)
+                return le32_to_cpu(load_le32_unaligned(p));
+        else
+                return ((uint32_t)p[3] << 24) | ((uint32_t)p[2] << 16) |
+                        ((uint32_t)p[1] << 8) | p[0];
+}
+
+static attrib_forceinline uint32_t
+get_unaligned_be32(const uint8_t *p)
+{
+        if (UNALIGNED_ACCESS_IS_FAST)
+                return be32_to_cpu(load_be32_unaligned(p));
+        else
+                return ((uint32_t)p[0] << 24) | ((uint32_t)p[1] << 16) |
+                        ((uint32_t)p[2] << 8) | p[3];
+}
+
+static attrib_forceinline void
+put_unaligned_le16(uint16_t v, uint8_t *p)
+{
+        if (UNALIGNED_ACCESS_IS_FAST) {
+                store_le16_unaligned(cpu_to_le16(v), p);
+        } else {
+                p[0] = (uint8_t)(v >> 0);
+                p[1] = (uint8_t)(v >> 8);
+        }
+}
+
+static attrib_forceinline void
+put_unaligned_le32(uint32_t v, uint8_t *p)
+{
+        if (UNALIGNED_ACCESS_IS_FAST) {
+                store_le32_unaligned(cpu_to_le32(v), p);
+        } else {
+                p[0] = (uint8_t)(v >> 0);
+                p[1] = (uint8_t)(v >> 8);
+                p[2] = (uint8_t)(v >> 16);
+                p[3] = (uint8_t)(v >> 24);
+        }
+}
+
+static attrib_forceinline void
+put_unaligned_be32(uint32_t v, uint8_t *p)
+{
+        if (UNALIGNED_ACCESS_IS_FAST) {
+                store_be32_unaligned(cpu_to_be32(v), p);
+        } else {
+                p[0] = (uint8_t)(v >> 24);
+                p[1] = (uint8_t)(v >> 16);
+                p[2] = (uint8_t)(v >> 8);
+                p[3] = (uint8_t)(v >> 0);
+        }
+}
+
+#endif /* _LIBLZX_UNALIGNED_H */
diff --git a/dlls/cabinet/liblzx_util.h b/dlls/cabinet/liblzx_util.h
new file mode 100644
index 00000000000..4d832ae0da6
--- /dev/null
+++ b/dlls/cabinet/liblzx_util.h
@@ -0,0 +1,20 @@
+/*
+ * util.h - utility functions and macros
+ */
+#ifndef _LIBLZX_UTIL_H
+#define _LIBLZX_UTIL_H
+
+#include "liblzx_compiler.h"
+#include "liblzx_types.h"
+
+/****************
+ * General macros
+ *****************/
+
+/* Calculate 'n / d', but round up instead of down.  */
+#define DIV_ROUND_UP(n, d)        (((n) + (d) - 1) / (d))
+
+/* Get the number of elements of an array type.  */
+#define ARRAY_LEN(array)        (sizeof(array) / sizeof((array)[0]))
+
+#endif /* _LIBLZX_UTIL_H */
diff --git a/dlls/cabinet/tests/extract.c b/dlls/cabinet/tests/extract.c
index 2e37dc7cda6..748115d67fd 100644
--- a/dlls/cabinet/tests/extract.c
+++ b/dlls/cabinet/tests/extract.c
@@ -104,9 +104,11 @@ static void create_test_files(void)

createTestFile("a.txt");
     createTestFile("b.txt");
+    createTestFile("c.txt");
+    createTestFile("d.txt");
     CreateDirectoryA("testdir", NULL);
-    createTestFile("testdir\c.txt");
-    createTestFile("testdir\d.txt");
+    createTestFile("testdir\e.txt");
+    createTestFile("testdir\f.txt");
     CreateDirectoryA("dest", NULL);
 }

@@ -114,11 +116,13 @@ static void delete_test_files(void)
 {
     DeleteFileA("a.txt");
     DeleteFileA("b.txt");
-    DeleteFileA("testdir\c.txt");
-    DeleteFileA("testdir\d.txt");
+    DeleteFileA("c.txt");
+    DeleteFileA("d.txt");
+    DeleteFileA("testdir\e.txt");
+    DeleteFileA("testdir\f.txt");
     RemoveDirectoryA("testdir");

-    DeleteFileA("extract.cab");
+    //DeleteFileA("extract.cab");
 }

/* the FCI callbacks */
@@ -269,7 +273,7 @@ static INT_PTR CDECL get_open_info(char *pszName, USHORT *pdate, USHORT *ptime,
     return (INT_PTR)handle;
 }

-static void add_file(HFCI hfci, char *file)
+static void add_file(HFCI hfci, char *file, TCOMP typeCompress)
 {
     char path[MAX_PATH];
     BOOL res;
@@ -279,7 +283,7 @@ static void add_file(HFCI hfci, char *file)
     lstrcatA(path, file);

res = FCIAddFile(hfci, path, file, FALSE, get_next_cabinet, progress,
-                     get_open_info, tcompTYPE_MSZIP);
+                     get_open_info, typeCompress);
     ok(res, "Expected FCIAddFile to succeed\n");
 }

@@ -300,10 +304,12 @@ static void create_cab_file(void)
     CCAB cabParams;
     HFCI hfci;
     ERF erf;
-    static CHAR a_txt[]         = "a.txt",
-                b_txt[]         = "b.txt",
-                testdir_c_txt[] = "testdir\c.txt",
-                testdir_d_txt[] = "testdir\d.txt";
+    static CHAR a_txt[]           = "a.txt",
+                b_txt[]           = "b.txt",
+                c_txt[]           = "c.txt",
+                d_txt[]           = "d.txt",
+                testdir_e_txt[]   = "testdir\e.txt",
+                testdir_f_txt[]   = "testdir\f.txt";
     BOOL res;

set_cab_parameters(&cabParams);
@@ -314,10 +320,12 @@ static void create_cab_file(void)

ok(hfci != NULL, "Failed to create an FCI context\n");

-    add_file(hfci, a_txt);
-    add_file(hfci, b_txt);
-    add_file(hfci, testdir_c_txt);
-    add_file(hfci, testdir_d_txt);
+    add_file(hfci, a_txt, tcompTYPE_MSZIP);
+    add_file(hfci, b_txt, tcompTYPE_MSZIP);
+    add_file(hfci, c_txt, TCOMPfromLZXWindow(21));
+    add_file(hfci, d_txt, TCOMPfromLZXWindow(21));
+    add_file(hfci, testdir_e_txt, tcompTYPE_MSZIP);
+    add_file(hfci, testdir_f_txt, tcompTYPE_MSZIP);

res = FCIFlushCabinet(hfci, FALSE, get_next_cabinet, progress);
     ok(res, "Failed to flush the cabinet\n");
@@ -380,26 +388,30 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 4, "Expected 4, got %d\n", session.FileCount);
+    ok(session.FileCount == 6, "Expected 6, got %d\n", session.FileCount);
     ok(session.Operation == (EXTRACT_FILLFILELIST | EXTRACT_EXTRACTFILES),
        "Expected EXTRACT_FILLFILELIST | EXTRACT_EXTRACTFILES, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(DeleteFileA("dest\a.txt"), "Expected dest\a.txt to exist\n");
     ok(DeleteFileA("dest\b.txt"), "Expected dest\b.txt to exist\n");
-    ok(DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to exist\n");
-    ok(DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to exist\n");
+    ok(DeleteFileA("dest\c.txt"), "Expected dest\c.txt to exist\n");
+    ok(DeleteFileA("dest\d.txt"), "Expected dest\d.txt to exist\n");
+    ok(DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to exist\n");
+    ok(DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to exist\n");
     ok(RemoveDirectoryA("dest\testdir"), "Expected dest\testdir to exist\n");
-    ok(check_list(&node, "testdir\d.txt", FALSE), "list entry wrong\n");
-    ok(check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\f.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "c.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "a.txt", FALSE), "list entry wrong\n");
     free_file_list(&session);
@@ -411,23 +423,25 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 4, "Expected 4, got %d\n", session.FileCount);
+    ok(session.FileCount == 6, "Expected 6, got %d\n", session.FileCount);
     ok(session.Operation == EXTRACT_FILLFILELIST,
        "Expected EXTRACT_FILLFILELIST, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(!DeleteFileA("dest\a.txt"), "Expected dest\a.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to not exist\n");
-    ok(check_list(&node, "testdir\d.txt", TRUE), "list entry wrong\n");
-    ok(check_list(&node, "testdir\c.txt", TRUE), "list entry wrong\n");
+    ok(!DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to not exist\n");
+    ok(check_list(&node, "testdir\f.txt", TRUE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\e.txt", TRUE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", TRUE), "list entry wrong\n");
+    ok(check_list(&node, "c.txt", TRUE), "list entry wrong\n");
     ok(check_list(&node, "b.txt", TRUE), "list entry wrong\n");
     ok(check_list(&node, "a.txt", TRUE), "list entry wrong\n");

@@ -436,27 +450,31 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 4, "Expected 4, got %d\n", session.FileCount);
+    ok(session.FileCount == 6, "Expected 6, got %d\n", session.FileCount);
     ok(session.Operation == EXTRACT_EXTRACTFILES,
        "Expected EXTRACT_EXTRACTFILES, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(DeleteFileA("dest\a.txt"), "Expected dest\a.txt to exist\n");
     ok(DeleteFileA("dest\b.txt"), "Expected dest\b.txt to exist\n");
-    ok(DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to exist\n");
-    ok(DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to exist\n");
+    ok(DeleteFileA("dest\c.txt"), "Expected dest\c.txt to exist\n");
+    ok(DeleteFileA("dest\d.txt"), "Expected dest\d.txt to exist\n");
+    ok(DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to exist\n");
+    ok(DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to exist\n");
     ok(RemoveDirectoryA("dest\testdir"), "Expected dest\testdir to exist\n");
     ok(RemoveDirectoryA("dest"), "Expected dest to exist\n");
-    ok(check_list(&node, "testdir\d.txt", FALSE), "list entry wrong\n");
-    ok(check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\f.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "c.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "a.txt", FALSE), "list entry wrong\n");

@@ -464,25 +482,29 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 4, "Expected 4, got %d\n", session.FileCount);
+    ok(session.FileCount == 6, "Expected 6, got %d\n", session.FileCount);
     ok(session.Operation == EXTRACT_EXTRACTFILES,
        "Expected EXTRACT_EXTRACTFILES, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(!DeleteFileA("dest\a.txt"), "Expected dest\a.txt to not exist\n");
     ok(!DeleteFileA("dest\b.txt"), "Expected dest\b.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to not exist\n");
-    ok(check_list(&node, "testdir\d.txt", FALSE), "list entry wrong\n");
-    ok(check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
+    ok(!DeleteFileA("dest\c.txt"), "Expected dest\a.txt to not exist\n");
+    ok(!DeleteFileA("dest\d.txt"), "Expected dest\b.txt to not exist\n");
+    ok(!DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to not exist\n");
+    ok(!DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to not exist\n");
+    ok(check_list(&node, "testdir\f.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "c.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "a.txt", FALSE), "list entry wrong\n");

@@ -497,26 +519,30 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 4, "Expected 4, got %d\n", session.FileCount);
+    ok(session.FileCount == 6, "Expected 6, got %d\n", session.FileCount);
     ok(session.Operation == EXTRACT_EXTRACTFILES,
        "Expected EXTRACT_EXTRACTFILES, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(DeleteFileA("dest\a.txt"), "Expected dest\a.txt to exist\n");
-    ok(DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to exist\n");
-    ok(!DeleteFileA("dest\b.txt"), "Expected dest\b.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to not exist\n");
-    ok(check_list(&node, "testdir\d.txt", FALSE), "list entry wrong\n");
-    ok(!check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
-    ok(check_list(&node, "b.txt", FALSE), "list entry wrong\n");
+    ok(DeleteFileA("dest\b.txt"), "Expected dest\b.txt to exist\n");
+    ok(DeleteFileA("dest\c.txt"), "Expected dest\c.txt to exist\n");
+    ok(DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to exist\n");
+    ok(!DeleteFileA("dest\d.txt"), "Expected dest\d.txt to not exist\n");
+    ok(!DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to not exist\n");
+    ok(check_list(&node, "testdir\f.txt", FALSE), "list entry wrong\n");
+    ok(!check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", FALSE), "list entry wrong\n");
+    ok(!check_list(&node, "c.txt", FALSE), "list entry wrong\n");
+    ok(!check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(!check_list(&node, "a.txt", FALSE), "list entry wrong\n");
     free_file_list(&session);

@@ -525,25 +551,25 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 8, "Expected 8, got %d\n", session.FileCount);
+    ok(session.FileCount == 12, "Expected 12, got %d\n", session.FileCount);
     ok(session.Operation == EXTRACT_FILLFILELIST,
        "Expected EXTRACT_FILLFILELIST, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(!DeleteFileA("dest\a.txt"), "Expected dest\a.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to not exist\n");
+    ok(!DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to not exist\n");
     ok(!DeleteFileA("dest\b.txt"), "Expected dest\b.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to not exist\n");
-    ok(check_list(&node, "testdir\d.txt", TRUE), "list entry wrong\n");
-    ok(!check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
+    ok(!DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to not exist\n");
+    ok(check_list(&node, "testdir\f.txt", TRUE), "list entry wrong\n");
+    ok(!check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
     ok(!check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(!check_list(&node, "a.txt", FALSE), "list entry wrong\n");

@@ -551,24 +577,26 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 8, "Expected 8, got %d\n", session.FileCount);
+    ok(session.FileCount == 12, "Expected 12, got %d\n", session.FileCount);
     ok(session.Operation == 0, "Expected 0, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(!DeleteFileA("dest\a.txt"), "Expected dest\a.txt to exist\n");
-    ok(!DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to exist\n");
+    ok(!DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to exist\n");
     ok(!DeleteFileA("dest\b.txt"), "Expected dest\b.txt to exist\n");
-    ok(!DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to exist\n");
-    ok(check_list(&node, "testdir\d.txt", TRUE), "list entry wrong\n");
-    ok(check_list(&node, "testdir\c.txt", TRUE), "list entry wrong\n");
+    ok(!DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to exist\n");
+    ok(check_list(&node, "testdir\f.txt", TRUE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\e.txt", TRUE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", TRUE), "list entry wrong\n");
+    ok(check_list(&node, "c.txt", TRUE), "list entry wrong\n");
     ok(check_list(&node, "b.txt", TRUE), "list entry wrong\n");
     ok(check_list(&node, "a.txt", TRUE), "list entry wrong\n");

@@ -577,28 +605,32 @@ static void test_Extract(void)
     res = pExtract(&session, "extract.cab");
     node = session.FileList;
     ok(res == S_OK, "Expected S_OK, got %ld\n", res);
-    ok(session.FileSize == 40, "Expected 40, got %d\n", session.FileSize);
+    ok(session.FileSize == 52, "Expected 52, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_NONE,
        "Expected FDIERROR_NONE, got %d\n", session.Error.erfOper);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Error.fError == FALSE, "Expected FALSE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 8, "Expected 8, got %d\n", session.FileCount);
+    ok(session.FileCount == 12, "Expected 12, got %d\n", session.FileCount);
     ok(session.Operation == 0, "Expected 0, got %d\n", session.Operation);
     ok(!lstrcmpA(session.Destination, "dest"), "Expected dest, got %s\n", session.Destination);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\d.txt"),
-       "Expected dest\testdir\d.txt, got %s\n", session.CurrentFile);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\f.txt"),
+       "Expected dest\testdir\f.txt, got %s\n", session.CurrentFile);
     ok(!*session.Reserved, "Expected empty string, got %s\n", session.Reserved);
     ok(DeleteFileA("dest\a.txt"), "Expected dest\a.txt to exist\n");
-    ok(DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to exist\n");
+    ok(DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to exist\n");
     ok(DeleteFileA("dest\b.txt"), "Expected dest\b.txt to exist\n");
-    ok(DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to exist\n");
-    ok(check_list(&node, "testdir\d.txt", FALSE), "list entry wrong\n");
-    ok(check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
+    ok(DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to exist\n");
+    ok(check_list(&node, "testdir\f.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "c.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "a.txt", FALSE), "list entry wrong\n");
     node = session.FilterList;
-    ok(check_list(&node, "testdir\d.txt", FALSE), "list entry wrong\n");
-    ok(check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\f.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "d.txt", FALSE), "list entry wrong\n");
+    ok(check_list(&node, "c.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(check_list(&node, "a.txt", FALSE), "list entry wrong\n");
     free_file_list(&session);
@@ -625,10 +657,10 @@ static void test_Extract(void)
     ok(!session.FilterList, "Expected empty filter list\n");
     ok(!DeleteFileA("dest\a.txt"), "Expected dest\a.txt to not exist\n");
     ok(!DeleteFileA("dest\b.txt"), "Expected dest\b.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to not exist\n");
-    ok(!check_list(&node, "testdir\d.txt", FALSE), "list entry should not exist\n");
-    ok(!check_list(&node, "testdir\c.txt", FALSE), "list entry should not exist\n");
+    ok(!DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to not exist\n");
+    ok(!DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to not exist\n");
+    ok(!check_list(&node, "testdir\f.txt", FALSE), "list entry should not exist\n");
+    ok(!check_list(&node, "testdir\e.txt", FALSE), "list entry should not exist\n");
     ok(!check_list(&node, "b.txt", FALSE), "list entry should not exist\n");
     ok(!check_list(&node, "a.txt", FALSE), "list entry should not exist\n");
     free_file_list(&session);
@@ -660,10 +692,10 @@ static void test_Extract(void)
     ok(getFileSize("dest\a.txt") == 11, "Expected dest\a.txt to be 11 bytes\n");
     ok(!DeleteFileA("dest\a.txt"), "Expected dest\a.txt to be read-only\n");
     ok(!DeleteFileA("dest\b.txt"), "Expected dest\b.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to not exist\n");
-    ok(!DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to not exist\n");
-    ok(!check_list(&node, "testdir\d.txt", FALSE), "list entry should not exist\n");
-    ok(!check_list(&node, "testdir\c.txt", FALSE), "list entry should not exist\n");
+    ok(!DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to not exist\n");
+    ok(!DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to not exist\n");
+    ok(!check_list(&node, "testdir\f.txt", FALSE), "list entry should not exist\n");
+    ok(!check_list(&node, "testdir\e.txt", FALSE), "list entry should not exist\n");
     ok(!check_list(&node, "b.txt", FALSE), "list entry should not exist\n");
     ok(!check_list(&node, "a.txt", FALSE), "list entry should not exist\n");
     free_file_list(&session);
@@ -673,8 +705,8 @@ static void test_Extract(void)

/* first file exists and is writable, third file exists but is read-only */
     createTestFile("dest\a.txt");
-    createTestFile("dest\testdir\c.txt");
-    SetFileAttributesA("dest\testdir\c.txt", FILE_ATTRIBUTE_READONLY);
+    createTestFile("dest\testdir\e.txt");
+    SetFileAttributesA("dest\testdir\e.txt", FILE_ATTRIBUTE_READONLY);
     ZeroMemory(&session, sizeof(SESSION));
     lstrcpyA(session.Destination, "dest");
     session.Operation = EXTRACT_FILLFILELIST | EXTRACT_EXTRACTFILES;
@@ -682,12 +714,12 @@ static void test_Extract(void)
     node = session.FileList;
     ok(res == HRESULT_FROM_WIN32(ERROR_ACCESS_DENIED) || res == E_FAIL,
        "Expected HRESULT_FROM_WIN32(ERROR_ACCESS_DENIED) or E_FAIL, got %08lx\n", res);
-    ok(session.FileSize == 26, "Expected 26, got %d\n", session.FileSize);
+    ok(session.FileSize == 38, "Expected 38, got %d\n", session.FileSize);
     ok(session.Error.erfOper == FDIERROR_USER_ABORT,
        "Expected FDIERROR_USER_ABORT, got %d\n", session.Error.erfOper);
     ok(session.Error.fError == TRUE, "Expected TRUE, got %d\n", session.Error.fError);
-    ok(session.FileCount == 3, "Expected 3, got %d\n", session.FileCount);
-    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\c.txt"),
+    ok(session.FileCount == 5, "Expected 5, got %d\n", session.FileCount);
+    ok(!lstrcmpA(session.CurrentFile, "dest\testdir\e.txt"),
        "Expected dest\c.txt, got %s\n", session.CurrentFile);
     ok(session.Error.erfType == 0, "Expected 0, got %d\n", session.Error.erfType);
     ok(session.Operation == (EXTRACT_FILLFILELIST | EXTRACT_EXTRACTFILES),
@@ -698,19 +730,21 @@ static void test_Extract(void)
     ok(getFileSize("dest\a.txt") == 6, "Expected dest\a.txt to be 6 bytes\n");
     ok(DeleteFileA("dest\a.txt"), "Expected dest\a.txt to exist\n");
     ok(DeleteFileA("dest\b.txt"), "Expected dest\b.txt to exist\n");
-    ok(!DeleteFileA("dest\testdir\c.txt"), "Expected dest\testdir\c.txt to be read-only\n");
-    ok(!DeleteFileA("dest\testdir\d.txt"), "Expected dest\testdir\d.txt to not exist\n");
-    ok(!check_list(&node, "testdir\d.txt", FALSE), "list entry should not exist\n");
-    ok(!check_list(&node, "testdir\c.txt", FALSE), "list entry wrong\n");
+    ok(DeleteFileA("dest\c.txt"), "Expected dest\a.txt to exist\n");
+    ok(DeleteFileA("dest\d.txt"), "Expected dest\b.txt to exist\n");
+    ok(!DeleteFileA("dest\testdir\e.txt"), "Expected dest\testdir\e.txt to be read-only\n");
+    ok(!DeleteFileA("dest\testdir\f.txt"), "Expected dest\testdir\f.txt to not exist\n");
+    ok(!check_list(&node, "testdir\f.txt", FALSE), "list entry should not exist\n");
+    ok(!check_list(&node, "testdir\e.txt", FALSE), "list entry wrong\n");
     ok(!check_list(&node, "b.txt", FALSE), "list entry wrong\n");
     ok(!check_list(&node, "a.txt", TRUE), "list entry wrong\n");
     free_file_list(&session);

-    SetFileAttributesA("dest\testdir\c.txt", FILE_ATTRIBUTE_NORMAL);
-    DeleteFileA("dest\testdir\c.txt");
+    SetFileAttributesA("dest\testdir\e.txt", FILE_ATTRIBUTE_NORMAL);
+    DeleteFileA("dest\testdir\e.txt");

-    ok(RemoveDirectoryA("dest\testdir"), "Expected dest\testdir to exist\n");
-    ok(RemoveDirectoryA("dest"), "Expected dest to exist\n");
+    ok(RemoveDirectoryA("dest\testdir"), "Expected dest\testdir to exist and be empty\n");
+    ok(RemoveDirectoryA("dest"), "Expected dest to exist and be empty\n");
 }

START_TEST(extract)