decompressors: add XZ decompressor module

In userspace, the .lzma format has become mostly a legacy file format that got superseded by the .xz format. Similarly, LZMA Utils was superseded by XZ Utils. These patches add support for XZ decompression into the kernel. Most of the code is as is from XZ Embedded <http://tukaani.org/xz/embedded.html>. It was written for the Linux kernel but is usable in other projects too. Advantages of XZ over the current LZMA code in the kernel: - Nice API that can be used by other kernel modules; it's not limited to kernel, initramfs, and initrd decompression. - Integrity check support (CRC32) - BCJ filters improve compression of executable code on certain architectures. These together with LZMA2 can produce a few percent smaller kernel or Squashfs images than plain LZMA without making the decompression slower. This patch: Add the main decompression code (xz_dec), testing module (xz_dec_test), wrapper script (xz_wrap.sh) for the xz command line tool, and documentation. The xz_dec module is enough to have a usable XZ decompressor e.g. for Squashfs. Signed-off-by: Lasse Collin <lasse.collin@tukaani.org> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Alain Knaff <alain@knaff.lu> Cc: Albin Tonnerre <albin.tonnerre@free-electrons.com> Cc: Phillip Lougher <phillip@lougher.demon.co.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

decompressors: add XZ decompressor module
In userspace, the .lzma format has become mostly a legacy file format that got superseded by the .xz format. Similarly, LZMA Utils was superseded by XZ Utils. These patches add support for XZ decompression into the kernel. Most of the code is as is from XZ Embedded <http://tukaani.org/xz/embedded.html>. It was written for the Linux kernel but is usable in other projects too. Advantages of XZ over the current LZMA code in the kernel: - Nice API that can be used by other kernel modules; it's not limited to kernel, initramfs, and initrd decompression. - Integrity check support (CRC32) - BCJ filters improve compression of executable code on certain architectures. These together with LZMA2 can produce a few percent smaller kernel or Squashfs images than plain LZMA without making the decompression slower. This patch: Add the main decompression code (xz_dec), testing module (xz_dec_test), wrapper script (xz_wrap.sh) for the xz command line tool, and documentation. The xz_dec module is enough to have a usable XZ decompressor e.g. for Squashfs. Signed-off-by: Lasse Collin <lasse.collin@tukaani.org> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Alain Knaff <alain@knaff.lu> Cc: Albin Tonnerre <albin.tonnerre@free-electrons.com> Cc: Phillip Lougher <phillip@lougher.demon.co.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Lasse Collin · Linus Torvalds
1 parent fb7fa589fd
Showing 17 changed files with 3783 additions and 0 deletions Side-by-side Diff
Documentation/xz.txt
include/linux/xz.h
lib/Kconfig
lib/Makefile
lib/xz/Kconfig
lib/xz/Makefile
lib/xz/xz_crc32.c
lib/xz/xz_dec_bcj.c
lib/xz/xz_dec_lzma2.c
lib/xz/xz_dec_stream.c
lib/xz/xz_dec_syms.c
lib/xz/xz_dec_test.c
lib/xz/xz_lzma2.h
lib/xz/xz_private.h
lib/xz/xz_stream.h
scripts/Makefile.lib
scripts/xz_wrap.sh
+
+XZ data compression in Linux
+============================
+
+Introduction
+
+    XZ is a general purpose data compression format with high compression
+    ratio and relatively fast decompression. The primary compression
+    algorithm (filter) is LZMA2. Additional filters can be used to improve
+    compression ratio even further. E.g. Branch/Call/Jump (BCJ) filters
+    improve compression ratio of executable data.
+
+    The XZ decompressor in Linux is called XZ Embedded. It supports
+    the LZMA2 filter and optionally also BCJ filters. CRC32 is supported
+    for integrity checking. The home page of XZ Embedded is at
+    <http://tukaani.org/xz/embedded.html>, where you can find the
+    latest version and also information about using the code outside
+    the Linux kernel.
+
+    For userspace, XZ Utils provide a zlib-like compression library
+    and a gzip-like command line tool. XZ Utils can be downloaded from
+    <http://tukaani.org/xz/>.
+
+XZ related components in the kernel
+
+    The xz_dec module provides XZ decompressor with single-call (buffer
+    to buffer) and multi-call (stateful) APIs. The usage of the xz_dec
+    module is documented in include/linux/xz.h.
+
+    The xz_dec_test module is for testing xz_dec. xz_dec_test is not
+    useful unless you are hacking the XZ decompressor. xz_dec_test
+    allocates a char device major dynamically to which one can write
+    .xz files from userspace. The decompressed output is thrown away.
+    Keep an eye on dmesg to see diagnostics printed by xz_dec_test.
+    See the xz_dec_test source code for the details.
+
+    For decompressing the kernel image, initramfs, and initrd, there
+    is a wrapper function in lib/decompress_unxz.c. Its API is the
+    same as in other decompress_*.c files, which is defined in
+    include/linux/decompress/generic.h.
+
+    scripts/xz_wrap.sh is a wrapper for the xz command line tool found
+    from XZ Utils. The wrapper sets compression options to values suitable
+    for compressing the kernel image.
+
+    For kernel makefiles, two commands are provided for use with
+    $(call if_needed). The kernel image should be compressed with
+    $(call if_needed,xzkern) which will use a BCJ filter and a big LZMA2
+    dictionary. It will also append a four-byte trailer containing the
+    uncompressed size of the file, which is needed by the boot code.
+    Other things should be compressed with $(call if_needed,xzmisc)
+    which will use no BCJ filter and 1 MiB LZMA2 dictionary.
+
+Notes on compression options
+
+    Since the XZ Embedded supports only streams with no integrity check or
+    CRC32, make sure that you don't use some other integrity check type
+    when encoding files that are supposed to be decoded by the kernel. With
+    liblzma, you need to use either LZMA_CHECK_NONE or LZMA_CHECK_CRC32
+    when encoding. With the xz command line tool, use --check=none or
+    --check=crc32.
+
+    Using CRC32 is strongly recommended unless there is some other layer
+    which will verify the integrity of the uncompressed data anyway.
+    Double checking the integrity would probably be waste of CPU cycles.
+    Note that the headers will always have a CRC32 which will be validated
+    by the decoder; you can only change the integrity check type (or
+    disable it) for the actual uncompressed data.
+
+    In userspace, LZMA2 is typically used with dictionary sizes of several
+    megabytes. The decoder needs to have the dictionary in RAM, thus big
+    dictionaries cannot be used for files that are intended to be decoded
+    by the kernel. 1 MiB is probably the maximum reasonable dictionary
+    size for in-kernel use (maybe more is OK for initramfs). The presets
+    in XZ Utils may not be optimal when creating files for the kernel,
+    so don't hesitate to use custom settings. Example:
+
+        xz --check=crc32 --lzma2=dict=512KiB inputfile
+
+    An exception to above dictionary size limitation is when the decoder
+    is used in single-call mode. Decompressing the kernel itself is an
+    example of this situation. In single-call mode, the memory usage
+    doesn't depend on the dictionary size, and it is perfectly fine to
+    use a big dictionary: for maximum compression, the dictionary should
+    be at least as big as the uncompressed data itself.
+
+Future plans
+
+    Creating a limited XZ encoder may be considered if people think it is
+    useful. LZMA2 is slower to compress than e.g. Deflate or LZO even at
+    the fastest settings, so it isn't clear if LZMA2 encoder is wanted
+    into the kernel.
+
+    Support for limited random-access reading is planned for the
+    decompression code. I don't know if it could have any use in the
+    kernel, but I know that it would be useful in some embedded projects
+    outside the Linux kernel.
+
+Conformance to the .xz file format specification
+
+    There are a couple of corner cases where things have been simplified
+    at expense of detecting errors as early as possible. These should not
+    matter in practice all, since they don't cause security issues. But
+    it is good to know this if testing the code e.g. with the test files
+    from XZ Utils.
+
+Reporting bugs
+
+    Before reporting a bug, please check that it's not fixed already
+    at upstream. See <http://tukaani.org/xz/embedded.html> to get the
+    latest code.
+
+    Report bugs to <lasse.collin@tukaani.org> or visit #tukaani on
+    Freenode and talk to Larhzu. I don't actively read LKML or other
+    kernel-related mailing lists, so if there's something I should know,
+    you should email to me personally or use IRC.
+
+    Don't bother Igor Pavlov with questions about the XZ implementation
+    in the kernel or about XZ Utils. While these two implementations
+    include essential code that is directly based on Igor Pavlov's code,
+    these implementations aren't maintained nor supported by him.
+/*
+ * XZ decompressor
+ *
+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
+ *          Igor Pavlov <http://7-zip.org/>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#ifndef XZ_H
+#define XZ_H
+
+#ifdef __KERNEL__
+#	include <linux/stddef.h>
+#	include <linux/types.h>
+#else
+#	include <stddef.h>
+#	include <stdint.h>
+#endif
+
+/* In Linux, this is used to make extern functions static when needed. */
+#ifndef XZ_EXTERN
+#	define XZ_EXTERN extern
+#endif
+
+/**
+ * enum xz_mode - Operation mode
+ *
+ * @XZ_SINGLE:              Single-call mode. This uses less RAM than
+ *                          than multi-call modes, because the LZMA2
+ *                          dictionary doesn't need to be allocated as
+ *                          part of the decoder state. All required data
+ *                          structures are allocated at initialization,
+ *                          so xz_dec_run() cannot return XZ_MEM_ERROR.
+ * @XZ_PREALLOC:            Multi-call mode with preallocated LZMA2
+ *                          dictionary buffer. All data structures are
+ *                          allocated at initialization, so xz_dec_run()
+ *                          cannot return XZ_MEM_ERROR.
+ * @XZ_DYNALLOC:            Multi-call mode. The LZMA2 dictionary is
+ *                          allocated once the required size has been
+ *                          parsed from the stream headers. If the
+ *                          allocation fails, xz_dec_run() will return
+ *                          XZ_MEM_ERROR.
+ *
+ * It is possible to enable support only for a subset of the above
+ * modes at compile time by defining XZ_DEC_SINGLE, XZ_DEC_PREALLOC,
+ * or XZ_DEC_DYNALLOC. The xz_dec kernel module is always compiled
+ * with support for all operation modes, but the preboot code may
+ * be built with fewer features to minimize code size.
+ */
+enum xz_mode {
+	XZ_SINGLE,
+	XZ_PREALLOC,
+	XZ_DYNALLOC
+};
+
+/**
+ * enum xz_ret - Return codes
+ * @XZ_OK:                  Everything is OK so far. More input or more
+ *                          output space is required to continue. This
+ *                          return code is possible only in multi-call mode
+ *                          (XZ_PREALLOC or XZ_DYNALLOC).
+ * @XZ_STREAM_END:          Operation finished successfully.
+ * @XZ_UNSUPPORTED_CHECK:   Integrity check type is not supported. Decoding
+ *                          is still possible in multi-call mode by simply
+ *                          calling xz_dec_run() again.
+ *                          Note that this return value is used only if
+ *                          XZ_DEC_ANY_CHECK was defined at build time,
+ *                          which is not used in the kernel. Unsupported
+ *                          check types return XZ_OPTIONS_ERROR if
+ *                          XZ_DEC_ANY_CHECK was not defined at build time.
+ * @XZ_MEM_ERROR:           Allocating memory failed. This return code is
+ *                          possible only if the decoder was initialized
+ *                          with XZ_DYNALLOC. The amount of memory that was
+ *                          tried to be allocated was no more than the
+ *                          dict_max argument given to xz_dec_init().
+ * @XZ_MEMLIMIT_ERROR:      A bigger LZMA2 dictionary would be needed than
+ *                          allowed by the dict_max argument given to
+ *                          xz_dec_init(). This return value is possible
+ *                          only in multi-call mode (XZ_PREALLOC or
+ *                          XZ_DYNALLOC); the single-call mode (XZ_SINGLE)
+ *                          ignores the dict_max argument.
+ * @XZ_FORMAT_ERROR:        File format was not recognized (wrong magic
+ *                          bytes).
+ * @XZ_OPTIONS_ERROR:       This implementation doesn't support the requested
+ *                          compression options. In the decoder this means
+ *                          that the header CRC32 matches, but the header
+ *                          itself specifies something that we don't support.
+ * @XZ_DATA_ERROR:          Compressed data is corrupt.
+ * @XZ_BUF_ERROR:           Cannot make any progress. Details are slightly
+ *                          different between multi-call and single-call
+ *                          mode; more information below.
+ *
+ * In multi-call mode, XZ_BUF_ERROR is returned when two consecutive calls
+ * to XZ code cannot consume any input and cannot produce any new output.
+ * This happens when there is no new input available, or the output buffer
+ * is full while at least one output byte is still pending. Assuming your
+ * code is not buggy, you can get this error only when decoding a compressed
+ * stream that is truncated or otherwise corrupt.
+ *
+ * In single-call mode, XZ_BUF_ERROR is returned only when the output buffer
+ * is too small or the compressed input is corrupt in a way that makes the
+ * decoder produce more output than the caller expected. When it is
+ * (relatively) clear that the compressed input is truncated, XZ_DATA_ERROR
+ * is used instead of XZ_BUF_ERROR.
+ */
+enum xz_ret {
+	XZ_OK,
+	XZ_STREAM_END,
+	XZ_UNSUPPORTED_CHECK,
+	XZ_MEM_ERROR,
+	XZ_MEMLIMIT_ERROR,
+	XZ_FORMAT_ERROR,
+	XZ_OPTIONS_ERROR,
+	XZ_DATA_ERROR,
+	XZ_BUF_ERROR
+};
+
+/**
+ * struct xz_buf - Passing input and output buffers to XZ code
+ * @in:         Beginning of the input buffer. This may be NULL if and only
+ *              if in_pos is equal to in_size.
+ * @in_pos:     Current position in the input buffer. This must not exceed
+ *              in_size.
+ * @in_size:    Size of the input buffer
+ * @out:        Beginning of the output buffer. This may be NULL if and only
+ *              if out_pos is equal to out_size.
+ * @out_pos:    Current position in the output buffer. This must not exceed
+ *              out_size.
+ * @out_size:   Size of the output buffer
+ *
+ * Only the contents of the output buffer from out[out_pos] onward, and
+ * the variables in_pos and out_pos are modified by the XZ code.
+ */
+struct xz_buf {
+	const uint8_t *in;
+	size_t in_pos;
+	size_t in_size;
+
+	uint8_t *out;
+	size_t out_pos;
+	size_t out_size;
+};
+
+/**
+ * struct xz_dec - Opaque type to hold the XZ decoder state
+ */
+struct xz_dec;
+
+/**
+ * xz_dec_init() - Allocate and initialize a XZ decoder state
+ * @mode:       Operation mode
+ * @dict_max:   Maximum size of the LZMA2 dictionary (history buffer) for
+ *              multi-call decoding. This is ignored in single-call mode
+ *              (mode == XZ_SINGLE). LZMA2 dictionary is always 2^n bytes
+ *              or 2^n + 2^(n-1) bytes (the latter sizes are less common
+ *              in practice), so other values for dict_max don't make sense.
+ *              In the kernel, dictionary sizes of 64 KiB, 128 KiB, 256 KiB,
+ *              512 KiB, and 1 MiB are probably the only reasonable values,
+ *              except for kernel and initramfs images where a bigger
+ *              dictionary can be fine and useful.
+ *
+ * Single-call mode (XZ_SINGLE): xz_dec_run() decodes the whole stream at
+ * once. The caller must provide enough output space or the decoding will
+ * fail. The output space is used as the dictionary buffer, which is why
+ * there is no need to allocate the dictionary as part of the decoder's
+ * internal state.
+ *
+ * Because the output buffer is used as the workspace, streams encoded using
+ * a big dictionary are not a problem in single-call mode. It is enough that
+ * the output buffer is big enough to hold the actual uncompressed data; it
+ * can be smaller than the dictionary size stored in the stream headers.
+ *
+ * Multi-call mode with preallocated dictionary (XZ_PREALLOC): dict_max bytes
+ * of memory is preallocated for the LZMA2 dictionary. This way there is no
+ * risk that xz_dec_run() could run out of memory, since xz_dec_run() will
+ * never allocate any memory. Instead, if the preallocated dictionary is too
+ * small for decoding the given input stream, xz_dec_run() will return
+ * XZ_MEMLIMIT_ERROR. Thus, it is important to know what kind of data will be
+ * decoded to avoid allocating excessive amount of memory for the dictionary.
+ *
+ * Multi-call mode with dynamically allocated dictionary (XZ_DYNALLOC):
+ * dict_max specifies the maximum allowed dictionary size that xz_dec_run()
+ * may allocate once it has parsed the dictionary size from the stream
+ * headers. This way excessive allocations can be avoided while still
+ * limiting the maximum memory usage to a sane value to prevent running the
+ * system out of memory when decompressing streams from untrusted sources.
+ *
+ * On success, xz_dec_init() returns a pointer to struct xz_dec, which is
+ * ready to be used with xz_dec_run(). If memory allocation fails,
+ * xz_dec_init() returns NULL.
+ */
+XZ_EXTERN struct xz_dec *xz_dec_init(enum xz_mode mode, uint32_t dict_max);
+
+/**
+ * xz_dec_run() - Run the XZ decoder
+ * @s:          Decoder state allocated using xz_dec_init()
+ * @b:          Input and output buffers
+ *
+ * The possible return values depend on build options and operation mode.
+ * See enum xz_ret for details.
+ *
+ * Note that if an error occurs in single-call mode (return value is not
+ * XZ_STREAM_END), b->in_pos and b->out_pos are not modified and the
+ * contents of the output buffer from b->out[b->out_pos] onward are
+ * undefined. This is true even after XZ_BUF_ERROR, because with some filter
+ * chains, there may be a second pass over the output buffer, and this pass
+ * cannot be properly done if the output buffer is truncated. Thus, you
+ * cannot give the single-call decoder a too small buffer and then expect to
+ * get that amount valid data from the beginning of the stream. You must use
+ * the multi-call decoder if you don't want to uncompress the whole stream.
+ */
+XZ_EXTERN enum xz_ret xz_dec_run(struct xz_dec *s, struct xz_buf *b);
+
+/**
+ * xz_dec_reset() - Reset an already allocated decoder state
+ * @s:          Decoder state allocated using xz_dec_init()
+ *
+ * This function can be used to reset the multi-call decoder state without
+ * freeing and reallocating memory with xz_dec_end() and xz_dec_init().
+ *
+ * In single-call mode, xz_dec_reset() is always called in the beginning of
+ * xz_dec_run(). Thus, explicit call to xz_dec_reset() is useful only in
+ * multi-call mode.
+ */
+XZ_EXTERN void xz_dec_reset(struct xz_dec *s);
+
+/**
+ * xz_dec_end() - Free the memory allocated for the decoder state
+ * @s:          Decoder state allocated using xz_dec_init(). If s is NULL,
+ *              this function does nothing.
+ */
+XZ_EXTERN void xz_dec_end(struct xz_dec *s);
+
+/*
+ * Standalone build (userspace build or in-kernel build for boot time use)
+ * needs a CRC32 implementation. For normal in-kernel use, kernel's own
+ * CRC32 module is used instead, and users of this module don't need to
+ * care about the functions below.
+ */
+#ifndef XZ_INTERNAL_CRC32
+#	ifdef __KERNEL__
+#		define XZ_INTERNAL_CRC32 0
+#	else
+#		define XZ_INTERNAL_CRC32 1
+#	endif
+#endif
+
+#if XZ_INTERNAL_CRC32
+/*
+ * This must be called before any other xz_* function to initialize
+ * the CRC32 lookup table.
+ */
+XZ_EXTERN void xz_crc32_init(void);
+
+/*
+ * Update CRC32 value using the polynomial from IEEE-802.3. To start a new
+ * calculation, the third argument must be zero. To continue the calculation,
+ * the previously returned value is passed as the third argument.
+ */
+XZ_EXTERN uint32_t xz_crc32(const uint8_t *buf, size_t size, uint32_t crc);
+#endif
+#endif
@@ -106,6 +106,8 @@
 config LZO_DECOMPRESS
 	tristate
  
+source "lib/xz/Kconfig"
+
 #
 # These all provide a common interface (hence the apparent duplication with
 # ZLIB_INFLATE; DECOMPRESS_GZIP is just a wrapper.)
@@ -69,6 +69,7 @@
 obj-$(CONFIG_REED_SOLOMON) += reed_solomon/
 obj-$(CONFIG_LZO_COMPRESS) += lzo/
 obj-$(CONFIG_LZO_DECOMPRESS) += lzo/
+obj-$(CONFIG_XZ_DEC) += xz/
 obj-$(CONFIG_RAID6_PQ) += raid6/
  
 lib-$(CONFIG_DECOMPRESS_GZIP) += decompress_inflate.o
+config XZ_DEC
+	tristate "XZ decompression support"
+	select CRC32
+	help
+	  LZMA2 compression algorithm and BCJ filters are supported using
+	  the .xz file format as the container. For integrity checking,
+	  CRC32 is supported. See Documentation/xz.txt for more information.
+
+config XZ_DEC_X86
+	bool "x86 BCJ filter decoder" if EMBEDDED
+	default y
+	depends on XZ_DEC
+	select XZ_DEC_BCJ
+
+config XZ_DEC_POWERPC
+	bool "PowerPC BCJ filter decoder" if EMBEDDED
+	default y
+	depends on XZ_DEC
+	select XZ_DEC_BCJ
+
+config XZ_DEC_IA64
+	bool "IA-64 BCJ filter decoder" if EMBEDDED
+	default y
+	depends on XZ_DEC
+	select XZ_DEC_BCJ
+
+config XZ_DEC_ARM
+	bool "ARM BCJ filter decoder" if EMBEDDED
+	default y
+	depends on XZ_DEC
+	select XZ_DEC_BCJ
+
+config XZ_DEC_ARMTHUMB
+	bool "ARM-Thumb BCJ filter decoder" if EMBEDDED
+	default y
+	depends on XZ_DEC
+	select XZ_DEC_BCJ
+
+config XZ_DEC_SPARC
+	bool "SPARC BCJ filter decoder" if EMBEDDED
+	default y
+	depends on XZ_DEC
+	select XZ_DEC_BCJ
+
+config XZ_DEC_BCJ
+	bool
+	default n
+
+config XZ_DEC_TEST
+	tristate "XZ decompressor tester"
+	default n
+	depends on XZ_DEC
+	help
+	  This allows passing .xz files to the in-kernel XZ decoder via
+	  a character special file. It calculates CRC32 of the decompressed
+	  data and writes diagnostics to the system log.
+
+	  Unless you are developing the XZ decoder, you don't need this
+	  and should say N.
+obj-$(CONFIG_XZ_DEC) += xz_dec.o
+xz_dec-y := xz_dec_syms.o xz_dec_stream.o xz_dec_lzma2.o
+xz_dec-$(CONFIG_XZ_DEC_BCJ) += xz_dec_bcj.o
+
+obj-$(CONFIG_XZ_DEC_TEST) += xz_dec_test.o
+/*
+ * CRC32 using the polynomial from IEEE-802.3
+ *
+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
+ *          Igor Pavlov <http://7-zip.org/>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+/*
+ * This is not the fastest implementation, but it is pretty compact.
+ * The fastest versions of xz_crc32() on modern CPUs without hardware
+ * accelerated CRC instruction are 3-5 times as fast as this version,
+ * but they are bigger and use more memory for the lookup table.
+ */
+
+#include "xz_private.h"
+
+/*
+ * STATIC_RW_DATA is used in the pre-boot environment on some architectures.
+ * See <linux/decompress/mm.h> for details.
+ */
+#ifndef STATIC_RW_DATA
+#	define STATIC_RW_DATA static
+#endif
+
+STATIC_RW_DATA uint32_t xz_crc32_table[256];
+
+XZ_EXTERN void xz_crc32_init(void)
+{
+	const uint32_t poly = 0xEDB88320;
+
+	uint32_t i;
+	uint32_t j;
+	uint32_t r;
+
+	for (i = 0; i < 256; ++i) {
+		r = i;
+		for (j = 0; j < 8; ++j)
+			r = (r >> 1) ^ (poly & ~((r & 1) - 1));
+
+		xz_crc32_table[i] = r;
+	}
+
+	return;
+}
+
+XZ_EXTERN uint32_t xz_crc32(const uint8_t *buf, size_t size, uint32_t crc)
+{
+	crc = ~crc;
+
+	while (size != 0) {
+		crc = xz_crc32_table[*buf++ ^ (crc & 0xFF)] ^ (crc >> 8);
+		--size;
+	}
+
+	return ~crc;
+}
+/*
+ * Branch/Call/Jump (BCJ) filter decoders
+ *
+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
+ *          Igor Pavlov <http://7-zip.org/>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#include "xz_private.h"
+
+/*
+ * The rest of the file is inside this ifdef. It makes things a little more
+ * convenient when building without support for any BCJ filters.
+ */
+#ifdef XZ_DEC_BCJ
+
+struct xz_dec_bcj {
+	/* Type of the BCJ filter being used */
+	enum {
+		BCJ_X86 = 4,        /* x86 or x86-64 */
+		BCJ_POWERPC = 5,    /* Big endian only */
+		BCJ_IA64 = 6,       /* Big or little endian */
+		BCJ_ARM = 7,        /* Little endian only */
+		BCJ_ARMTHUMB = 8,   /* Little endian only */
+		BCJ_SPARC = 9       /* Big or little endian */
+	} type;
+
+	/*
+	 * Return value of the next filter in the chain. We need to preserve
+	 * this information across calls, because we must not call the next
+	 * filter anymore once it has returned XZ_STREAM_END.
+	 */
+	enum xz_ret ret;
+
+	/* True if we are operating in single-call mode. */
+	bool single_call;
+
+	/*
+	 * Absolute position relative to the beginning of the uncompressed
+	 * data (in a single .xz Block). We care only about the lowest 32
+	 * bits so this doesn't need to be uint64_t even with big files.
+	 */
+	uint32_t pos;
+
+	/* x86 filter state */
+	uint32_t x86_prev_mask;
+
+	/* Temporary space to hold the variables from struct xz_buf */
+	uint8_t *out;
+	size_t out_pos;
+	size_t out_size;
+
+	struct {
+		/* Amount of already filtered data in the beginning of buf */
+		size_t filtered;
+
+		/* Total amount of data currently stored in buf  */
+		size_t size;
+
+		/*
+		 * Buffer to hold a mix of filtered and unfiltered data. This
+		 * needs to be big enough to hold Alignment + 2 * Look-ahead:
+		 *
+		 * Type         Alignment   Look-ahead
+		 * x86              1           4
+		 * PowerPC          4           0
+		 * IA-64           16           0
+		 * ARM              4           0
+		 * ARM-Thumb        2           2
+		 * SPARC            4           0
+		 */
+		uint8_t buf[16];
+	} temp;
+};
+
+#ifdef XZ_DEC_X86
+/*
+ * This is used to test the most significant byte of a memory address
+ * in an x86 instruction.
+ */
+static inline int bcj_x86_test_msbyte(uint8_t b)
+{
+	return b == 0x00 || b == 0xFF;
+}
+
+static size_t bcj_x86(struct xz_dec_bcj *s, uint8_t *buf, size_t size)
+{
+	static const bool mask_to_allowed_status[8]
+		= { true, true, true, false, true, false, false, false };
+
+	static const uint8_t mask_to_bit_num[8] = { 0, 1, 2, 2, 3, 3, 3, 3 };
+
+	size_t i;
+	size_t prev_pos = (size_t)-1;
+	uint32_t prev_mask = s->x86_prev_mask;
+	uint32_t src;
+	uint32_t dest;
+	uint32_t j;
+	uint8_t b;
+
+	if (size <= 4)
+		return 0;
+
+	size -= 4;
+	for (i = 0; i < size; ++i) {
+		if ((buf[i] & 0xFE) != 0xE8)
+			continue;
+
+		prev_pos = i - prev_pos;
+		if (prev_pos > 3) {
+			prev_mask = 0;
+		} else {
+			prev_mask = (prev_mask << (prev_pos - 1)) & 7;
+			if (prev_mask != 0) {
+				b = buf[i + 4 - mask_to_bit_num[prev_mask]];
+				if (!mask_to_allowed_status[prev_mask]
+						|| bcj_x86_test_msbyte(b)) {
+					prev_pos = i;
+					prev_mask = (prev_mask << 1) | 1;
+					continue;
+				}
+			}
+		}
+
+		prev_pos = i;
+
+		if (bcj_x86_test_msbyte(buf[i + 4])) {
+			src = get_unaligned_le32(buf + i + 1);
+			while (true) {
+				dest = src - (s->pos + (uint32_t)i + 5);
+				if (prev_mask == 0)
+					break;
+
+				j = mask_to_bit_num[prev_mask] * 8;
+				b = (uint8_t)(dest >> (24 - j));
+				if (!bcj_x86_test_msbyte(b))
+					break;
+
+				src = dest ^ (((uint32_t)1 << (32 - j)) - 1);
+			}
+
+			dest &= 0x01FFFFFF;
+			dest |= (uint32_t)0 - (dest & 0x01000000);
+			put_unaligned_le32(dest, buf + i + 1);
+			i += 4;
+		} else {
+			prev_mask = (prev_mask << 1) | 1;
+		}
+	}
+
+	prev_pos = i - prev_pos;
+	s->x86_prev_mask = prev_pos > 3 ? 0 : prev_mask << (prev_pos - 1);
+	return i;
+}
+#endif
+
+#ifdef XZ_DEC_POWERPC
+static size_t bcj_powerpc(struct xz_dec_bcj *s, uint8_t *buf, size_t size)
+{
+	size_t i;
+	uint32_t instr;
+
+	for (i = 0; i + 4 <= size; i += 4) {
+		instr = get_unaligned_be32(buf + i);
+		if ((instr & 0xFC000003) == 0x48000001) {
+			instr &= 0x03FFFFFC;
+			instr -= s->pos + (uint32_t)i;
+			instr &= 0x03FFFFFC;
+			instr |= 0x48000001;
+			put_unaligned_be32(instr, buf + i);
+		}
+	}
+
+	return i;
+}
+#endif
+
+#ifdef XZ_DEC_IA64
+static size_t bcj_ia64(struct xz_dec_bcj *s, uint8_t *buf, size_t size)
+{
+	static const uint8_t branch_table[32] = {
+		0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0,
+		4, 4, 6, 6, 0, 0, 7, 7,
+		4, 4, 0, 0, 4, 4, 0, 0
+	};
+
+	/*
+	 * The local variables take a little bit stack space, but it's less
+	 * than what LZMA2 decoder takes, so it doesn't make sense to reduce
+	 * stack usage here without doing that for the LZMA2 decoder too.
+	 */
+
+	/* Loop counters */
+	size_t i;
+	size_t j;
+
+	/* Instruction slot (0, 1, or 2) in the 128-bit instruction word */
+	uint32_t slot;
+
+	/* Bitwise offset of the instruction indicated by slot */
+	uint32_t bit_pos;
+
+	/* bit_pos split into byte and bit parts */
+	uint32_t byte_pos;
+	uint32_t bit_res;
+
+	/* Address part of an instruction */
+	uint32_t addr;
+
+	/* Mask used to detect which instructions to convert */
+	uint32_t mask;
+
+	/* 41-bit instruction stored somewhere in the lowest 48 bits */
+	uint64_t instr;
+
+	/* Instruction normalized with bit_res for easier manipulation */
+	uint64_t norm;
+
+	for (i = 0; i + 16 <= size; i += 16) {
+		mask = branch_table[buf[i] & 0x1F];
+		for (slot = 0, bit_pos = 5; slot < 3; ++slot, bit_pos += 41) {
+			if (((mask >> slot) & 1) == 0)
+				continue;
+
+			byte_pos = bit_pos >> 3;
+			bit_res = bit_pos & 7;
+			instr = 0;
+			for (j = 0; j < 6; ++j)
+				instr |= (uint64_t)(buf[i + j + byte_pos])
+						<< (8 * j);
+
+			norm = instr >> bit_res;
+
+			if (((norm >> 37) & 0x0F) == 0x05
+					&& ((norm >> 9) & 0x07) == 0) {
+				addr = (norm >> 13) & 0x0FFFFF;
+				addr |= ((uint32_t)(norm >> 36) & 1) << 20;
+				addr <<= 4;
+				addr -= s->pos + (uint32_t)i;
+				addr >>= 4;
+
+				norm &= ~((uint64_t)0x8FFFFF << 13);
+				norm |= (uint64_t)(addr & 0x0FFFFF) << 13;
+				norm |= (uint64_t)(addr & 0x100000)
+						<< (36 - 20);
+
+				instr &= (1 << bit_res) - 1;
+				instr |= norm << bit_res;
+
+				for (j = 0; j < 6; j++)
+					buf[i + j + byte_pos]
+						= (uint8_t)(instr >> (8 * j));
+			}
+		}
+	}
+
+	return i;
+}
+#endif
+
+#ifdef XZ_DEC_ARM
+static size_t bcj_arm(struct xz_dec_bcj *s, uint8_t *buf, size_t size)
+{
+	size_t i;
+	uint32_t addr;
+
+	for (i = 0; i + 4 <= size; i += 4) {
+		if (buf[i + 3] == 0xEB) {
+			addr = (uint32_t)buf[i] | ((uint32_t)buf[i + 1] << 8)
+					| ((uint32_t)buf[i + 2] << 16);
+			addr <<= 2;
+			addr -= s->pos + (uint32_t)i + 8;
+			addr >>= 2;
+			buf[i] = (uint8_t)addr;
+			buf[i + 1] = (uint8_t)(addr >> 8);
+			buf[i + 2] = (uint8_t)(addr >> 16);
+		}
+	}
+
+	return i;
+}
+#endif
+
+#ifdef XZ_DEC_ARMTHUMB
+static size_t bcj_armthumb(struct xz_dec_bcj *s, uint8_t *buf, size_t size)
+{
+	size_t i;
+	uint32_t addr;
+
+	for (i = 0; i + 4 <= size; i += 2) {
+		if ((buf[i + 1] & 0xF8) == 0xF0
+				&& (buf[i + 3] & 0xF8) == 0xF8) {
+			addr = (((uint32_t)buf[i + 1] & 0x07) << 19)
+					| ((uint32_t)buf[i] << 11)
+					| (((uint32_t)buf[i + 3] & 0x07) << 8)
+					| (uint32_t)buf[i + 2];
+			addr <<= 1;
+			addr -= s->pos + (uint32_t)i + 4;
+			addr >>= 1;
+			buf[i + 1] = (uint8_t)(0xF0 | ((addr >> 19) & 0x07));
+			buf[i] = (uint8_t)(addr >> 11);
+			buf[i + 3] = (uint8_t)(0xF8 | ((addr >> 8) & 0x07));
+			buf[i + 2] = (uint8_t)addr;
+			i += 2;
+		}
+	}
+
+	return i;
+}
+#endif
+
+#ifdef XZ_DEC_SPARC
+static size_t bcj_sparc(struct xz_dec_bcj *s, uint8_t *buf, size_t size)
+{
+	size_t i;
+	uint32_t instr;
+
+	for (i = 0; i + 4 <= size; i += 4) {
+		instr = get_unaligned_be32(buf + i);
+		if ((instr >> 22) == 0x100 || (instr >> 22) == 0x1FF) {
+			instr <<= 2;
+			instr -= s->pos + (uint32_t)i;
+			instr >>= 2;
+			instr = ((uint32_t)0x40000000 - (instr & 0x400000))
+					| 0x40000000 | (instr & 0x3FFFFF);
+			put_unaligned_be32(instr, buf + i);
+		}
+	}
+
+	return i;
+}
+#endif
+
+/*
+ * Apply the selected BCJ filter. Update *pos and s->pos to match the amount
+ * of data that got filtered.
+ *
+ * NOTE: This is implemented as a switch statement to avoid using function
+ * pointers, which could be problematic in the kernel boot code, which must
+ * avoid pointers to static data (at least on x86).
+ */
+static void bcj_apply(struct xz_dec_bcj *s,
+		      uint8_t *buf, size_t *pos, size_t size)
+{
+	size_t filtered;
+
+	buf += *pos;
+	size -= *pos;
+
+	switch (s->type) {
+#ifdef XZ_DEC_X86
+	case BCJ_X86:
+		filtered = bcj_x86(s, buf, size);
+		break;
+#endif
+#ifdef XZ_DEC_POWERPC
+	case BCJ_POWERPC:
+		filtered = bcj_powerpc(s, buf, size);
+		break;
+#endif
+#ifdef XZ_DEC_IA64
+	case BCJ_IA64:
+		filtered = bcj_ia64(s, buf, size);
+		break;
+#endif
+#ifdef XZ_DEC_ARM
+	case BCJ_ARM:
+		filtered = bcj_arm(s, buf, size);
+		break;
+#endif
+#ifdef XZ_DEC_ARMTHUMB
+	case BCJ_ARMTHUMB:
+		filtered = bcj_armthumb(s, buf, size);
+		break;
+#endif
+#ifdef XZ_DEC_SPARC
+	case BCJ_SPARC:
+		filtered = bcj_sparc(s, buf, size);
+		break;
+#endif
+	default:
+		/* Never reached but silence compiler warnings. */
+		filtered = 0;
+		break;
+	}
+
+	*pos += filtered;
+	s->pos += filtered;
+}
+
+/*
+ * Flush pending filtered data from temp to the output buffer.
+ * Move the remaining mixture of possibly filtered and unfiltered
+ * data to the beginning of temp.
+ */
+static void bcj_flush(struct xz_dec_bcj *s, struct xz_buf *b)
+{
+	size_t copy_size;
+
+	copy_size = min_t(size_t, s->temp.filtered, b->out_size - b->out_pos);
+	memcpy(b->out + b->out_pos, s->temp.buf, copy_size);
+	b->out_pos += copy_size;
+
+	s->temp.filtered -= copy_size;
+	s->temp.size -= copy_size;
+	memmove(s->temp.buf, s->temp.buf + copy_size, s->temp.size);
+}
+
+/*
+ * The BCJ filter functions are primitive in sense that they process the
+ * data in chunks of 1-16 bytes. To hide this issue, this function does
+ * some buffering.
+ */
+XZ_EXTERN enum xz_ret xz_dec_bcj_run(struct xz_dec_bcj *s,
+				     struct xz_dec_lzma2 *lzma2,
+				     struct xz_buf *b)
+{
+	size_t out_start;
+
+	/*
+	 * Flush pending already filtered data to the output buffer. Return
+	 * immediatelly if we couldn't flush everything, or if the next
+	 * filter in the chain had already returned XZ_STREAM_END.
+	 */
+	if (s->temp.filtered > 0) {
+		bcj_flush(s, b);
+		if (s->temp.filtered > 0)
+			return XZ_OK;
+
+		if (s->ret == XZ_STREAM_END)
+			return XZ_STREAM_END;
+	}
+
+	/*
+	 * If we have more output space than what is currently pending in
+	 * temp, copy the unfiltered data from temp to the output buffer
+	 * and try to fill the output buffer by decoding more data from the
+	 * next filter in the chain. Apply the BCJ filter on the new data
+	 * in the output buffer. If everything cannot be filtered, copy it
+	 * to temp and rewind the output buffer position accordingly.
+	 */
+	if (s->temp.size < b->out_size - b->out_pos) {
+		out_start = b->out_pos;
+		memcpy(b->out + b->out_pos, s->temp.buf, s->temp.size);
+		b->out_pos += s->temp.size;
+
+		s->ret = xz_dec_lzma2_run(lzma2, b);
+		if (s->ret != XZ_STREAM_END
+				&& (s->ret != XZ_OK || s->single_call))
+			return s->ret;
+
+		bcj_apply(s, b->out, &out_start, b->out_pos);
+
+		/*
+		 * As an exception, if the next filter returned XZ_STREAM_END,
+		 * we can do that too, since the last few bytes that remain
+		 * unfiltered are meant to remain unfiltered.
+		 */
+		if (s->ret == XZ_STREAM_END)
+			return XZ_STREAM_END;
+
+		s->temp.size = b->out_pos - out_start;
+		b->out_pos -= s->temp.size;
+		memcpy(s->temp.buf, b->out + b->out_pos, s->temp.size);
+	}
+
+	/*
+	 * If we have unfiltered data in temp, try to fill by decoding more
+	 * data from the next filter. Apply the BCJ filter on temp. Then we
+	 * hopefully can fill the actual output buffer by copying filtered
+	 * data from temp. A mix of filtered and unfiltered data may be left
+	 * in temp; it will be taken care on the next call to this function.
+	 */
+	if (s->temp.size > 0) {
+		/* Make b->out{,_pos,_size} temporarily point to s->temp. */
+		s->out = b->out;
+		s->out_pos = b->out_pos;
+		s->out_size = b->out_size;
+		b->out = s->temp.buf;
+		b->out_pos = s->temp.size;
+		b->out_size = sizeof(s->temp.buf);
+
+		s->ret = xz_dec_lzma2_run(lzma2, b);
+
+		s->temp.size = b->out_pos;
+		b->out = s->out;
+		b->out_pos = s->out_pos;
+		b->out_size = s->out_size;
+
+		if (s->ret != XZ_OK && s->ret != XZ_STREAM_END)
+			return s->ret;
+
+		bcj_apply(s, s->temp.buf, &s->temp.filtered, s->temp.size);
+
+		/*
+		 * If the next filter returned XZ_STREAM_END, we mark that
+		 * everything is filtered, since the last unfiltered bytes
+		 * of the stream are meant to be left as is.
+		 */
+		if (s->ret == XZ_STREAM_END)
+			s->temp.filtered = s->temp.size;
+
+		bcj_flush(s, b);
+		if (s->temp.filtered > 0)
+			return XZ_OK;
+	}
+
+	return s->ret;
+}
+
+XZ_EXTERN struct xz_dec_bcj *xz_dec_bcj_create(bool single_call)
+{
+	struct xz_dec_bcj *s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (s != NULL)
+		s->single_call = single_call;
+
+	return s;
+}
+
+XZ_EXTERN enum xz_ret xz_dec_bcj_reset(struct xz_dec_bcj *s, uint8_t id)
+{
+	switch (id) {
+#ifdef XZ_DEC_X86
+	case BCJ_X86:
+#endif
+#ifdef XZ_DEC_POWERPC
+	case BCJ_POWERPC:
+#endif
+#ifdef XZ_DEC_IA64
+	case BCJ_IA64:
+#endif
+#ifdef XZ_DEC_ARM
+	case BCJ_ARM:
+#endif
+#ifdef XZ_DEC_ARMTHUMB
+	case BCJ_ARMTHUMB:
+#endif
+#ifdef XZ_DEC_SPARC
+	case BCJ_SPARC:
+#endif
+		break;
+
+	default:
+		/* Unsupported Filter ID */
+		return XZ_OPTIONS_ERROR;
+	}
+
+	s->type = id;
+	s->ret = XZ_OK;
+	s->pos = 0;
+	s->x86_prev_mask = 0;
+	s->temp.filtered = 0;
+	s->temp.size = 0;
+
+	return XZ_OK;
+}
+
+#endif
+/*
+ * LZMA2 decoder
+ *
+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
+ *          Igor Pavlov <http://7-zip.org/>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#include "xz_private.h"
+#include "xz_lzma2.h"
+
+/*
+ * Range decoder initialization eats the first five bytes of each LZMA chunk.
+ */
+#define RC_INIT_BYTES 5
+
+/*
+ * Minimum number of usable input buffer to safely decode one LZMA symbol.
+ * The worst case is that we decode 22 bits using probabilities and 26
+ * direct bits. This may decode at maximum of 20 bytes of input. However,
+ * lzma_main() does an extra normalization before returning, thus we
+ * need to put 21 here.
+ */
+#define LZMA_IN_REQUIRED 21
+
+/*
+ * Dictionary (history buffer)
+ *
+ * These are always true:
+ *    start <= pos <= full <= end
+ *    pos <= limit <= end
+ *
+ * In multi-call mode, also these are true:
+ *    end == size
+ *    size <= size_max
+ *    allocated <= size
+ *
+ * Most of these variables are size_t to support single-call mode,
+ * in which the dictionary variables address the actual output
+ * buffer directly.
+ */
+struct dictionary {
+	/* Beginning of the history buffer */
+	uint8_t *buf;
+
+	/* Old position in buf (before decoding more data) */
+	size_t start;
+
+	/* Position in buf */
+	size_t pos;
+
+	/*
+	 * How full dictionary is. This is used to detect corrupt input that
+	 * would read beyond the beginning of the uncompressed stream.
+	 */
+	size_t full;
+
+	/* Write limit; we don't write to buf[limit] or later bytes. */
+	size_t limit;
+
+	/*
+	 * End of the dictionary buffer. In multi-call mode, this is
+	 * the same as the dictionary size. In single-call mode, this
+	 * indicates the size of the output buffer.
+	 */
+	size_t end;
+
+	/*
+	 * Size of the dictionary as specified in Block Header. This is used
+	 * together with "full" to detect corrupt input that would make us
+	 * read beyond the beginning of the uncompressed stream.
+	 */
+	uint32_t size;
+
+	/*
+	 * Maximum allowed dictionary size in multi-call mode.
+	 * This is ignored in single-call mode.
+	 */
+	uint32_t size_max;
+
+	/*
+	 * Amount of memory currently allocated for the dictionary.
+	 * This is used only with XZ_DYNALLOC. (With XZ_PREALLOC,
+	 * size_max is always the same as the allocated size.)
+	 */
+	uint32_t allocated;
+
+	/* Operation mode */
+	enum xz_mode mode;
+};
+
+/* Range decoder */
+struct rc_dec {
+	uint32_t range;
+	uint32_t code;
+
+	/*
+	 * Number of initializing bytes remaining to be read
+	 * by rc_read_init().
+	 */
+	uint32_t init_bytes_left;
+
+	/*
+	 * Buffer from which we read our input. It can be either
+	 * temp.buf or the caller-provided input buffer.
+	 */
+	const uint8_t *in;
+	size_t in_pos;
+	size_t in_limit;
+};
+
+/* Probabilities for a length decoder. */
+struct lzma_len_dec {
+	/* Probability of match length being at least 10 */
+	uint16_t choice;
+
+	/* Probability of match length being at least 18 */
+	uint16_t choice2;
+
+	/* Probabilities for match lengths 2-9 */
+	uint16_t low[POS_STATES_MAX][LEN_LOW_SYMBOLS];
+
+	/* Probabilities for match lengths 10-17 */
+	uint16_t mid[POS_STATES_MAX][LEN_MID_SYMBOLS];
+
+	/* Probabilities for match lengths 18-273 */
+	uint16_t high[LEN_HIGH_SYMBOLS];
+};
+
+struct lzma_dec {
+	/* Distances of latest four matches */
+	uint32_t rep0;
+	uint32_t rep1;
+	uint32_t rep2;
+	uint32_t rep3;
+
+	/* Types of the most recently seen LZMA symbols */
+	enum lzma_state state;
+
+	/*
+	 * Length of a match. This is updated so that dict_repeat can
+	 * be called again to finish repeating the whole match.
+	 */
+	uint32_t len;
+
+	/*
+	 * LZMA properties or related bit masks (number of literal
+	 * context bits, a mask dervied from the number of literal
+	 * position bits, and a mask dervied from the number
+	 * position bits)
+	 */
+	uint32_t lc;
+	uint32_t literal_pos_mask; /* (1 << lp) - 1 */
+	uint32_t pos_mask;         /* (1 << pb) - 1 */
+
+	/* If 1, it's a match. Otherwise it's a single 8-bit literal. */
+	uint16_t is_match[STATES][POS_STATES_MAX];
+
+	/* If 1, it's a repeated match. The distance is one of rep0 .. rep3. */
+	uint16_t is_rep[STATES];
+
+	/*
+	 * If 0, distance of a repeated match is rep0.
+	 * Otherwise check is_rep1.
+	 */
+	uint16_t is_rep0[STATES];
+
+	/*
+	 * If 0, distance of a repeated match is rep1.
+	 * Otherwise check is_rep2.
+	 */
+	uint16_t is_rep1[STATES];
+
+	/* If 0, distance of a repeated match is rep2. Otherwise it is rep3. */
+	uint16_t is_rep2[STATES];
+
+	/*
+	 * If 1, the repeated match has length of one byte. Otherwise
+	 * the length is decoded from rep_len_decoder.
+	 */
+	uint16_t is_rep0_long[STATES][POS_STATES_MAX];
+
+	/*
+	 * Probability tree for the highest two bits of the match
+	 * distance. There is a separate probability tree for match
+	 * lengths of 2 (i.e. MATCH_LEN_MIN), 3, 4, and [5, 273].
+	 */
+	uint16_t dist_slot[DIST_STATES][DIST_SLOTS];
+
+	/*
+	 * Probility trees for additional bits for match distance
+	 * when the distance is in the range [4, 127].
+	 */
+	uint16_t dist_special[FULL_DISTANCES - DIST_MODEL_END];
+
+	/*
+	 * Probability tree for the lowest four bits of a match
+	 * distance that is equal to or greater than 128.
+	 */
+	uint16_t dist_align[ALIGN_SIZE];
+
+	/* Length of a normal match */
+	struct lzma_len_dec match_len_dec;
+
+	/* Length of a repeated match */
+	struct lzma_len_dec rep_len_dec;
+
+	/* Probabilities of literals */
+	uint16_t literal[LITERAL_CODERS_MAX][LITERAL_CODER_SIZE];
+};
+
+struct lzma2_dec {
+	/* Position in xz_dec_lzma2_run(). */
+	enum lzma2_seq {
+		SEQ_CONTROL,
+		SEQ_UNCOMPRESSED_1,
+		SEQ_UNCOMPRESSED_2,
+		SEQ_COMPRESSED_0,
+		SEQ_COMPRESSED_1,
+		SEQ_PROPERTIES,
+		SEQ_LZMA_PREPARE,
+		SEQ_LZMA_RUN,
+		SEQ_COPY
+	} sequence;
+
+	/* Next position after decoding the compressed size of the chunk. */
+	enum lzma2_seq next_sequence;
+
+	/* Uncompressed size of LZMA chunk (2 MiB at maximum) */
+	uint32_t uncompressed;
+
+	/*
+	 * Compressed size of LZMA chunk or compressed/uncompressed
+	 * size of uncompressed chunk (64 KiB at maximum)
+	 */
+	uint32_t compressed;
+
+	/*
+	 * True if dictionary reset is needed. This is false before
+	 * the first chunk (LZMA or uncompressed).
+	 */
+	bool need_dict_reset;
+
+	/*
+	 * True if new LZMA properties are needed. This is false
+	 * before the first LZMA chunk.
+	 */
+	bool need_props;
+};
+
+struct xz_dec_lzma2 {
+	/*
+	 * The order below is important on x86 to reduce code size and
+	 * it shouldn't hurt on other platforms. Everything up to and
+	 * including lzma.pos_mask are in the first 128 bytes on x86-32,
+	 * which allows using smaller instructions to access those
+	 * variables. On x86-64, fewer variables fit into the first 128
+	 * bytes, but this is still the best order without sacrificing
+	 * the readability by splitting the structures.
+	 */
+	struct rc_dec rc;
+	struct dictionary dict;
+	struct lzma2_dec lzma2;
+	struct lzma_dec lzma;
+
+	/*
+	 * Temporary buffer which holds small number of input bytes between
+	 * decoder calls. See lzma2_lzma() for details.
+	 */
+	struct {
+		uint32_t size;
+		uint8_t buf[3 * LZMA_IN_REQUIRED];
+	} temp;
+};
+
+/**************
+ * Dictionary *
+ **************/
+
+/*
+ * Reset the dictionary state. When in single-call mode, set up the beginning
+ * of the dictionary to point to the actual output buffer.
+ */
+static void dict_reset(struct dictionary *dict, struct xz_buf *b)
+{
+	if (DEC_IS_SINGLE(dict->mode)) {
+		dict->buf = b->out + b->out_pos;
+		dict->end = b->out_size - b->out_pos;
+	}
+
+	dict->start = 0;
+	dict->pos = 0;
+	dict->limit = 0;
+	dict->full = 0;
+}
+
+/* Set dictionary write limit */
+static void dict_limit(struct dictionary *dict, size_t out_max)
+{
+	if (dict->end - dict->pos <= out_max)
+		dict->limit = dict->end;
+	else
+		dict->limit = dict->pos + out_max;
+}
+
+/* Return true if at least one byte can be written into the dictionary. */
+static inline bool dict_has_space(const struct dictionary *dict)
+{
+	return dict->pos < dict->limit;
+}
+
+/*
+ * Get a byte from the dictionary at the given distance. The distance is
+ * assumed to valid, or as a special case, zero when the dictionary is
+ * still empty. This special case is needed for single-call decoding to
+ * avoid writing a '\0' to the end of the destination buffer.
+ */
+static inline uint32_t dict_get(const struct dictionary *dict, uint32_t dist)
+{
+	size_t offset = dict->pos - dist - 1;
+
+	if (dist >= dict->pos)
+		offset += dict->end;
+
+	return dict->full > 0 ? dict->buf[offset] : 0;
+}
+
+/*
+ * Put one byte into the dictionary. It is assumed that there is space for it.
+ */
+static inline void dict_put(struct dictionary *dict, uint8_t byte)
+{
+	dict->buf[dict->pos++] = byte;
+
+	if (dict->full < dict->pos)
+		dict->full = dict->pos;
+}
+
+/*
+ * Repeat given number of bytes from the given distance. If the distance is
+ * invalid, false is returned. On success, true is returned and *len is
+ * updated to indicate how many bytes were left to be repeated.
+ */
+static bool dict_repeat(struct dictionary *dict, uint32_t *len, uint32_t dist)
+{
+	size_t back;
+	uint32_t left;
+
+	if (dist >= dict->full || dist >= dict->size)
+		return false;
+
+	left = min_t(size_t, dict->limit - dict->pos, *len);
+	*len -= left;
+
+	back = dict->pos - dist - 1;
+	if (dist >= dict->pos)
+		back += dict->end;
+
+	do {
+		dict->buf[dict->pos++] = dict->buf[back++];
+		if (back == dict->end)
+			back = 0;
+	} while (--left > 0);
+
+	if (dict->full < dict->pos)
+		dict->full = dict->pos;
+
+	return true;
+}
+
+/* Copy uncompressed data as is from input to dictionary and output buffers. */
+static void dict_uncompressed(struct dictionary *dict, struct xz_buf *b,
+			      uint32_t *left)
+{
+	size_t copy_size;
+
+	while (*left > 0 && b->in_pos < b->in_size
+			&& b->out_pos < b->out_size) {
+		copy_size = min(b->in_size - b->in_pos,
+				b->out_size - b->out_pos);
+		if (copy_size > dict->end - dict->pos)
+			copy_size = dict->end - dict->pos;
+		if (copy_size > *left)
+			copy_size = *left;
+
+		*left -= copy_size;
+
+		memcpy(dict->buf + dict->pos, b->in + b->in_pos, copy_size);
+		dict->pos += copy_size;
+
+		if (dict->full < dict->pos)
+			dict->full = dict->pos;
+
+		if (DEC_IS_MULTI(dict->mode)) {
+			if (dict->pos == dict->end)
+				dict->pos = 0;
+
+			memcpy(b->out + b->out_pos, b->in + b->in_pos,
+					copy_size);
+		}
+
+		dict->start = dict->pos;
+
+		b->out_pos += copy_size;
+		b->in_pos += copy_size;
+	}
+}
+
+/*
+ * Flush pending data from dictionary to b->out. It is assumed that there is
+ * enough space in b->out. This is guaranteed because caller uses dict_limit()
+ * before decoding data into the dictionary.
+ */
+static uint32_t dict_flush(struct dictionary *dict, struct xz_buf *b)
+{
+	size_t copy_size = dict->pos - dict->start;
+
+	if (DEC_IS_MULTI(dict->mode)) {
+		if (dict->pos == dict->end)
+			dict->pos = 0;
+
+		memcpy(b->out + b->out_pos, dict->buf + dict->start,
+				copy_size);
+	}
+
+	dict->start = dict->pos;
+	b->out_pos += copy_size;
+	return copy_size;
+}
+
+/*****************
+ * Range decoder *
+ *****************/
+
+/* Reset the range decoder. */
+static void rc_reset(struct rc_dec *rc)
+{
+	rc->range = (uint32_t)-1;
+	rc->code = 0;
+	rc->init_bytes_left = RC_INIT_BYTES;
+}
+
+/*
+ * Read the first five initial bytes into rc->code if they haven't been
+ * read already. (Yes, the first byte gets completely ignored.)
+ */
+static bool rc_read_init(struct rc_dec *rc, struct xz_buf *b)
+{
+	while (rc->init_bytes_left > 0) {
+		if (b->in_pos == b->in_size)
+			return false;
+
+		rc->code = (rc->code << 8) + b->in[b->in_pos++];
+		--rc->init_bytes_left;
+	}
+
+	return true;
+}
+
+/* Return true if there may not be enough input for the next decoding loop. */
+static inline bool rc_limit_exceeded(const struct rc_dec *rc)
+{
+	return rc->in_pos > rc->in_limit;
+}
+
+/*
+ * Return true if it is possible (from point of view of range decoder) that
+ * we have reached the end of the LZMA chunk.
+ */
+static inline bool rc_is_finished(const struct rc_dec *rc)
+{
+	return rc->code == 0;
+}
+
+/* Read the next input byte if needed. */
+static __always_inline void rc_normalize(struct rc_dec *rc)
+{
+	if (rc->range < RC_TOP_VALUE) {
+		rc->range <<= RC_SHIFT_BITS;
+		rc->code = (rc->code << RC_SHIFT_BITS) + rc->in[rc->in_pos++];
+	}
+}
+
+/*
+ * Decode one bit. In some versions, this function has been splitted in three
+ * functions so that the compiler is supposed to be able to more easily avoid
+ * an extra branch. In this particular version of the LZMA decoder, this
+ * doesn't seem to be a good idea (tested with GCC 3.3.6, 3.4.6, and 4.3.3
+ * on x86). Using a non-splitted version results in nicer looking code too.
+ *
+ * NOTE: This must return an int. Do not make it return a bool or the speed
+ * of the code generated by GCC 3.x decreases 10-15 %. (GCC 4.3 doesn't care,
+ * and it generates 10-20 % faster code than GCC 3.x from this file anyway.)
+ */
+static __always_inline int rc_bit(struct rc_dec *rc, uint16_t *prob)
+{
+	uint32_t bound;
+	int bit;
+
+	rc_normalize(rc);
+	bound = (rc->range >> RC_BIT_MODEL_TOTAL_BITS) * *prob;
+	if (rc->code < bound) {
+		rc->range = bound;
+		*prob += (RC_BIT_MODEL_TOTAL - *prob) >> RC_MOVE_BITS;
+		bit = 0;
+	} else {
+		rc->range -= bound;
+		rc->code -= bound;
+		*prob -= *prob >> RC_MOVE_BITS;
+		bit = 1;
+	}
+
+	return bit;
+}
+
+/* Decode a bittree starting from the most significant bit. */
+static __always_inline uint32_t rc_bittree(struct rc_dec *rc,
+					   uint16_t *probs, uint32_t limit)
+{
+	uint32_t symbol = 1;
+
+	do {
+		if (rc_bit(rc, &probs[symbol]))
+			symbol = (symbol << 1) + 1;
+		else
+			symbol <<= 1;
+	} while (symbol < limit);
+
+	return symbol;
+}
+
+/* Decode a bittree starting from the least significant bit. */
+static __always_inline void rc_bittree_reverse(struct rc_dec *rc,
+					       uint16_t *probs,
+					       uint32_t *dest, uint32_t limit)
+{
+	uint32_t symbol = 1;
+	uint32_t i = 0;
+
+	do {
+		if (rc_bit(rc, &probs[symbol])) {
+			symbol = (symbol << 1) + 1;
+			*dest += 1 << i;
+		} else {
+			symbol <<= 1;
+		}
+	} while (++i < limit);
+}
+
+/* Decode direct bits (fixed fifty-fifty probability) */
+static inline void rc_direct(struct rc_dec *rc, uint32_t *dest, uint32_t limit)
+{
+	uint32_t mask;
+
+	do {
+		rc_normalize(rc);
+		rc->range >>= 1;
+		rc->code -= rc->range;
+		mask = (uint32_t)0 - (rc->code >> 31);
+		rc->code += rc->range & mask;
+		*dest = (*dest << 1) + (mask + 1);
+	} while (--limit > 0);
+}
+
+/********
+ * LZMA *
+ ********/
+
+/* Get pointer to literal coder probability array. */
+static uint16_t *lzma_literal_probs(struct xz_dec_lzma2 *s)
+{
+	uint32_t prev_byte = dict_get(&s->dict, 0);
+	uint32_t low = prev_byte >> (8 - s->lzma.lc);
+	uint32_t high = (s->dict.pos & s->lzma.literal_pos_mask) << s->lzma.lc;
+	return s->lzma.literal[low + high];
+}
+
+/* Decode a literal (one 8-bit byte) */
+static void lzma_literal(struct xz_dec_lzma2 *s)
+{
+	uint16_t *probs;
+	uint32_t symbol;
+	uint32_t match_byte;
+	uint32_t match_bit;
+	uint32_t offset;
+	uint32_t i;
+
+	probs = lzma_literal_probs(s);
+
+	if (lzma_state_is_literal(s->lzma.state)) {
+		symbol = rc_bittree(&s->rc, probs, 0x100);
+	} else {
+		symbol = 1;
+		match_byte = dict_get(&s->dict, s->lzma.rep0) << 1;
+		offset = 0x100;
+
+		do {
+			match_bit = match_byte & offset;
+			match_byte <<= 1;
+			i = offset + match_bit + symbol;
+
+			if (rc_bit(&s->rc, &probs[i])) {
+				symbol = (symbol << 1) + 1;
+				offset &= match_bit;
+			} else {
+				symbol <<= 1;
+				offset &= ~match_bit;
+			}
+		} while (symbol < 0x100);
+	}
+
+	dict_put(&s->dict, (uint8_t)symbol);
+	lzma_state_literal(&s->lzma.state);
+}
+
+/* Decode the length of the match into s->lzma.len. */
+static void lzma_len(struct xz_dec_lzma2 *s, struct lzma_len_dec *l,
+		     uint32_t pos_state)
+{
+	uint16_t *probs;
+	uint32_t limit;
+
+	if (!rc_bit(&s->rc, &l->choice)) {
+		probs = l->low[pos_state];
+		limit = LEN_LOW_SYMBOLS;
+		s->lzma.len = MATCH_LEN_MIN;
+	} else {
+		if (!rc_bit(&s->rc, &l->choice2)) {
+			probs = l->mid[pos_state];
+			limit = LEN_MID_SYMBOLS;
+			s->lzma.len = MATCH_LEN_MIN + LEN_LOW_SYMBOLS;
+		} else {
+			probs = l->high;
+			limit = LEN_HIGH_SYMBOLS;
+			s->lzma.len = MATCH_LEN_MIN + LEN_LOW_SYMBOLS
+					+ LEN_MID_SYMBOLS;
+		}
+	}
+
+	s->lzma.len += rc_bittree(&s->rc, probs, limit) - limit;
+}
+
+/* Decode a match. The distance will be stored in s->lzma.rep0. */
+static void lzma_match(struct xz_dec_lzma2 *s, uint32_t pos_state)
+{
+	uint16_t *probs;
+	uint32_t dist_slot;
+	uint32_t limit;
+
+	lzma_state_match(&s->lzma.state);
+
+	s->lzma.rep3 = s->lzma.rep2;
+	s->lzma.rep2 = s->lzma.rep1;
+	s->lzma.rep1 = s->lzma.rep0;
+
+	lzma_len(s, &s->lzma.match_len_dec, pos_state);
+
+	probs = s->lzma.dist_slot[lzma_get_dist_state(s->lzma.len)];
+	dist_slot = rc_bittree(&s->rc, probs, DIST_SLOTS) - DIST_SLOTS;
+
+	if (dist_slot < DIST_MODEL_START) {
+		s->lzma.rep0 = dist_slot;
+	} else {
+		limit = (dist_slot >> 1) - 1;
+		s->lzma.rep0 = 2 + (dist_slot & 1);
+
+		if (dist_slot < DIST_MODEL_END) {
+			s->lzma.rep0 <<= limit;
+			probs = s->lzma.dist_special + s->lzma.rep0
+					- dist_slot - 1;
+			rc_bittree_reverse(&s->rc, probs,
+					&s->lzma.rep0, limit);
+		} else {
+			rc_direct(&s->rc, &s->lzma.rep0, limit - ALIGN_BITS);
+			s->lzma.rep0 <<= ALIGN_BITS;
+			rc_bittree_reverse(&s->rc, s->lzma.dist_align,
+					&s->lzma.rep0, ALIGN_BITS);
+		}
+	}
+}
+
+/*
+ * Decode a repeated match. The distance is one of the four most recently
+ * seen matches. The distance will be stored in s->lzma.rep0.
+ */
+static void lzma_rep_match(struct xz_dec_lzma2 *s, uint32_t pos_state)
+{
+	uint32_t tmp;
+
+	if (!rc_bit(&s->rc, &s->lzma.is_rep0[s->lzma.state])) {
+		if (!rc_bit(&s->rc, &s->lzma.is_rep0_long[
+				s->lzma.state][pos_state])) {
+			lzma_state_short_rep(&s->lzma.state);
+			s->lzma.len = 1;
+			return;
+		}
+	} else {
+		if (!rc_bit(&s->rc, &s->lzma.is_rep1[s->lzma.state])) {
+			tmp = s->lzma.rep1;
+		} else {
+			if (!rc_bit(&s->rc, &s->lzma.is_rep2[s->lzma.state])) {
+				tmp = s->lzma.rep2;
+			} else {
+				tmp = s->lzma.rep3;
+				s->lzma.rep3 = s->lzma.rep2;
+			}
+
+			s->lzma.rep2 = s->lzma.rep1;
+		}
+
+		s->lzma.rep1 = s->lzma.rep0;
+		s->lzma.rep0 = tmp;
+	}
+
+	lzma_state_long_rep(&s->lzma.state);
+	lzma_len(s, &s->lzma.rep_len_dec, pos_state);
+}
+
+/* LZMA decoder core */
+static bool lzma_main(struct xz_dec_lzma2 *s)
+{
+	uint32_t pos_state;
+
+	/*
+	 * If the dictionary was reached during the previous call, try to
+	 * finish the possibly pending repeat in the dictionary.
+	 */
+	if (dict_has_space(&s->dict) && s->lzma.len > 0)
+		dict_repeat(&s->dict, &s->lzma.len, s->lzma.rep0);
+
+	/*
+	 * Decode more LZMA symbols. One iteration may consume up to
+	 * LZMA_IN_REQUIRED - 1 bytes.
+	 */
+	while (dict_has_space(&s->dict) && !rc_limit_exceeded(&s->rc)) {
+		pos_state = s->dict.pos & s->lzma.pos_mask;
+
+		if (!rc_bit(&s->rc, &s->lzma.is_match[
+				s->lzma.state][pos_state])) {
+			lzma_literal(s);
+		} else {
+			if (rc_bit(&s->rc, &s->lzma.is_rep[s->lzma.state]))
+				lzma_rep_match(s, pos_state);
+			else
+				lzma_match(s, pos_state);
+
+			if (!dict_repeat(&s->dict, &s->lzma.len, s->lzma.rep0))
+				return false;
+		}
+	}
+
+	/*
+	 * Having the range decoder always normalized when we are outside
+	 * this function makes it easier to correctly handle end of the chunk.
+	 */
+	rc_normalize(&s->rc);
+
+	return true;
+}
+
+/*
+ * Reset the LZMA decoder and range decoder state. Dictionary is nore reset
+ * here, because LZMA state may be reset without resetting the dictionary.
+ */
+static void lzma_reset(struct xz_dec_lzma2 *s)
+{
+	uint16_t *probs;
+	size_t i;
+
+	s->lzma.state = STATE_LIT_LIT;
+	s->lzma.rep0 = 0;
+	s->lzma.rep1 = 0;
+	s->lzma.rep2 = 0;
+	s->lzma.rep3 = 0;
+
+	/*
+	 * All probabilities are initialized to the same value. This hack
+	 * makes the code smaller by avoiding a separate loop for each
+	 * probability array.
+	 *
+	 * This could be optimized so that only that part of literal
+	 * probabilities that are actually required. In the common case
+	 * we would write 12 KiB less.
+	 */
+	probs = s->lzma.is_match[0];
+	for (i = 0; i < PROBS_TOTAL; ++i)
+		probs[i] = RC_BIT_MODEL_TOTAL / 2;
+
+	rc_reset(&s->rc);
+}
+
+/*
+ * Decode and validate LZMA properties (lc/lp/pb) and calculate the bit masks
+ * from the decoded lp and pb values. On success, the LZMA decoder state is
+ * reset and true is returned.
+ */
+static bool lzma_props(struct xz_dec_lzma2 *s, uint8_t props)
+{
+	if (props > (4 * 5 + 4) * 9 + 8)
+		return false;
+
+	s->lzma.pos_mask = 0;
+	while (props >= 9 * 5) {
+		props -= 9 * 5;
+		++s->lzma.pos_mask;
+	}
+
+	s->lzma.pos_mask = (1 << s->lzma.pos_mask) - 1;
+
+	s->lzma.literal_pos_mask = 0;
+	while (props >= 9) {
+		props -= 9;
+		++s->lzma.literal_pos_mask;
+	}
+
+	s->lzma.lc = props;
+
+	if (s->lzma.lc + s->lzma.literal_pos_mask > 4)
+		return false;
+
+	s->lzma.literal_pos_mask = (1 << s->lzma.literal_pos_mask) - 1;
+
+	lzma_reset(s);
+
+	return true;
+}
+
+/*********
+ * LZMA2 *
+ *********/
+
+/*
+ * The LZMA decoder assumes that if the input limit (s->rc.in_limit) hasn't
+ * been exceeded, it is safe to read up to LZMA_IN_REQUIRED bytes. This
+ * wrapper function takes care of making the LZMA decoder's assumption safe.
+ *
+ * As long as there is plenty of input left to be decoded in the current LZMA
+ * chunk, we decode directly from the caller-supplied input buffer until
+ * there's LZMA_IN_REQUIRED bytes left. Those remaining bytes are copied into
+ * s->temp.buf, which (hopefully) gets filled on the next call to this
+ * function. We decode a few bytes from the temporary buffer so that we can
+ * continue decoding from the caller-supplied input buffer again.
+ */
+static bool lzma2_lzma(struct xz_dec_lzma2 *s, struct xz_buf *b)
+{
+	size_t in_avail;
+	uint32_t tmp;
+
+	in_avail = b->in_size - b->in_pos;
+	if (s->temp.size > 0 || s->lzma2.compressed == 0) {
+		tmp = 2 * LZMA_IN_REQUIRED - s->temp.size;
+		if (tmp > s->lzma2.compressed - s->temp.size)
+			tmp = s->lzma2.compressed - s->temp.size;
+		if (tmp > in_avail)
+			tmp = in_avail;
+
+		memcpy(s->temp.buf + s->temp.size, b->in + b->in_pos, tmp);
+
+		if (s->temp.size + tmp == s->lzma2.compressed) {
+			memzero(s->temp.buf + s->temp.size + tmp,
+					sizeof(s->temp.buf)
+						- s->temp.size - tmp);
+			s->rc.in_limit = s->temp.size + tmp;
+		} else if (s->temp.size + tmp < LZMA_IN_REQUIRED) {
+			s->temp.size += tmp;
+			b->in_pos += tmp;
+			return true;
+		} else {
+			s->rc.in_limit = s->temp.size + tmp - LZMA_IN_REQUIRED;
+		}
+
+		s->rc.in = s->temp.buf;
+		s->rc.in_pos = 0;
+
+		if (!lzma_main(s) || s->rc.in_pos > s->temp.size + tmp)
+			return false;
+
+		s->lzma2.compressed -= s->rc.in_pos;
+
+		if (s->rc.in_pos < s->temp.size) {
+			s->temp.size -= s->rc.in_pos;
+			memmove(s->temp.buf, s->temp.buf + s->rc.in_pos,
+					s->temp.size);
+			return true;
+		}
+
+		b->in_pos += s->rc.in_pos - s->temp.size;
+		s->temp.size = 0;
+	}
+
+	in_avail = b->in_size - b->in_pos;
+	if (in_avail >= LZMA_IN_REQUIRED) {
+		s->rc.in = b->in;
+		s->rc.in_pos = b->in_pos;
+
+		if (in_avail >= s->lzma2.compressed + LZMA_IN_REQUIRED)
+			s->rc.in_limit = b->in_pos + s->lzma2.compressed;
+		else
+			s->rc.in_limit = b->in_size - LZMA_IN_REQUIRED;
+
+		if (!lzma_main(s))
+			return false;
+
+		in_avail = s->rc.in_pos - b->in_pos;
+		if (in_avail > s->lzma2.compressed)
+			return false;
+
+		s->lzma2.compressed -= in_avail;
+		b->in_pos = s->rc.in_pos;
+	}
+
+	in_avail = b->in_size - b->in_pos;
+	if (in_avail < LZMA_IN_REQUIRED) {
+		if (in_avail > s->lzma2.compressed)
+			in_avail = s->lzma2.compressed;
+
+		memcpy(s->temp.buf, b->in + b->in_pos, in_avail);
+		s->temp.size = in_avail;
+		b->in_pos += in_avail;
+	}
+
+	return true;
+}
+
+/*
+ * Take care of the LZMA2 control layer, and forward the job of actual LZMA
+ * decoding or copying of uncompressed chunks to other functions.
+ */
+XZ_EXTERN enum xz_ret xz_dec_lzma2_run(struct xz_dec_lzma2 *s,
+				       struct xz_buf *b)
+{
+	uint32_t tmp;
+
+	while (b->in_pos < b->in_size || s->lzma2.sequence == SEQ_LZMA_RUN) {
+		switch (s->lzma2.sequence) {
+		case SEQ_CONTROL:
+			/*
+			 * LZMA2 control byte
+			 *
+			 * Exact values:
+			 *   0x00   End marker
+			 *   0x01   Dictionary reset followed by
+			 *          an uncompressed chunk
+			 *   0x02   Uncompressed chunk (no dictionary reset)
+			 *
+			 * Highest three bits (s->control & 0xE0):
+			 *   0xE0   Dictionary reset, new properties and state
+			 *          reset, followed by LZMA compressed chunk
+			 *   0xC0   New properties and state reset, followed
+			 *          by LZMA compressed chunk (no dictionary
+			 *          reset)
+			 *   0xA0   State reset using old properties,
+			 *          followed by LZMA compressed chunk (no
+			 *          dictionary reset)
+			 *   0x80   LZMA chunk (no dictionary or state reset)
+			 *
+			 * For LZMA compressed chunks, the lowest five bits
+			 * (s->control & 1F) are the highest bits of the
+			 * uncompressed size (bits 16-20).
+			 *
+			 * A new LZMA2 stream must begin with a dictionary
+			 * reset. The first LZMA chunk must set new
+			 * properties and reset the LZMA state.
+			 *
+			 * Values that don't match anything described above
+			 * are invalid and we return XZ_DATA_ERROR.
+			 */
+			tmp = b->in[b->in_pos++];
+
+			if (tmp >= 0xE0 || tmp == 0x01) {
+				s->lzma2.need_props = true;
+				s->lzma2.need_dict_reset = false;
+				dict_reset(&s->dict, b);
+			} else if (s->lzma2.need_dict_reset) {
+				return XZ_DATA_ERROR;
+			}
+
+			if (tmp >= 0x80) {
+				s->lzma2.uncompressed = (tmp & 0x1F) << 16;
+				s->lzma2.sequence = SEQ_UNCOMPRESSED_1;
+
+				if (tmp >= 0xC0) {
+					/*
+					 * When there are new properties,
+					 * state reset is done at
+					 * SEQ_PROPERTIES.
+					 */
+					s->lzma2.need_props = false;
+					s->lzma2.next_sequence
+							= SEQ_PROPERTIES;
+
+				} else if (s->lzma2.need_props) {
+					return XZ_DATA_ERROR;
+
+				} else {
+					s->lzma2.next_sequence
+							= SEQ_LZMA_PREPARE;
+					if (tmp >= 0xA0)
+						lzma_reset(s);
+				}
+			} else {
+				if (tmp == 0x00)
+					return XZ_STREAM_END;
+
+				if (tmp > 0x02)
+					return XZ_DATA_ERROR;
+
+				s->lzma2.sequence = SEQ_COMPRESSED_0;
+				s->lzma2.next_sequence = SEQ_COPY;
+			}
+
+			break;
+
+		case SEQ_UNCOMPRESSED_1:
+			s->lzma2.uncompressed
+					+= (uint32_t)b->in[b->in_pos++] << 8;
+			s->lzma2.sequence = SEQ_UNCOMPRESSED_2;
+			break;
+
+		case SEQ_UNCOMPRESSED_2:
+			s->lzma2.uncompressed
+					+= (uint32_t)b->in[b->in_pos++] + 1;
+			s->lzma2.sequence = SEQ_COMPRESSED_0;
+			break;
+
+		case SEQ_COMPRESSED_0:
+			s->lzma2.compressed
+					= (uint32_t)b->in[b->in_pos++] << 8;
+			s->lzma2.sequence = SEQ_COMPRESSED_1;
+			break;
+
+		case SEQ_COMPRESSED_1:
+			s->lzma2.compressed
+					+= (uint32_t)b->in[b->in_pos++] + 1;
+			s->lzma2.sequence = s->lzma2.next_sequence;
+			break;
+
+		case SEQ_PROPERTIES:
+			if (!lzma_props(s, b->in[b->in_pos++]))
+				return XZ_DATA_ERROR;
+
+			s->lzma2.sequence = SEQ_LZMA_PREPARE;
+
+		case SEQ_LZMA_PREPARE:
+			if (s->lzma2.compressed < RC_INIT_BYTES)
+				return XZ_DATA_ERROR;
+
+			if (!rc_read_init(&s->rc, b))
+				return XZ_OK;
+
+			s->lzma2.compressed -= RC_INIT_BYTES;
+			s->lzma2.sequence = SEQ_LZMA_RUN;
+
+		case SEQ_LZMA_RUN:
+			/*
+			 * Set dictionary limit to indicate how much we want
+			 * to be encoded at maximum. Decode new data into the
+			 * dictionary. Flush the new data from dictionary to
+			 * b->out. Check if we finished decoding this chunk.
+			 * In case the dictionary got full but we didn't fill
+			 * the output buffer yet, we may run this loop
+			 * multiple times without changing s->lzma2.sequence.
+			 */
+			dict_limit(&s->dict, min_t(size_t,
+					b->out_size - b->out_pos,
+					s->lzma2.uncompressed));
+			if (!lzma2_lzma(s, b))
+				return XZ_DATA_ERROR;
+
+			s->lzma2.uncompressed -= dict_flush(&s->dict, b);
+
+			if (s->lzma2.uncompressed == 0) {
+				if (s->lzma2.compressed > 0 || s->lzma.len > 0
+						|| !rc_is_finished(&s->rc))
+					return XZ_DATA_ERROR;
+
+				rc_reset(&s->rc);
+				s->lzma2.sequence = SEQ_CONTROL;
+
+			} else if (b->out_pos == b->out_size
+					|| (b->in_pos == b->in_size
+						&& s->temp.size
+						< s->lzma2.compressed)) {
+				return XZ_OK;
+			}
+
+			break;
+
+		case SEQ_COPY:
+			dict_uncompressed(&s->dict, b, &s->lzma2.compressed);
+			if (s->lzma2.compressed > 0)
+				return XZ_OK;
+
+			s->lzma2.sequence = SEQ_CONTROL;
+			break;
+		}
+	}
+
+	return XZ_OK;
+}
+
+XZ_EXTERN struct xz_dec_lzma2 *xz_dec_lzma2_create(enum xz_mode mode,
+						   uint32_t dict_max)
+{
+	struct xz_dec_lzma2 *s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (s == NULL)
+		return NULL;
+
+	s->dict.mode = mode;
+	s->dict.size_max = dict_max;
+
+	if (DEC_IS_PREALLOC(mode)) {
+		s->dict.buf = vmalloc(dict_max);
+		if (s->dict.buf == NULL) {
+			kfree(s);
+			return NULL;
+		}
+	} else if (DEC_IS_DYNALLOC(mode)) {
+		s->dict.buf = NULL;
+		s->dict.allocated = 0;
+	}
+
+	return s;
+}
+
+XZ_EXTERN enum xz_ret xz_dec_lzma2_reset(struct xz_dec_lzma2 *s, uint8_t props)
+{
+	/* This limits dictionary size to 3 GiB to keep parsing simpler. */
+	if (props > 39)
+		return XZ_OPTIONS_ERROR;
+
+	s->dict.size = 2 + (props & 1);
+	s->dict.size <<= (props >> 1) + 11;
+
+	if (DEC_IS_MULTI(s->dict.mode)) {
+		if (s->dict.size > s->dict.size_max)
+			return XZ_MEMLIMIT_ERROR;
+
+		s->dict.end = s->dict.size;
+
+		if (DEC_IS_DYNALLOC(s->dict.mode)) {
+			if (s->dict.allocated < s->dict.size) {
+				vfree(s->dict.buf);
+				s->dict.buf = vmalloc(s->dict.size);
+				if (s->dict.buf == NULL) {
+					s->dict.allocated = 0;
+					return XZ_MEM_ERROR;
+				}
+			}
+		}
+	}
+
+	s->lzma.len = 0;
+
+	s->lzma2.sequence = SEQ_CONTROL;
+	s->lzma2.need_dict_reset = true;
+
+	s->temp.size = 0;
+
+	return XZ_OK;
+}
+
+XZ_EXTERN void xz_dec_lzma2_end(struct xz_dec_lzma2 *s)
+{
+	if (DEC_IS_MULTI(s->dict.mode))
+		vfree(s->dict.buf);
+
+	kfree(s);
+}
+/*
+ * .xz Stream decoder
+ *
+ * Author: Lasse Collin <lasse.collin@tukaani.org>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#include "xz_private.h"
+#include "xz_stream.h"
+
+/* Hash used to validate the Index field */
+struct xz_dec_hash {
+	vli_type unpadded;
+	vli_type uncompressed;
+	uint32_t crc32;
+};
+
+struct xz_dec {
+	/* Position in dec_main() */
+	enum {
+		SEQ_STREAM_HEADER,
+		SEQ_BLOCK_START,
+		SEQ_BLOCK_HEADER,
+		SEQ_BLOCK_UNCOMPRESS,
+		SEQ_BLOCK_PADDING,
+		SEQ_BLOCK_CHECK,
+		SEQ_INDEX,
+		SEQ_INDEX_PADDING,
+		SEQ_INDEX_CRC32,
+		SEQ_STREAM_FOOTER
+	} sequence;
+
+	/* Position in variable-length integers and Check fields */
+	uint32_t pos;
+
+	/* Variable-length integer decoded by dec_vli() */
+	vli_type vli;
+
+	/* Saved in_pos and out_pos */
+	size_t in_start;
+	size_t out_start;
+
+	/* CRC32 value in Block or Index */
+	uint32_t crc32;
+
+	/* Type of the integrity check calculated from uncompressed data */
+	enum xz_check check_type;
+
+	/* Operation mode */
+	enum xz_mode mode;
+
+	/*
+	 * True if the next call to xz_dec_run() is allowed to return
+	 * XZ_BUF_ERROR.
+	 */
+	bool allow_buf_error;
+
+	/* Information stored in Block Header */
+	struct {
+		/*
+		 * Value stored in the Compressed Size field, or
+		 * VLI_UNKNOWN if Compressed Size is not present.
+		 */
+		vli_type compressed;
+
+		/*
+		 * Value stored in the Uncompressed Size field, or
+		 * VLI_UNKNOWN if Uncompressed Size is not present.
+		 */
+		vli_type uncompressed;
+
+		/* Size of the Block Header field */
+		uint32_t size;
+	} block_header;
+
+	/* Information collected when decoding Blocks */
+	struct {
+		/* Observed compressed size of the current Block */
+		vli_type compressed;
+
+		/* Observed uncompressed size of the current Block */
+		vli_type uncompressed;
+
+		/* Number of Blocks decoded so far */
+		vli_type count;
+
+		/*
+		 * Hash calculated from the Block sizes. This is used to
+		 * validate the Index field.
+		 */
+		struct xz_dec_hash hash;
+	} block;
+
+	/* Variables needed when verifying the Index field */
+	struct {
+		/* Position in dec_index() */
+		enum {
+			SEQ_INDEX_COUNT,
+			SEQ_INDEX_UNPADDED,
+			SEQ_INDEX_UNCOMPRESSED
+		} sequence;
+
+		/* Size of the Index in bytes */
+		vli_type size;
+
+		/* Number of Records (matches block.count in valid files) */
+		vli_type count;
+
+		/*
+		 * Hash calculated from the Records (matches block.hash in
+		 * valid files).
+		 */
+		struct xz_dec_hash hash;
+	} index;
+
+	/*
+	 * Temporary buffer needed to hold Stream Header, Block Header,
+	 * and Stream Footer. The Block Header is the biggest (1 KiB)
+	 * so we reserve space according to that. buf[] has to be aligned
+	 * to a multiple of four bytes; the size_t variables before it
+	 * should guarantee this.
+	 */
+	struct {
+		size_t pos;
+		size_t size;
+		uint8_t buf[1024];
+	} temp;
+
+	struct xz_dec_lzma2 *lzma2;
+
+#ifdef XZ_DEC_BCJ
+	struct xz_dec_bcj *bcj;
+	bool bcj_active;
+#endif
+};
+
+#ifdef XZ_DEC_ANY_CHECK
+/* Sizes of the Check field with different Check IDs */
+static const uint8_t check_sizes[16] = {
+	0,
+	4, 4, 4,
+	8, 8, 8,
+	16, 16, 16,
+	32, 32, 32,
+	64, 64, 64
+};
+#endif
+
+/*
+ * Fill s->temp by copying data starting from b->in[b->in_pos]. Caller
+ * must have set s->temp.pos to indicate how much data we are supposed
+ * to copy into s->temp.buf. Return true once s->temp.pos has reached
+ * s->temp.size.
+ */
+static bool fill_temp(struct xz_dec *s, struct xz_buf *b)
+{
+	size_t copy_size = min_t(size_t,
+			b->in_size - b->in_pos, s->temp.size - s->temp.pos);
+
+	memcpy(s->temp.buf + s->temp.pos, b->in + b->in_pos, copy_size);
+	b->in_pos += copy_size;
+	s->temp.pos += copy_size;
+
+	if (s->temp.pos == s->temp.size) {
+		s->temp.pos = 0;
+		return true;
+	}
+
+	return false;
+}
+
+/* Decode a variable-length integer (little-endian base-128 encoding) */
+static enum xz_ret dec_vli(struct xz_dec *s, const uint8_t *in,
+			   size_t *in_pos, size_t in_size)
+{
+	uint8_t byte;
+
+	if (s->pos == 0)
+		s->vli = 0;
+
+	while (*in_pos < in_size) {
+		byte = in[*in_pos];
+		++*in_pos;
+
+		s->vli |= (vli_type)(byte & 0x7F) << s->pos;
+
+		if ((byte & 0x80) == 0) {
+			/* Don't allow non-minimal encodings. */
+			if (byte == 0 && s->pos != 0)
+				return XZ_DATA_ERROR;
+
+			s->pos = 0;
+			return XZ_STREAM_END;
+		}
+
+		s->pos += 7;
+		if (s->pos == 7 * VLI_BYTES_MAX)
+			return XZ_DATA_ERROR;
+	}
+
+	return XZ_OK;
+}
+
+/*
+ * Decode the Compressed Data field from a Block. Update and validate
+ * the observed compressed and uncompressed sizes of the Block so that
+ * they don't exceed the values possibly stored in the Block Header
+ * (validation assumes that no integer overflow occurs, since vli_type
+ * is normally uint64_t). Update the CRC32 if presence of the CRC32
+ * field was indicated in Stream Header.
+ *
+ * Once the decoding is finished, validate that the observed sizes match
+ * the sizes possibly stored in the Block Header. Update the hash and
+ * Block count, which are later used to validate the Index field.
+ */
+static enum xz_ret dec_block(struct xz_dec *s, struct xz_buf *b)
+{
+	enum xz_ret ret;
+
+	s->in_start = b->in_pos;
+	s->out_start = b->out_pos;
+
+#ifdef XZ_DEC_BCJ
+	if (s->bcj_active)
+		ret = xz_dec_bcj_run(s->bcj, s->lzma2, b);
+	else
+#endif
+		ret = xz_dec_lzma2_run(s->lzma2, b);
+
+	s->block.compressed += b->in_pos - s->in_start;
+	s->block.uncompressed += b->out_pos - s->out_start;
+
+	/*
+	 * There is no need to separately check for VLI_UNKNOWN, since
+	 * the observed sizes are always smaller than VLI_UNKNOWN.
+	 */
+	if (s->block.compressed > s->block_header.compressed
+			|| s->block.uncompressed
+				> s->block_header.uncompressed)
+		return XZ_DATA_ERROR;
+
+	if (s->check_type == XZ_CHECK_CRC32)
+		s->crc32 = xz_crc32(b->out + s->out_start,
+				b->out_pos - s->out_start, s->crc32);
+
+	if (ret == XZ_STREAM_END) {
+		if (s->block_header.compressed != VLI_UNKNOWN
+				&& s->block_header.compressed
+					!= s->block.compressed)
+			return XZ_DATA_ERROR;
+
+		if (s->block_header.uncompressed != VLI_UNKNOWN
+				&& s->block_header.uncompressed
+					!= s->block.uncompressed)
+			return XZ_DATA_ERROR;
+
+		s->block.hash.unpadded += s->block_header.size
+				+ s->block.compressed;
+
+#ifdef XZ_DEC_ANY_CHECK
+		s->block.hash.unpadded += check_sizes[s->check_type];
+#else
+		if (s->check_type == XZ_CHECK_CRC32)
+			s->block.hash.unpadded += 4;
+#endif
+
+		s->block.hash.uncompressed += s->block.uncompressed;
+		s->block.hash.crc32 = xz_crc32(
+				(const uint8_t *)&s->block.hash,
+				sizeof(s->block.hash), s->block.hash.crc32);
+
+		++s->block.count;
+	}
+
+	return ret;
+}
+
+/* Update the Index size and the CRC32 value. */
+static void index_update(struct xz_dec *s, const struct xz_buf *b)
+{
+	size_t in_used = b->in_pos - s->in_start;
+	s->index.size += in_used;
+	s->crc32 = xz_crc32(b->in + s->in_start, in_used, s->crc32);
+}
+
+/*
+ * Decode the Number of Records, Unpadded Size, and Uncompressed Size
+ * fields from the Index field. That is, Index Padding and CRC32 are not
+ * decoded by this function.
+ *
+ * This can return XZ_OK (more input needed), XZ_STREAM_END (everything
+ * successfully decoded), or XZ_DATA_ERROR (input is corrupt).
+ */
+static enum xz_ret dec_index(struct xz_dec *s, struct xz_buf *b)
+{
+	enum xz_ret ret;
+
+	do {
+		ret = dec_vli(s, b->in, &b->in_pos, b->in_size);
+		if (ret != XZ_STREAM_END) {
+			index_update(s, b);
+			return ret;
+		}
+
+		switch (s->index.sequence) {
+		case SEQ_INDEX_COUNT:
+			s->index.count = s->vli;
+
+			/*
+			 * Validate that the Number of Records field
+			 * indicates the same number of Records as
+			 * there were Blocks in the Stream.
+			 */
+			if (s->index.count != s->block.count)
+				return XZ_DATA_ERROR;
+
+			s->index.sequence = SEQ_INDEX_UNPADDED;
+			break;
+
+		case SEQ_INDEX_UNPADDED:
+			s->index.hash.unpadded += s->vli;
+			s->index.sequence = SEQ_INDEX_UNCOMPRESSED;
+			break;
+
+		case SEQ_INDEX_UNCOMPRESSED:
+			s->index.hash.uncompressed += s->vli;
+			s->index.hash.crc32 = xz_crc32(
+					(const uint8_t *)&s->index.hash,
+					sizeof(s->index.hash),
+					s->index.hash.crc32);
+			--s->index.count;
+			s->index.sequence = SEQ_INDEX_UNPADDED;
+			break;
+		}
+	} while (s->index.count > 0);
+
+	return XZ_STREAM_END;
+}
+
+/*
+ * Validate that the next four input bytes match the value of s->crc32.
+ * s->pos must be zero when starting to validate the first byte.
+ */
+static enum xz_ret crc32_validate(struct xz_dec *s, struct xz_buf *b)
+{
+	do {
+		if (b->in_pos == b->in_size)
+			return XZ_OK;
+
+		if (((s->crc32 >> s->pos) & 0xFF) != b->in[b->in_pos++])
+			return XZ_DATA_ERROR;
+
+		s->pos += 8;
+
+	} while (s->pos < 32);
+
+	s->crc32 = 0;
+	s->pos = 0;
+
+	return XZ_STREAM_END;
+}
+
+#ifdef XZ_DEC_ANY_CHECK
+/*
+ * Skip over the Check field when the Check ID is not supported.
+ * Returns true once the whole Check field has been skipped over.
+ */
+static bool check_skip(struct xz_dec *s, struct xz_buf *b)
+{
+	while (s->pos < check_sizes[s->check_type]) {
+		if (b->in_pos == b->in_size)
+			return false;
+
+		++b->in_pos;
+		++s->pos;
+	}
+
+	s->pos = 0;
+
+	return true;
+}
+#endif
+
+/* Decode the Stream Header field (the first 12 bytes of the .xz Stream). */
+static enum xz_ret dec_stream_header(struct xz_dec *s)
+{
+	if (!memeq(s->temp.buf, HEADER_MAGIC, HEADER_MAGIC_SIZE))
+		return XZ_FORMAT_ERROR;
+
+	if (xz_crc32(s->temp.buf + HEADER_MAGIC_SIZE, 2, 0)
+			!= get_le32(s->temp.buf + HEADER_MAGIC_SIZE + 2))
+		return XZ_DATA_ERROR;
+
+	if (s->temp.buf[HEADER_MAGIC_SIZE] != 0)
+		return XZ_OPTIONS_ERROR;
+
+	/*
+	 * Of integrity checks, we support only none (Check ID = 0) and
+	 * CRC32 (Check ID = 1). However, if XZ_DEC_ANY_CHECK is defined,
+	 * we will accept other check types too, but then the check won't
+	 * be verified and a warning (XZ_UNSUPPORTED_CHECK) will be given.
+	 */
+	s->check_type = s->temp.buf[HEADER_MAGIC_SIZE + 1];
+
+#ifdef XZ_DEC_ANY_CHECK
+	if (s->check_type > XZ_CHECK_MAX)
+		return XZ_OPTIONS_ERROR;
+
+	if (s->check_type > XZ_CHECK_CRC32)
+		return XZ_UNSUPPORTED_CHECK;
+#else
+	if (s->check_type > XZ_CHECK_CRC32)
+		return XZ_OPTIONS_ERROR;
+#endif
+
+	return XZ_OK;
+}
+
+/* Decode the Stream Footer field (the last 12 bytes of the .xz Stream) */
+static enum xz_ret dec_stream_footer(struct xz_dec *s)
+{
+	if (!memeq(s->temp.buf + 10, FOOTER_MAGIC, FOOTER_MAGIC_SIZE))
+		return XZ_DATA_ERROR;
+
+	if (xz_crc32(s->temp.buf + 4, 6, 0) != get_le32(s->temp.buf))
+		return XZ_DATA_ERROR;
+
+	/*
+	 * Validate Backward Size. Note that we never added the size of the
+	 * Index CRC32 field to s->index.size, thus we use s->index.size / 4
+	 * instead of s->index.size / 4 - 1.
+	 */
+	if ((s->index.size >> 2) != get_le32(s->temp.buf + 4))
+		return XZ_DATA_ERROR;
+
+	if (s->temp.buf[8] != 0 || s->temp.buf[9] != s->check_type)
+		return XZ_DATA_ERROR;
+
+	/*
+	 * Use XZ_STREAM_END instead of XZ_OK to be more convenient
+	 * for the caller.
+	 */
+	return XZ_STREAM_END;
+}
+
+/* Decode the Block Header and initialize the filter chain. */
+static enum xz_ret dec_block_header(struct xz_dec *s)
+{
+	enum xz_ret ret;
+
+	/*
+	 * Validate the CRC32. We know that the temp buffer is at least
+	 * eight bytes so this is safe.
+	 */
+	s->temp.size -= 4;
+	if (xz_crc32(s->temp.buf, s->temp.size, 0)
+			!= get_le32(s->temp.buf + s->temp.size))
+		return XZ_DATA_ERROR;
+
+	s->temp.pos = 2;
+
+	/*
+	 * Catch unsupported Block Flags. We support only one or two filters
+	 * in the chain, so we catch that with the same test.
+	 */
+#ifdef XZ_DEC_BCJ
+	if (s->temp.buf[1] & 0x3E)
+#else
+	if (s->temp.buf[1] & 0x3F)
+#endif
+		return XZ_OPTIONS_ERROR;
+
+	/* Compressed Size */
+	if (s->temp.buf[1] & 0x40) {
+		if (dec_vli(s, s->temp.buf, &s->temp.pos, s->temp.size)
+					!= XZ_STREAM_END)
+			return XZ_DATA_ERROR;
+
+		s->block_header.compressed = s->vli;
+	} else {
+		s->block_header.compressed = VLI_UNKNOWN;
+	}
+
+	/* Uncompressed Size */
+	if (s->temp.buf[1] & 0x80) {
+		if (dec_vli(s, s->temp.buf, &s->temp.pos, s->temp.size)
+				!= XZ_STREAM_END)
+			return XZ_DATA_ERROR;
+
+		s->block_header.uncompressed = s->vli;
+	} else {
+		s->block_header.uncompressed = VLI_UNKNOWN;
+	}
+
+#ifdef XZ_DEC_BCJ
+	/* If there are two filters, the first one must be a BCJ filter. */
+	s->bcj_active = s->temp.buf[1] & 0x01;
+	if (s->bcj_active) {
+		if (s->temp.size - s->temp.pos < 2)
+			return XZ_OPTIONS_ERROR;
+
+		ret = xz_dec_bcj_reset(s->bcj, s->temp.buf[s->temp.pos++]);
+		if (ret != XZ_OK)
+			return ret;
+
+		/*
+		 * We don't support custom start offset,
+		 * so Size of Properties must be zero.
+		 */
+		if (s->temp.buf[s->temp.pos++] != 0x00)
+			return XZ_OPTIONS_ERROR;
+	}
+#endif
+
+	/* Valid Filter Flags always take at least two bytes. */
+	if (s->temp.size - s->temp.pos < 2)
+		return XZ_DATA_ERROR;
+
+	/* Filter ID = LZMA2 */
+	if (s->temp.buf[s->temp.pos++] != 0x21)
+		return XZ_OPTIONS_ERROR;
+
+	/* Size of Properties = 1-byte Filter Properties */
+	if (s->temp.buf[s->temp.pos++] != 0x01)
+		return XZ_OPTIONS_ERROR;
+
+	/* Filter Properties contains LZMA2 dictionary size. */
+	if (s->temp.size - s->temp.pos < 1)
+		return XZ_DATA_ERROR;
+
+	ret = xz_dec_lzma2_reset(s->lzma2, s->temp.buf[s->temp.pos++]);
+	if (ret != XZ_OK)
+		return ret;
+
+	/* The rest must be Header Padding. */
+	while (s->temp.pos < s->temp.size)
+		if (s->temp.buf[s->temp.pos++] != 0x00)
+			return XZ_OPTIONS_ERROR;
+
+	s->temp.pos = 0;
+	s->block.compressed = 0;
+	s->block.uncompressed = 0;
+
+	return XZ_OK;
+}
+
+static enum xz_ret dec_main(struct xz_dec *s, struct xz_buf *b)
+{
+	enum xz_ret ret;
+
+	/*
+	 * Store the start position for the case when we are in the middle
+	 * of the Index field.
+	 */
+	s->in_start = b->in_pos;
+
+	while (true) {
+		switch (s->sequence) {
+		case SEQ_STREAM_HEADER:
+			/*
+			 * Stream Header is copied to s->temp, and then
+			 * decoded from there. This way if the caller
+			 * gives us only little input at a time, we can
+			 * still keep the Stream Header decoding code
+			 * simple. Similar approach is used in many places
+			 * in this file.
+			 */
+			if (!fill_temp(s, b))
+				return XZ_OK;
+
+			/*
+			 * If dec_stream_header() returns
+			 * XZ_UNSUPPORTED_CHECK, it is still possible
+			 * to continue decoding if working in multi-call
+			 * mode. Thus, update s->sequence before calling
+			 * dec_stream_header().
+			 */
+			s->sequence = SEQ_BLOCK_START;
+
+			ret = dec_stream_header(s);
+			if (ret != XZ_OK)
+				return ret;
+
+		case SEQ_BLOCK_START:
+			/* We need one byte of input to continue. */
+			if (b->in_pos == b->in_size)
+				return XZ_OK;
+
+			/* See if this is the beginning of the Index field. */
+			if (b->in[b->in_pos] == 0) {
+				s->in_start = b->in_pos++;
+				s->sequence = SEQ_INDEX;
+				break;
+			}
+
+			/*
+			 * Calculate the size of the Block Header and
+			 * prepare to decode it.
+			 */
+			s->block_header.size
+				= ((uint32_t)b->in[b->in_pos] + 1) * 4;
+
+			s->temp.size = s->block_header.size;
+			s->temp.pos = 0;
+			s->sequence = SEQ_BLOCK_HEADER;
+
+		case SEQ_BLOCK_HEADER:
+			if (!fill_temp(s, b))
+				return XZ_OK;
+
+			ret = dec_block_header(s);
+			if (ret != XZ_OK)
+				return ret;
+
+			s->sequence = SEQ_BLOCK_UNCOMPRESS;
+
+		case SEQ_BLOCK_UNCOMPRESS:
+			ret = dec_block(s, b);
+			if (ret != XZ_STREAM_END)
+				return ret;
+
+			s->sequence = SEQ_BLOCK_PADDING;
+
+		case SEQ_BLOCK_PADDING:
+			/*
+			 * Size of Compressed Data + Block Padding
+			 * must be a multiple of four. We don't need
+			 * s->block.compressed for anything else
+			 * anymore, so we use it here to test the size
+			 * of the Block Padding field.
+			 */
+			while (s->block.compressed & 3) {
+				if (b->in_pos == b->in_size)
+					return XZ_OK;
+
+				if (b->in[b->in_pos++] != 0)
+					return XZ_DATA_ERROR;
+
+				++s->block.compressed;
+			}
+
+			s->sequence = SEQ_BLOCK_CHECK;
+
+		case SEQ_BLOCK_CHECK:
+			if (s->check_type == XZ_CHECK_CRC32) {
+				ret = crc32_validate(s, b);
+				if (ret != XZ_STREAM_END)
+					return ret;
+			}
+#ifdef XZ_DEC_ANY_CHECK
+			else if (!check_skip(s, b)) {
+				return XZ_OK;
+			}
+#endif
+
+			s->sequence = SEQ_BLOCK_START;
+			break;
+
+		case SEQ_INDEX:
+			ret = dec_index(s, b);
+			if (ret != XZ_STREAM_END)
+				return ret;
+
+			s->sequence = SEQ_INDEX_PADDING;
+
+		case SEQ_INDEX_PADDING:
+			while ((s->index.size + (b->in_pos - s->in_start))
+					& 3) {
+				if (b->in_pos == b->in_size) {
+					index_update(s, b);
+					return XZ_OK;
+				}
+
+				if (b->in[b->in_pos++] != 0)
+					return XZ_DATA_ERROR;
+			}
+
+			/* Finish the CRC32 value and Index size. */
+			index_update(s, b);
+
+			/* Compare the hashes to validate the Index field. */
+			if (!memeq(&s->block.hash, &s->index.hash,
+					sizeof(s->block.hash)))
+				return XZ_DATA_ERROR;
+
+			s->sequence = SEQ_INDEX_CRC32;
+
+		case SEQ_INDEX_CRC32:
+			ret = crc32_validate(s, b);
+			if (ret != XZ_STREAM_END)
+				return ret;
+
+			s->temp.size = STREAM_HEADER_SIZE;
+			s->sequence = SEQ_STREAM_FOOTER;
+
+		case SEQ_STREAM_FOOTER:
+			if (!fill_temp(s, b))
+				return XZ_OK;
+
+			return dec_stream_footer(s);
+		}
+	}
+
+	/* Never reached */
+}
+
+/*
+ * xz_dec_run() is a wrapper for dec_main() to handle some special cases in
+ * multi-call and single-call decoding.
+ *
+ * In multi-call mode, we must return XZ_BUF_ERROR when it seems clear that we
+ * are not going to make any progress anymore. This is to prevent the caller
+ * from calling us infinitely when the input file is truncated or otherwise
+ * corrupt. Since zlib-style API allows that the caller fills the input buffer
+ * only when the decoder doesn't produce any new output, we have to be careful
+ * to avoid returning XZ_BUF_ERROR too easily: XZ_BUF_ERROR is returned only
+ * after the second consecutive call to xz_dec_run() that makes no progress.
+ *
+ * In single-call mode, if we couldn't decode everything and no error
+ * occurred, either the input is truncated or the output buffer is too small.
+ * Since we know that the last input byte never produces any output, we know
+ * that if all the input was consumed and decoding wasn't finished, the file
+ * must be corrupt. Otherwise the output buffer has to be too small or the
+ * file is corrupt in a way that decoding it produces too big output.
+ *
+ * If single-call decoding fails, we reset b->in_pos and b->out_pos back to
+ * their original values. This is because with some filter chains there won't
+ * be any valid uncompressed data in the output buffer unless the decoding
+ * actually succeeds (that's the price to pay of using the output buffer as
+ * the workspace).
+ */
+XZ_EXTERN enum xz_ret xz_dec_run(struct xz_dec *s, struct xz_buf *b)
+{
+	size_t in_start;
+	size_t out_start;
+	enum xz_ret ret;
+
+	if (DEC_IS_SINGLE(s->mode))
+		xz_dec_reset(s);
+
+	in_start = b->in_pos;
+	out_start = b->out_pos;
+	ret = dec_main(s, b);
+
+	if (DEC_IS_SINGLE(s->mode)) {
+		if (ret == XZ_OK)
+			ret = b->in_pos == b->in_size
+					? XZ_DATA_ERROR : XZ_BUF_ERROR;
+
+		if (ret != XZ_STREAM_END) {
+			b->in_pos = in_start;
+			b->out_pos = out_start;
+		}
+
+	} else if (ret == XZ_OK && in_start == b->in_pos
+			&& out_start == b->out_pos) {
+		if (s->allow_buf_error)
+			ret = XZ_BUF_ERROR;
+
+		s->allow_buf_error = true;
+	} else {
+		s->allow_buf_error = false;
+	}
+
+	return ret;
+}
+
+XZ_EXTERN struct xz_dec *xz_dec_init(enum xz_mode mode, uint32_t dict_max)
+{
+	struct xz_dec *s = kmalloc(sizeof(*s), GFP_KERNEL);
+	if (s == NULL)
+		return NULL;
+
+	s->mode = mode;
+
+#ifdef XZ_DEC_BCJ
+	s->bcj = xz_dec_bcj_create(DEC_IS_SINGLE(mode));
+	if (s->bcj == NULL)
+		goto error_bcj;
+#endif
+
+	s->lzma2 = xz_dec_lzma2_create(mode, dict_max);
+	if (s->lzma2 == NULL)
+		goto error_lzma2;
+
+	xz_dec_reset(s);
+	return s;
+
+error_lzma2:
+#ifdef XZ_DEC_BCJ
+	xz_dec_bcj_end(s->bcj);
+error_bcj:
+#endif
+	kfree(s);
+	return NULL;
+}
+
+XZ_EXTERN void xz_dec_reset(struct xz_dec *s)
+{
+	s->sequence = SEQ_STREAM_HEADER;
+	s->allow_buf_error = false;
+	s->pos = 0;
+	s->crc32 = 0;
+	memzero(&s->block, sizeof(s->block));
+	memzero(&s->index, sizeof(s->index));
+	s->temp.pos = 0;
+	s->temp.size = STREAM_HEADER_SIZE;
+}
+
+XZ_EXTERN void xz_dec_end(struct xz_dec *s)
+{
+	if (s != NULL) {
+		xz_dec_lzma2_end(s->lzma2);
+#ifdef XZ_DEC_BCJ
+		xz_dec_bcj_end(s->bcj);
+#endif
+		kfree(s);
+	}
+}
+/*
+ * XZ decoder module information
+ *
+ * Author: Lasse Collin <lasse.collin@tukaani.org>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#include <linux/module.h>
+#include <linux/xz.h>
+
+EXPORT_SYMBOL(xz_dec_init);
+EXPORT_SYMBOL(xz_dec_reset);
+EXPORT_SYMBOL(xz_dec_run);
+EXPORT_SYMBOL(xz_dec_end);
+
+MODULE_DESCRIPTION("XZ decompressor");
+MODULE_VERSION("1.0");
+MODULE_AUTHOR("Lasse Collin <lasse.collin@tukaani.org> and Igor Pavlov");
+
+/*
+ * This code is in the public domain, but in Linux it's simplest to just
+ * say it's GPL and consider the authors as the copyright holders.
+ */
+MODULE_LICENSE("GPL");
+/*
+ * XZ decoder tester
+ *
+ * Author: Lasse Collin <lasse.collin@tukaani.org>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/uaccess.h>
+#include <linux/crc32.h>
+#include <linux/xz.h>
+
+/* Maximum supported dictionary size */
+#define DICT_MAX (1 << 20)
+
+/* Device name to pass to register_chrdev(). */
+#define DEVICE_NAME "xz_dec_test"
+
+/* Dynamically allocated device major number */
+static int device_major;
+
+/*
+ * We reuse the same decoder state, and thus can decode only one
+ * file at a time.
+ */
+static bool device_is_open;
+
+/* XZ decoder state */
+static struct xz_dec *state;
+
+/*
+ * Return value of xz_dec_run(). We need to avoid calling xz_dec_run() after
+ * it has returned XZ_STREAM_END, so we make this static.
+ */
+static enum xz_ret ret;
+
+/*
+ * Input and output buffers. The input buffer is used as a temporary safe
+ * place for the data coming from the userspace.
+ */
+static uint8_t buffer_in[1024];
+static uint8_t buffer_out[1024];
+
+/*
+ * Structure to pass the input and output buffers to the XZ decoder.
+ * A few of the fields are never modified so we initialize them here.
+ */
+static struct xz_buf buffers = {
+	.in = buffer_in,
+	.out = buffer_out,
+	.out_size = sizeof(buffer_out)
+};
+
+/*
+ * CRC32 of uncompressed data. This is used to give the user a simple way
+ * to check that the decoder produces correct output.
+ */
+static uint32_t crc;
+
+static int xz_dec_test_open(struct inode *i, struct file *f)
+{
+	if (device_is_open)
+		return -EBUSY;
+
+	device_is_open = true;
+
+	xz_dec_reset(state);
+	ret = XZ_OK;
+	crc = 0xFFFFFFFF;
+
+	buffers.in_pos = 0;
+	buffers.in_size = 0;
+	buffers.out_pos = 0;
+
+	printk(KERN_INFO DEVICE_NAME ": opened\n");
+	return 0;
+}
+
+static int xz_dec_test_release(struct inode *i, struct file *f)
+{
+	device_is_open = false;
+
+	if (ret == XZ_OK)
+		printk(KERN_INFO DEVICE_NAME ": input was truncated\n");
+
+	printk(KERN_INFO DEVICE_NAME ": closed\n");
+	return 0;
+}
+
+/*
+ * Decode the data given to us from the userspace. CRC32 of the uncompressed
+ * data is calculated and is printed at the end of successful decoding. The
+ * uncompressed data isn't stored anywhere for further use.
+ *
+ * The .xz file must have exactly one Stream and no Stream Padding. The data
+ * after the first Stream is considered to be garbage.
+ */
+static ssize_t xz_dec_test_write(struct file *file, const char __user *buf,
+				 size_t size, loff_t *pos)
+{
+	size_t remaining;
+
+	if (ret != XZ_OK) {
+		if (size > 0)
+			printk(KERN_INFO DEVICE_NAME ": %zu bytes of "
+					"garbage at the end of the file\n",
+					size);
+
+		return -ENOSPC;
+	}
+
+	printk(KERN_INFO DEVICE_NAME ": decoding %zu bytes of input\n",
+			size);
+
+	remaining = size;
+	while ((remaining > 0 || buffers.out_pos == buffers.out_size)
+			&& ret == XZ_OK) {
+		if (buffers.in_pos == buffers.in_size) {
+			buffers.in_pos = 0;
+			buffers.in_size = min(remaining, sizeof(buffer_in));
+			if (copy_from_user(buffer_in, buf, buffers.in_size))
+				return -EFAULT;
+
+			buf += buffers.in_size;
+			remaining -= buffers.in_size;
+		}
+
+		buffers.out_pos = 0;
+		ret = xz_dec_run(state, &buffers);
+		crc = crc32(crc, buffer_out, buffers.out_pos);
+	}
+
+	switch (ret) {
+	case XZ_OK:
+		printk(KERN_INFO DEVICE_NAME ": XZ_OK\n");
+		return size;
+
+	case XZ_STREAM_END:
+		printk(KERN_INFO DEVICE_NAME ": XZ_STREAM_END, "
+				"CRC32 = 0x%08X\n", ~crc);
+		return size - remaining - (buffers.in_size - buffers.in_pos);
+
+	case XZ_MEMLIMIT_ERROR:
+		printk(KERN_INFO DEVICE_NAME ": XZ_MEMLIMIT_ERROR\n");
+		break;
+
+	case XZ_FORMAT_ERROR:
+		printk(KERN_INFO DEVICE_NAME ": XZ_FORMAT_ERROR\n");
+		break;
+
+	case XZ_OPTIONS_ERROR:
+		printk(KERN_INFO DEVICE_NAME ": XZ_OPTIONS_ERROR\n");
+		break;
+
+	case XZ_DATA_ERROR:
+		printk(KERN_INFO DEVICE_NAME ": XZ_DATA_ERROR\n");
+		break;
+
+	case XZ_BUF_ERROR:
+		printk(KERN_INFO DEVICE_NAME ": XZ_BUF_ERROR\n");
+		break;
+
+	default:
+		printk(KERN_INFO DEVICE_NAME ": Bug detected!\n");
+		break;
+	}
+
+	return -EIO;
+}
+
+/* Allocate the XZ decoder state and register the character device. */
+static int __init xz_dec_test_init(void)
+{
+	static const struct file_operations fileops = {
+		.owner = THIS_MODULE,
+		.open = &xz_dec_test_open,
+		.release = &xz_dec_test_release,
+		.write = &xz_dec_test_write
+	};
+
+	state = xz_dec_init(XZ_PREALLOC, DICT_MAX);
+	if (state == NULL)
+		return -ENOMEM;
+
+	device_major = register_chrdev(0, DEVICE_NAME, &fileops);
+	if (device_major < 0) {
+		xz_dec_end(state);
+		return device_major;
+	}
+
+	printk(KERN_INFO DEVICE_NAME ": module loaded\n");
+	printk(KERN_INFO DEVICE_NAME ": Create a device node with "
+			"'mknod " DEVICE_NAME " c %d 0' and write .xz files "
+			"to it.\n", device_major);
+	return 0;
+}
+
+static void __exit xz_dec_test_exit(void)
+{
+	unregister_chrdev(device_major, DEVICE_NAME);
+	xz_dec_end(state);
+	printk(KERN_INFO DEVICE_NAME ": module unloaded\n");
+}
+
+module_init(xz_dec_test_init);
+module_exit(xz_dec_test_exit);
+
+MODULE_DESCRIPTION("XZ decompressor tester");
+MODULE_VERSION("1.0");
+MODULE_AUTHOR("Lasse Collin <lasse.collin@tukaani.org>");
+
+/*
+ * This code is in the public domain, but in Linux it's simplest to just
+ * say it's GPL and consider the authors as the copyright holders.
+ */
+MODULE_LICENSE("GPL");
+/*
+ * LZMA2 definitions
+ *
+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
+ *          Igor Pavlov <http://7-zip.org/>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#ifndef XZ_LZMA2_H
+#define XZ_LZMA2_H
+
+/* Range coder constants */
+#define RC_SHIFT_BITS 8
+#define RC_TOP_BITS 24
+#define RC_TOP_VALUE (1 << RC_TOP_BITS)
+#define RC_BIT_MODEL_TOTAL_BITS 11
+#define RC_BIT_MODEL_TOTAL (1 << RC_BIT_MODEL_TOTAL_BITS)
+#define RC_MOVE_BITS 5
+
+/*
+ * Maximum number of position states. A position state is the lowest pb
+ * number of bits of the current uncompressed offset. In some places there
+ * are different sets of probabilities for different position states.
+ */
+#define POS_STATES_MAX (1 << 4)
+
+/*
+ * This enum is used to track which LZMA symbols have occurred most recently
+ * and in which order. This information is used to predict the next symbol.
+ *
+ * Symbols:
+ *  - Literal: One 8-bit byte
+ *  - Match: Repeat a chunk of data at some distance
+ *  - Long repeat: Multi-byte match at a recently seen distance
+ *  - Short repeat: One-byte repeat at a recently seen distance
+ *
+ * The symbol names are in from STATE_oldest_older_previous. REP means
+ * either short or long repeated match, and NONLIT means any non-literal.
+ */
+enum lzma_state {
+	STATE_LIT_LIT,
+	STATE_MATCH_LIT_LIT,
+	STATE_REP_LIT_LIT,
+	STATE_SHORTREP_LIT_LIT,
+	STATE_MATCH_LIT,
+	STATE_REP_LIT,
+	STATE_SHORTREP_LIT,
+	STATE_LIT_MATCH,
+	STATE_LIT_LONGREP,
+	STATE_LIT_SHORTREP,
+	STATE_NONLIT_MATCH,
+	STATE_NONLIT_REP
+};
+
+/* Total number of states */
+#define STATES 12
+
+/* The lowest 7 states indicate that the previous state was a literal. */
+#define LIT_STATES 7
+
+/* Indicate that the latest symbol was a literal. */
+static inline void lzma_state_literal(enum lzma_state *state)
+{
+	if (*state <= STATE_SHORTREP_LIT_LIT)
+		*state = STATE_LIT_LIT;
+	else if (*state <= STATE_LIT_SHORTREP)
+		*state -= 3;
+	else
+		*state -= 6;
+}
+
+/* Indicate that the latest symbol was a match. */
+static inline void lzma_state_match(enum lzma_state *state)
+{
+	*state = *state < LIT_STATES ? STATE_LIT_MATCH : STATE_NONLIT_MATCH;
+}
+
+/* Indicate that the latest state was a long repeated match. */
+static inline void lzma_state_long_rep(enum lzma_state *state)
+{
+	*state = *state < LIT_STATES ? STATE_LIT_LONGREP : STATE_NONLIT_REP;
+}
+
+/* Indicate that the latest symbol was a short match. */
+static inline void lzma_state_short_rep(enum lzma_state *state)
+{
+	*state = *state < LIT_STATES ? STATE_LIT_SHORTREP : STATE_NONLIT_REP;
+}
+
+/* Test if the previous symbol was a literal. */
+static inline bool lzma_state_is_literal(enum lzma_state state)
+{
+	return state < LIT_STATES;
+}
+
+/* Each literal coder is divided in three sections:
+ *   - 0x001-0x0FF: Without match byte
+ *   - 0x101-0x1FF: With match byte; match bit is 0
+ *   - 0x201-0x2FF: With match byte; match bit is 1
+ *
+ * Match byte is used when the previous LZMA symbol was something else than
+ * a literal (that is, it was some kind of match).
+ */
+#define LITERAL_CODER_SIZE 0x300
+
+/* Maximum number of literal coders */
+#define LITERAL_CODERS_MAX (1 << 4)
+
+/* Minimum length of a match is two bytes. */
+#define MATCH_LEN_MIN 2
+
+/* Match length is encoded with 4, 5, or 10 bits.
+ *
+ * Length   Bits
+ *  2-9      4 = Choice=0 + 3 bits
+ * 10-17     5 = Choice=1 + Choice2=0 + 3 bits
+ * 18-273   10 = Choice=1 + Choice2=1 + 8 bits
+ */
+#define LEN_LOW_BITS 3
+#define LEN_LOW_SYMBOLS (1 << LEN_LOW_BITS)
+#define LEN_MID_BITS 3
+#define LEN_MID_SYMBOLS (1 << LEN_MID_BITS)
+#define LEN_HIGH_BITS 8
+#define LEN_HIGH_SYMBOLS (1 << LEN_HIGH_BITS)
+#define LEN_SYMBOLS (LEN_LOW_SYMBOLS + LEN_MID_SYMBOLS + LEN_HIGH_SYMBOLS)
+
+/*
+ * Maximum length of a match is 273 which is a result of the encoding
+ * described above.
+ */
+#define MATCH_LEN_MAX (MATCH_LEN_MIN + LEN_SYMBOLS - 1)
+
+/*
+ * Different sets of probabilities are used for match distances that have
+ * very short match length: Lengths of 2, 3, and 4 bytes have a separate
+ * set of probabilities for each length. The matches with longer length
+ * use a shared set of probabilities.
+ */
+#define DIST_STATES 4
+
+/*
+ * Get the index of the appropriate probability array for decoding
+ * the distance slot.
+ */
+static inline uint32_t lzma_get_dist_state(uint32_t len)
+{
+	return len < DIST_STATES + MATCH_LEN_MIN
+			? len - MATCH_LEN_MIN : DIST_STATES - 1;
+}
+
+/*
+ * The highest two bits of a 32-bit match distance are encoded using six bits.
+ * This six-bit value is called a distance slot. This way encoding a 32-bit
+ * value takes 6-36 bits, larger values taking more bits.
+ */
+#define DIST_SLOT_BITS 6
+#define DIST_SLOTS (1 << DIST_SLOT_BITS)
+
+/* Match distances up to 127 are fully encoded using probabilities. Since
+ * the highest two bits (distance slot) are always encoded using six bits,
+ * the distances 0-3 don't need any additional bits to encode, since the
+ * distance slot itself is the same as the actual distance. DIST_MODEL_START
+ * indicates the first distance slot where at least one additional bit is
+ * needed.
+ */
+#define DIST_MODEL_START 4
+
+/*
+ * Match distances greater than 127 are encoded in three pieces:
+ *   - distance slot: the highest two bits
+ *   - direct bits: 2-26 bits below the highest two bits
+ *   - alignment bits: four lowest bits
+ *
+ * Direct bits don't use any probabilities.
+ *
+ * The distance slot value of 14 is for distances 128-191.
+ */
+#define DIST_MODEL_END 14
+
+/* Distance slots that indicate a distance <= 127. */
+#define FULL_DISTANCES_BITS (DIST_MODEL_END / 2)
+#define FULL_DISTANCES (1 << FULL_DISTANCES_BITS)
+
+/*
+ * For match distances greater than 127, only the highest two bits and the
+ * lowest four bits (alignment) is encoded using probabilities.
+ */
+#define ALIGN_BITS 4
+#define ALIGN_SIZE (1 << ALIGN_BITS)
+#define ALIGN_MASK (ALIGN_SIZE - 1)
+
+/* Total number of all probability variables */
+#define PROBS_TOTAL (1846 + LITERAL_CODERS_MAX * LITERAL_CODER_SIZE)
+
+/*
+ * LZMA remembers the four most recent match distances. Reusing these
+ * distances tends to take less space than re-encoding the actual
+ * distance value.
+ */
+#define REPS 4
+
+#endif
+/*
+ * Private includes and definitions
+ *
+ * Author: Lasse Collin <lasse.collin@tukaani.org>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#ifndef XZ_PRIVATE_H
+#define XZ_PRIVATE_H
+
+#ifdef __KERNEL__
+#	include <linux/xz.h>
+#	include <asm/byteorder.h>
+#	include <asm/unaligned.h>
+	/* XZ_PREBOOT may be defined only via decompress_unxz.c. */
+#	ifndef XZ_PREBOOT
+#		include <linux/slab.h>
+#		include <linux/vmalloc.h>
+#		include <linux/string.h>
+#		ifdef CONFIG_XZ_DEC_X86
+#			define XZ_DEC_X86
+#		endif
+#		ifdef CONFIG_XZ_DEC_POWERPC
+#			define XZ_DEC_POWERPC
+#		endif
+#		ifdef CONFIG_XZ_DEC_IA64
+#			define XZ_DEC_IA64
+#		endif
+#		ifdef CONFIG_XZ_DEC_ARM
+#			define XZ_DEC_ARM
+#		endif
+#		ifdef CONFIG_XZ_DEC_ARMTHUMB
+#			define XZ_DEC_ARMTHUMB
+#		endif
+#		ifdef CONFIG_XZ_DEC_SPARC
+#			define XZ_DEC_SPARC
+#		endif
+#		define memeq(a, b, size) (memcmp(a, b, size) == 0)
+#		define memzero(buf, size) memset(buf, 0, size)
+#	endif
+#	define get_le32(p) le32_to_cpup((const uint32_t *)(p))
+#else
+	/*
+	 * For userspace builds, use a separate header to define the required
+	 * macros and functions. This makes it easier to adapt the code into
+	 * different environments and avoids clutter in the Linux kernel tree.
+	 */
+#	include "xz_config.h"
+#endif
+
+/* If no specific decoding mode is requested, enable support for all modes. */
+#if !defined(XZ_DEC_SINGLE) && !defined(XZ_DEC_PREALLOC) \
+		&& !defined(XZ_DEC_DYNALLOC)
+#	define XZ_DEC_SINGLE
+#	define XZ_DEC_PREALLOC
+#	define XZ_DEC_DYNALLOC
+#endif
+
+/*
+ * The DEC_IS_foo(mode) macros are used in "if" statements. If only some
+ * of the supported modes are enabled, these macros will evaluate to true or
+ * false at compile time and thus allow the compiler to omit unneeded code.
+ */
+#ifdef XZ_DEC_SINGLE
+#	define DEC_IS_SINGLE(mode) ((mode) == XZ_SINGLE)
+#else
+#	define DEC_IS_SINGLE(mode) (false)
+#endif
+
+#ifdef XZ_DEC_PREALLOC
+#	define DEC_IS_PREALLOC(mode) ((mode) == XZ_PREALLOC)
+#else
+#	define DEC_IS_PREALLOC(mode) (false)
+#endif
+
+#ifdef XZ_DEC_DYNALLOC
+#	define DEC_IS_DYNALLOC(mode) ((mode) == XZ_DYNALLOC)
+#else
+#	define DEC_IS_DYNALLOC(mode) (false)
+#endif
+
+#if !defined(XZ_DEC_SINGLE)
+#	define DEC_IS_MULTI(mode) (true)
+#elif defined(XZ_DEC_PREALLOC) || defined(XZ_DEC_DYNALLOC)
+#	define DEC_IS_MULTI(mode) ((mode) != XZ_SINGLE)
+#else
+#	define DEC_IS_MULTI(mode) (false)
+#endif
+
+/*
+ * If any of the BCJ filter decoders are wanted, define XZ_DEC_BCJ.
+ * XZ_DEC_BCJ is used to enable generic support for BCJ decoders.
+ */
+#ifndef XZ_DEC_BCJ
+#	if defined(XZ_DEC_X86) || defined(XZ_DEC_POWERPC) \
+			|| defined(XZ_DEC_IA64) || defined(XZ_DEC_ARM) \
+			|| defined(XZ_DEC_ARM) || defined(XZ_DEC_ARMTHUMB) \
+			|| defined(XZ_DEC_SPARC)
+#		define XZ_DEC_BCJ
+#	endif
+#endif
+
+/*
+ * Allocate memory for LZMA2 decoder. xz_dec_lzma2_reset() must be used
+ * before calling xz_dec_lzma2_run().
+ */
+XZ_EXTERN struct xz_dec_lzma2 *xz_dec_lzma2_create(enum xz_mode mode,
+						   uint32_t dict_max);
+
+/*
+ * Decode the LZMA2 properties (one byte) and reset the decoder. Return
+ * XZ_OK on success, XZ_MEMLIMIT_ERROR if the preallocated dictionary is not
+ * big enough, and XZ_OPTIONS_ERROR if props indicates something that this
+ * decoder doesn't support.
+ */
+XZ_EXTERN enum xz_ret xz_dec_lzma2_reset(struct xz_dec_lzma2 *s,
+					 uint8_t props);
+
+/* Decode raw LZMA2 stream from b->in to b->out. */
+XZ_EXTERN enum xz_ret xz_dec_lzma2_run(struct xz_dec_lzma2 *s,
+				       struct xz_buf *b);
+
+/* Free the memory allocated for the LZMA2 decoder. */
+XZ_EXTERN void xz_dec_lzma2_end(struct xz_dec_lzma2 *s);
+
+#ifdef XZ_DEC_BCJ
+/*
+ * Allocate memory for BCJ decoders. xz_dec_bcj_reset() must be used before
+ * calling xz_dec_bcj_run().
+ */
+XZ_EXTERN struct xz_dec_bcj *xz_dec_bcj_create(bool single_call);
+
+/*
+ * Decode the Filter ID of a BCJ filter. This implementation doesn't
+ * support custom start offsets, so no decoding of Filter Properties
+ * is needed. Returns XZ_OK if the given Filter ID is supported.
+ * Otherwise XZ_OPTIONS_ERROR is returned.
+ */
+XZ_EXTERN enum xz_ret xz_dec_bcj_reset(struct xz_dec_bcj *s, uint8_t id);
+
+/*
+ * Decode raw BCJ + LZMA2 stream. This must be used only if there actually is
+ * a BCJ filter in the chain. If the chain has only LZMA2, xz_dec_lzma2_run()
+ * must be called directly.
+ */
+XZ_EXTERN enum xz_ret xz_dec_bcj_run(struct xz_dec_bcj *s,
+				     struct xz_dec_lzma2 *lzma2,
+				     struct xz_buf *b);
+
+/* Free the memory allocated for the BCJ filters. */
+#define xz_dec_bcj_end(s) kfree(s)
+#endif
+
+#endif
+/*
+ * Definitions for handling the .xz file format
+ *
+ * Author: Lasse Collin <lasse.collin@tukaani.org>
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#ifndef XZ_STREAM_H
+#define XZ_STREAM_H
+
+#if defined(__KERNEL__) && !XZ_INTERNAL_CRC32
+#	include <linux/crc32.h>
+#	undef crc32
+#	define xz_crc32(buf, size, crc) \
+		(~crc32_le(~(uint32_t)(crc), buf, size))
+#endif
+
+/*
+ * See the .xz file format specification at
+ * http://tukaani.org/xz/xz-file-format.txt
+ * to understand the container format.
+ */
+
+#define STREAM_HEADER_SIZE 12
+
+#define HEADER_MAGIC "\3757zXZ"
+#define HEADER_MAGIC_SIZE 6
+
+#define FOOTER_MAGIC "YZ"
+#define FOOTER_MAGIC_SIZE 2
+
+/*
+ * Variable-length integer can hold a 63-bit unsigned integer or a special
+ * value indicating that the value is unknown.
+ *
+ * Experimental: vli_type can be defined to uint32_t to save a few bytes
+ * in code size (no effect on speed). Doing so limits the uncompressed and
+ * compressed size of the file to less than 256 MiB and may also weaken
+ * error detection slightly.
+ */
+typedef uint64_t vli_type;
+
+#define VLI_MAX ((vli_type)-1 / 2)
+#define VLI_UNKNOWN ((vli_type)-1)
+
+/* Maximum encoded size of a VLI */
+#define VLI_BYTES_MAX (sizeof(vli_type) * 8 / 7)
+
+/* Integrity Check types */
+enum xz_check {
+	XZ_CHECK_NONE = 0,
+	XZ_CHECK_CRC32 = 1,
+	XZ_CHECK_CRC64 = 4,
+	XZ_CHECK_SHA256 = 10
+};
+
+/* Maximum possible Check ID */
+#define XZ_CHECK_MAX 15
+
+#endif
@@ -262,6 +262,34 @@
 	lzop -9 && $(call size_append, $(filter-out FORCE,$^))) > $@ || \
 	(rm -f $@ ; false)
  
+# XZ
+# ---------------------------------------------------------------------------
+# Use xzkern to compress the kernel image and xzmisc to compress other things.
+#
+# xzkern uses a big LZMA2 dictionary since it doesn't increase memory usage
+# of the kernel decompressor. A BCJ filter is used if it is available for
+# the target architecture. xzkern also appends uncompressed size of the data
+# using size_append. The .xz format has the size information available at
+# the end of the file too, but it's in more complex format and it's good to
+# avoid changing the part of the boot code that reads the uncompressed size.
+# Note that the bytes added by size_append will make the xz tool think that
+# the file is corrupt. This is expected.
+#
+# xzmisc doesn't use size_append, so it can be used to create normal .xz
+# files. xzmisc uses smaller LZMA2 dictionary than xzkern, because a very
+# big dictionary would increase the memory usage too much in the multi-call
+# decompression mode. A BCJ filter isn't used either.
+quiet_cmd_xzkern = XZKERN  $@
+cmd_xzkern = (cat $(filter-out FORCE,$^) | \
+	sh $(srctree)/scripts/xz_wrap.sh && \
+	$(call size_append, $(filter-out FORCE,$^))) > $@ || \
+	(rm -f $@ ; false)
+
+quiet_cmd_xzmisc = XZMISC  $@
+cmd_xzmisc = (cat $(filter-out FORCE,$^) | \
+	xz --check=crc32 --lzma2=dict=1MiB) > $@ || \
+	(rm -f $@ ; false)
+
 # misc stuff
 # ---------------------------------------------------------------------------
 quote:="
+#!/bin/sh
+#
+# This is a wrapper for xz to compress the kernel image using appropriate
+# compression options depending on the architecture.
+#
+# Author: Lasse Collin <lasse.collin@tukaani.org>
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+
+BCJ=
+LZMA2OPTS=
+
+case $ARCH in
+	x86|x86_64)     BCJ=--x86 ;;
+	powerpc)        BCJ=--powerpc ;;
+	ia64)           BCJ=--ia64; LZMA2OPTS=pb=4 ;;
+	arm)            BCJ=--arm ;;
+	sparc)          BCJ=--sparc ;;
+esac
+
+exec xz --check=crc32 $BCJ --lzma2=$LZMA2OPTS,dict=32MiB
	1	+
	2	+XZ data compression in Linux
	3	+============================
	4	+
	5	+Introduction
	6	+
	7	+ XZ is a general purpose data compression format with high compression
	8	+ ratio and relatively fast decompression. The primary compression
	9	+ algorithm (filter) is LZMA2. Additional filters can be used to improve
	10	+ compression ratio even further. E.g. Branch/Call/Jump (BCJ) filters
	11	+ improve compression ratio of executable data.
	12	+
	13	+ The XZ decompressor in Linux is called XZ Embedded. It supports
	14	+ the LZMA2 filter and optionally also BCJ filters. CRC32 is supported
	15	+ for integrity checking. The home page of XZ Embedded is at
	16	+ <http://tukaani.org/xz/embedded.html>, where you can find the
	17	+ latest version and also information about using the code outside
	18	+ the Linux kernel.
	19	+
	20	+ For userspace, XZ Utils provide a zlib-like compression library
	21	+ and a gzip-like command line tool. XZ Utils can be downloaded from
	22	+ <http://tukaani.org/xz/>.
	23	+
	24	+XZ related components in the kernel
	25	+
	26	+ The xz_dec module provides XZ decompressor with single-call (buffer
	27	+ to buffer) and multi-call (stateful) APIs. The usage of the xz_dec
	28	+ module is documented in include/linux/xz.h.
	29	+
	30	+ The xz_dec_test module is for testing xz_dec. xz_dec_test is not
	31	+ useful unless you are hacking the XZ decompressor. xz_dec_test
	32	+ allocates a char device major dynamically to which one can write
	33	+ .xz files from userspace. The decompressed output is thrown away.
	34	+ Keep an eye on dmesg to see diagnostics printed by xz_dec_test.
	35	+ See the xz_dec_test source code for the details.
	36	+
	37	+ For decompressing the kernel image, initramfs, and initrd, there
	38	+ is a wrapper function in lib/decompress_unxz.c. Its API is the
	39	+ same as in other decompress_*.c files, which is defined in
	40	+ include/linux/decompress/generic.h.
	41	+
	42	+ scripts/xz_wrap.sh is a wrapper for the xz command line tool found
	43	+ from XZ Utils. The wrapper sets compression options to values suitable
	44	+ for compressing the kernel image.
	45	+
	46	+ For kernel makefiles, two commands are provided for use with
	47	+ $(call if_needed). The kernel image should be compressed with
	48	+ $(call if_needed,xzkern) which will use a BCJ filter and a big LZMA2
	49	+ dictionary. It will also append a four-byte trailer containing the
	50	+ uncompressed size of the file, which is needed by the boot code.
	51	+ Other things should be compressed with $(call if_needed,xzmisc)
	52	+ which will use no BCJ filter and 1 MiB LZMA2 dictionary.
	53	+
	54	+Notes on compression options
	55	+
	56	+ Since the XZ Embedded supports only streams with no integrity check or
	57	+ CRC32, make sure that you don't use some other integrity check type
	58	+ when encoding files that are supposed to be decoded by the kernel. With
	59	+ liblzma, you need to use either LZMA_CHECK_NONE or LZMA_CHECK_CRC32
	60	+ when encoding. With the xz command line tool, use --check=none or
	61	+ --check=crc32.
	62	+
	63	+ Using CRC32 is strongly recommended unless there is some other layer
	64	+ which will verify the integrity of the uncompressed data anyway.
	65	+ Double checking the integrity would probably be waste of CPU cycles.
	66	+ Note that the headers will always have a CRC32 which will be validated
	67	+ by the decoder; you can only change the integrity check type (or
	68	+ disable it) for the actual uncompressed data.
	69	+
	70	+ In userspace, LZMA2 is typically used with dictionary sizes of several
	71	+ megabytes. The decoder needs to have the dictionary in RAM, thus big
	72	+ dictionaries cannot be used for files that are intended to be decoded
	73	+ by the kernel. 1 MiB is probably the maximum reasonable dictionary
	74	+ size for in-kernel use (maybe more is OK for initramfs). The presets
	75	+ in XZ Utils may not be optimal when creating files for the kernel,
	76	+ so don't hesitate to use custom settings. Example:
	77	+
	78	+ xz --check=crc32 --lzma2=dict=512KiB inputfile
	79	+
	80	+ An exception to above dictionary size limitation is when the decoder
	81	+ is used in single-call mode. Decompressing the kernel itself is an
	82	+ example of this situation. In single-call mode, the memory usage
	83	+ doesn't depend on the dictionary size, and it is perfectly fine to
	84	+ use a big dictionary: for maximum compression, the dictionary should
	85	+ be at least as big as the uncompressed data itself.
	86	+
	87	+Future plans
	88	+
	89	+ Creating a limited XZ encoder may be considered if people think it is
	90	+ useful. LZMA2 is slower to compress than e.g. Deflate or LZO even at
	91	+ the fastest settings, so it isn't clear if LZMA2 encoder is wanted
	92	+ into the kernel.
	93	+
	94	+ Support for limited random-access reading is planned for the
	95	+ decompression code. I don't know if it could have any use in the
	96	+ kernel, but I know that it would be useful in some embedded projects
	97	+ outside the Linux kernel.
	98	+
	99	+Conformance to the .xz file format specification
	100	+
	101	+ There are a couple of corner cases where things have been simplified
	102	+ at expense of detecting errors as early as possible. These should not
	103	+ matter in practice all, since they don't cause security issues. But
	104	+ it is good to know this if testing the code e.g. with the test files
	105	+ from XZ Utils.
	106	+
	107	+Reporting bugs
	108	+
	109	+ Before reporting a bug, please check that it's not fixed already
	110	+ at upstream. See <http://tukaani.org/xz/embedded.html> to get the
	111	+ latest code.
	112	+
	113	+ Report bugs to <lasse.collin@tukaani.org> or visit #tukaani on
	114	+ Freenode and talk to Larhzu. I don't actively read LKML or other
	115	+ kernel-related mailing lists, so if there's something I should know,
	116	+ you should email to me personally or use IRC.
	117	+
	118	+ Don't bother Igor Pavlov with questions about the XZ implementation
	119	+ in the kernel or about XZ Utils. While these two implementations
	120	+ include essential code that is directly based on Igor Pavlov's code,
	121	+ these implementations aren't maintained nor supported by him.
	1	+/*
	2	+ * XZ decompressor
	3	+ *
	4	+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
	5	+ * Igor Pavlov <http://7-zip.org/>
	6	+ *
	7	+ * This file has been put into the public domain.
	8	+ * You can do whatever you want with this file.
	9	+ */
	10	+
	11	+#ifndef XZ_H
	12	+#define XZ_H
	13	+
	14	+#ifdef __KERNEL__
	15	+# include <linux/stddef.h>
	16	+# include <linux/types.h>
	17	+#else
	18	+# include <stddef.h>
	19	+# include <stdint.h>
	20	+#endif
	21	+
	22	+/* In Linux, this is used to make extern functions static when needed. */
	23	+#ifndef XZ_EXTERN
	24	+# define XZ_EXTERN extern
	25	+#endif
	26	+
	27	+/**
	28	+ * enum xz_mode - Operation mode
	29	+ *
	30	+ * @XZ_SINGLE: Single-call mode. This uses less RAM than
	31	+ * than multi-call modes, because the LZMA2
	32	+ * dictionary doesn't need to be allocated as
	33	+ * part of the decoder state. All required data
	34	+ * structures are allocated at initialization,
	35	+ * so xz_dec_run() cannot return XZ_MEM_ERROR.
	36	+ * @XZ_PREALLOC: Multi-call mode with preallocated LZMA2
	37	+ * dictionary buffer. All data structures are
	38	+ * allocated at initialization, so xz_dec_run()
	39	+ * cannot return XZ_MEM_ERROR.
	40	+ * @XZ_DYNALLOC: Multi-call mode. The LZMA2 dictionary is
	41	+ * allocated once the required size has been
	42	+ * parsed from the stream headers. If the
	43	+ * allocation fails, xz_dec_run() will return
	44	+ * XZ_MEM_ERROR.
	45	+ *
	46	+ * It is possible to enable support only for a subset of the above
	47	+ * modes at compile time by defining XZ_DEC_SINGLE, XZ_DEC_PREALLOC,
	48	+ * or XZ_DEC_DYNALLOC. The xz_dec kernel module is always compiled
	49	+ * with support for all operation modes, but the preboot code may
	50	+ * be built with fewer features to minimize code size.
	51	+ */
	52	+enum xz_mode {
	53	+ XZ_SINGLE,
	54	+ XZ_PREALLOC,
	55	+ XZ_DYNALLOC
	56	+};
	57	+
	58	+/**
	59	+ * enum xz_ret - Return codes
	60	+ * @XZ_OK: Everything is OK so far. More input or more
	61	+ * output space is required to continue. This
	62	+ * return code is possible only in multi-call mode
	63	+ * (XZ_PREALLOC or XZ_DYNALLOC).
	64	+ * @XZ_STREAM_END: Operation finished successfully.
	65	+ * @XZ_UNSUPPORTED_CHECK: Integrity check type is not supported. Decoding
	66	+ * is still possible in multi-call mode by simply
	67	+ * calling xz_dec_run() again.
	68	+ * Note that this return value is used only if
	69	+ * XZ_DEC_ANY_CHECK was defined at build time,
	70	+ * which is not used in the kernel. Unsupported
	71	+ * check types return XZ_OPTIONS_ERROR if
	72	+ * XZ_DEC_ANY_CHECK was not defined at build time.
	73	+ * @XZ_MEM_ERROR: Allocating memory failed. This return code is
	74	+ * possible only if the decoder was initialized
	75	+ * with XZ_DYNALLOC. The amount of memory that was
	76	+ * tried to be allocated was no more than the
	77	+ * dict_max argument given to xz_dec_init().
	78	+ * @XZ_MEMLIMIT_ERROR: A bigger LZMA2 dictionary would be needed than
	79	+ * allowed by the dict_max argument given to
	80	+ * xz_dec_init(). This return value is possible
	81	+ * only in multi-call mode (XZ_PREALLOC or
	82	+ * XZ_DYNALLOC); the single-call mode (XZ_SINGLE)
	83	+ * ignores the dict_max argument.
	84	+ * @XZ_FORMAT_ERROR: File format was not recognized (wrong magic
	85	+ * bytes).
	86	+ * @XZ_OPTIONS_ERROR: This implementation doesn't support the requested
	87	+ * compression options. In the decoder this means
	88	+ * that the header CRC32 matches, but the header
	89	+ * itself specifies something that we don't support.
	90	+ * @XZ_DATA_ERROR: Compressed data is corrupt.
	91	+ * @XZ_BUF_ERROR: Cannot make any progress. Details are slightly
	92	+ * different between multi-call and single-call
	93	+ * mode; more information below.
	94	+ *
	95	+ * In multi-call mode, XZ_BUF_ERROR is returned when two consecutive calls
	96	+ * to XZ code cannot consume any input and cannot produce any new output.
	97	+ * This happens when there is no new input available, or the output buffer
	98	+ * is full while at least one output byte is still pending. Assuming your
	99	+ * code is not buggy, you can get this error only when decoding a compressed
	100	+ * stream that is truncated or otherwise corrupt.
	101	+ *
	102	+ * In single-call mode, XZ_BUF_ERROR is returned only when the output buffer
	103	+ * is too small or the compressed input is corrupt in a way that makes the
	104	+ * decoder produce more output than the caller expected. When it is
	105	+ * (relatively) clear that the compressed input is truncated, XZ_DATA_ERROR
	106	+ * is used instead of XZ_BUF_ERROR.
	107	+ */
	108	+enum xz_ret {
	109	+ XZ_OK,
	110	+ XZ_STREAM_END,
	111	+ XZ_UNSUPPORTED_CHECK,
	112	+ XZ_MEM_ERROR,
	113	+ XZ_MEMLIMIT_ERROR,
	114	+ XZ_FORMAT_ERROR,
	115	+ XZ_OPTIONS_ERROR,
	116	+ XZ_DATA_ERROR,
	117	+ XZ_BUF_ERROR
	118	+};
	119	+
	120	+/**
	121	+ * struct xz_buf - Passing input and output buffers to XZ code
	122	+ * @in: Beginning of the input buffer. This may be NULL if and only
	123	+ * if in_pos is equal to in_size.
	124	+ * @in_pos: Current position in the input buffer. This must not exceed
	125	+ * in_size.
	126	+ * @in_size: Size of the input buffer
	127	+ * @out: Beginning of the output buffer. This may be NULL if and only
	128	+ * if out_pos is equal to out_size.
	129	+ * @out_pos: Current position in the output buffer. This must not exceed
	130	+ * out_size.
	131	+ * @out_size: Size of the output buffer
	132	+ *
	133	+ * Only the contents of the output buffer from out[out_pos] onward, and
	134	+ * the variables in_pos and out_pos are modified by the XZ code.
	135	+ */
	136	+struct xz_buf {
	137	+ const uint8_t *in;
	138	+ size_t in_pos;
	139	+ size_t in_size;
	140	+
	141	+ uint8_t *out;
	142	+ size_t out_pos;
	143	+ size_t out_size;
	144	+};
	145	+
	146	+/**
	147	+ * struct xz_dec - Opaque type to hold the XZ decoder state
	148	+ */
	149	+struct xz_dec;
	150	+
	151	+/**
	152	+ * xz_dec_init() - Allocate and initialize a XZ decoder state
	153	+ * @mode: Operation mode
	154	+ * @dict_max: Maximum size of the LZMA2 dictionary (history buffer) for
	155	+ * multi-call decoding. This is ignored in single-call mode
	156	+ * (mode == XZ_SINGLE). LZMA2 dictionary is always 2^n bytes
	157	+ * or 2^n + 2^(n-1) bytes (the latter sizes are less common
	158	+ * in practice), so other values for dict_max don't make sense.
	159	+ * In the kernel, dictionary sizes of 64 KiB, 128 KiB, 256 KiB,
	160	+ * 512 KiB, and 1 MiB are probably the only reasonable values,
	161	+ * except for kernel and initramfs images where a bigger
	162	+ * dictionary can be fine and useful.
	163	+ *
	164	+ * Single-call mode (XZ_SINGLE): xz_dec_run() decodes the whole stream at
	165	+ * once. The caller must provide enough output space or the decoding will
	166	+ * fail. The output space is used as the dictionary buffer, which is why
	167	+ * there is no need to allocate the dictionary as part of the decoder's
	168	+ * internal state.
	169	+ *
	170	+ * Because the output buffer is used as the workspace, streams encoded using
	171	+ * a big dictionary are not a problem in single-call mode. It is enough that
	172	+ * the output buffer is big enough to hold the actual uncompressed data; it
	173	+ * can be smaller than the dictionary size stored in the stream headers.
	174	+ *
	175	+ * Multi-call mode with preallocated dictionary (XZ_PREALLOC): dict_max bytes
	176	+ * of memory is preallocated for the LZMA2 dictionary. This way there is no
	177	+ * risk that xz_dec_run() could run out of memory, since xz_dec_run() will
	178	+ * never allocate any memory. Instead, if the preallocated dictionary is too
	179	+ * small for decoding the given input stream, xz_dec_run() will return
	180	+ * XZ_MEMLIMIT_ERROR. Thus, it is important to know what kind of data will be
	181	+ * decoded to avoid allocating excessive amount of memory for the dictionary.
	182	+ *
	183	+ * Multi-call mode with dynamically allocated dictionary (XZ_DYNALLOC):
	184	+ * dict_max specifies the maximum allowed dictionary size that xz_dec_run()
	185	+ * may allocate once it has parsed the dictionary size from the stream
	186	+ * headers. This way excessive allocations can be avoided while still
	187	+ * limiting the maximum memory usage to a sane value to prevent running the
	188	+ * system out of memory when decompressing streams from untrusted sources.
	189	+ *
	190	+ * On success, xz_dec_init() returns a pointer to struct xz_dec, which is
	191	+ * ready to be used with xz_dec_run(). If memory allocation fails,
	192	+ * xz_dec_init() returns NULL.
	193	+ */
	194	+XZ_EXTERN struct xz_dec *xz_dec_init(enum xz_mode mode, uint32_t dict_max);
	195	+
	196	+/**
	197	+ * xz_dec_run() - Run the XZ decoder
	198	+ * @s: Decoder state allocated using xz_dec_init()
	199	+ * @b: Input and output buffers
	200	+ *
	201	+ * The possible return values depend on build options and operation mode.
	202	+ * See enum xz_ret for details.
	203	+ *
	204	+ * Note that if an error occurs in single-call mode (return value is not
	205	+ * XZ_STREAM_END), b->in_pos and b->out_pos are not modified and the
	206	+ * contents of the output buffer from b->out[b->out_pos] onward are
	207	+ * undefined. This is true even after XZ_BUF_ERROR, because with some filter
	208	+ * chains, there may be a second pass over the output buffer, and this pass
	209	+ * cannot be properly done if the output buffer is truncated. Thus, you
	210	+ * cannot give the single-call decoder a too small buffer and then expect to
	211	+ * get that amount valid data from the beginning of the stream. You must use
	212	+ * the multi-call decoder if you don't want to uncompress the whole stream.
	213	+ */
	214	+XZ_EXTERN enum xz_ret xz_dec_run(struct xz_dec s, struct xz_buf b);
	215	+
	216	+/**
	217	+ * xz_dec_reset() - Reset an already allocated decoder state
	218	+ * @s: Decoder state allocated using xz_dec_init()
	219	+ *
	220	+ * This function can be used to reset the multi-call decoder state without
	221	+ * freeing and reallocating memory with xz_dec_end() and xz_dec_init().
	222	+ *
	223	+ * In single-call mode, xz_dec_reset() is always called in the beginning of
	224	+ * xz_dec_run(). Thus, explicit call to xz_dec_reset() is useful only in
	225	+ * multi-call mode.
	226	+ */
	227	+XZ_EXTERN void xz_dec_reset(struct xz_dec *s);
	228	+
	229	+/**
	230	+ * xz_dec_end() - Free the memory allocated for the decoder state
	231	+ * @s: Decoder state allocated using xz_dec_init(). If s is NULL,
	232	+ * this function does nothing.
	233	+ */
	234	+XZ_EXTERN void xz_dec_end(struct xz_dec *s);
	235	+
	236	+/*
	237	+ * Standalone build (userspace build or in-kernel build for boot time use)
	238	+ * needs a CRC32 implementation. For normal in-kernel use, kernel's own
	239	+ * CRC32 module is used instead, and users of this module don't need to
	240	+ * care about the functions below.
	241	+ */
	242	+#ifndef XZ_INTERNAL_CRC32
	243	+# ifdef __KERNEL__
	244	+# define XZ_INTERNAL_CRC32 0
	245	+# else
	246	+# define XZ_INTERNAL_CRC32 1
	247	+# endif
	248	+#endif
	249	+
	250	+#if XZ_INTERNAL_CRC32
	251	+/*
	252	+ * This must be called before any other xz_* function to initialize
	253	+ * the CRC32 lookup table.
	254	+ */
	255	+XZ_EXTERN void xz_crc32_init(void);
	256	+
	257	+/*
	258	+ * Update CRC32 value using the polynomial from IEEE-802.3. To start a new
	259	+ * calculation, the third argument must be zero. To continue the calculation,
	260	+ * the previously returned value is passed as the third argument.
	261	+ */
	262	+XZ_EXTERN uint32_t xz_crc32(const uint8_t *buf, size_t size, uint32_t crc);
	263	+#endif
	264	+#endif
...	...	@@ -106,6 +106,8 @@
106	106	config LZO_DECOMPRESS
107	107	tristate
108	108
	109	+source "lib/xz/Kconfig"
	110	+
109	111	#
110	112	# These all provide a common interface (hence the apparent duplication with
111	113	# ZLIB_INFLATE; DECOMPRESS_GZIP is just a wrapper.)
...	...	@@ -69,6 +69,7 @@
69	69	obj-$(CONFIG_REED_SOLOMON) += reed_solomon/
70	70	obj-$(CONFIG_LZO_COMPRESS) += lzo/
71	71	obj-$(CONFIG_LZO_DECOMPRESS) += lzo/
	72	+obj-$(CONFIG_XZ_DEC) += xz/
72	73	obj-$(CONFIG_RAID6_PQ) += raid6/
73	74
74	75	lib-$(CONFIG_DECOMPRESS_GZIP) += decompress_inflate.o
	1	+config XZ_DEC
	2	+ tristate "XZ decompression support"
	3	+ select CRC32
	4	+ help
	5	+ LZMA2 compression algorithm and BCJ filters are supported using
	6	+ the .xz file format as the container. For integrity checking,
	7	+ CRC32 is supported. See Documentation/xz.txt for more information.
	8	+
	9	+config XZ_DEC_X86
	10	+ bool "x86 BCJ filter decoder" if EMBEDDED
	11	+ default y
	12	+ depends on XZ_DEC
	13	+ select XZ_DEC_BCJ
	14	+
	15	+config XZ_DEC_POWERPC
	16	+ bool "PowerPC BCJ filter decoder" if EMBEDDED
	17	+ default y
	18	+ depends on XZ_DEC
	19	+ select XZ_DEC_BCJ
	20	+
	21	+config XZ_DEC_IA64
	22	+ bool "IA-64 BCJ filter decoder" if EMBEDDED
	23	+ default y
	24	+ depends on XZ_DEC
	25	+ select XZ_DEC_BCJ
	26	+
	27	+config XZ_DEC_ARM
	28	+ bool "ARM BCJ filter decoder" if EMBEDDED
	29	+ default y
	30	+ depends on XZ_DEC
	31	+ select XZ_DEC_BCJ
	32	+
	33	+config XZ_DEC_ARMTHUMB
	34	+ bool "ARM-Thumb BCJ filter decoder" if EMBEDDED
	35	+ default y
	36	+ depends on XZ_DEC
	37	+ select XZ_DEC_BCJ
	38	+
	39	+config XZ_DEC_SPARC
	40	+ bool "SPARC BCJ filter decoder" if EMBEDDED
	41	+ default y
	42	+ depends on XZ_DEC
	43	+ select XZ_DEC_BCJ
	44	+
	45	+config XZ_DEC_BCJ
	46	+ bool
	47	+ default n
	48	+
	49	+config XZ_DEC_TEST
	50	+ tristate "XZ decompressor tester"
	51	+ default n
	52	+ depends on XZ_DEC
	53	+ help
	54	+ This allows passing .xz files to the in-kernel XZ decoder via
	55	+ a character special file. It calculates CRC32 of the decompressed
	56	+ data and writes diagnostics to the system log.
	57	+
	58	+ Unless you are developing the XZ decoder, you don't need this
	59	+ and should say N.
	1	+obj-$(CONFIG_XZ_DEC) += xz_dec.o
	2	+xz_dec-y := xz_dec_syms.o xz_dec_stream.o xz_dec_lzma2.o
	3	+xz_dec-$(CONFIG_XZ_DEC_BCJ) += xz_dec_bcj.o
	4	+
	5	+obj-$(CONFIG_XZ_DEC_TEST) += xz_dec_test.o
	1	+/*
	2	+ * CRC32 using the polynomial from IEEE-802.3
	3	+ *
	4	+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
	5	+ * Igor Pavlov <http://7-zip.org/>
	6	+ *
	7	+ * This file has been put into the public domain.
	8	+ * You can do whatever you want with this file.
	9	+ */
	10	+
	11	+/*
	12	+ * This is not the fastest implementation, but it is pretty compact.
	13	+ * The fastest versions of xz_crc32() on modern CPUs without hardware
	14	+ * accelerated CRC instruction are 3-5 times as fast as this version,
	15	+ * but they are bigger and use more memory for the lookup table.
	16	+ */
	17	+
	18	+#include "xz_private.h"
	19	+
	20	+/*
	21	+ * STATIC_RW_DATA is used in the pre-boot environment on some architectures.
	22	+ * See <linux/decompress/mm.h> for details.
	23	+ */
	24	+#ifndef STATIC_RW_DATA
	25	+# define STATIC_RW_DATA static
	26	+#endif
	27	+
	28	+STATIC_RW_DATA uint32_t xz_crc32_table[256];
	29	+
	30	+XZ_EXTERN void xz_crc32_init(void)
	31	+{
	32	+ const uint32_t poly = 0xEDB88320;
	33	+
	34	+ uint32_t i;
	35	+ uint32_t j;
	36	+ uint32_t r;
	37	+
	38	+ for (i = 0; i < 256; ++i) {
	39	+ r = i;
	40	+ for (j = 0; j < 8; ++j)
	41	+ r = (r >> 1) ^ (poly & ~((r & 1) - 1));
	42	+
	43	+ xz_crc32_table[i] = r;
	44	+ }
	45	+
	46	+ return;
	47	+}
	48	+
	49	+XZ_EXTERN uint32_t xz_crc32(const uint8_t *buf, size_t size, uint32_t crc)
	50	+{
	51	+ crc = ~crc;
	52	+
	53	+ while (size != 0) {
	54	+ crc = xz_crc32_table[*buf++ ^ (crc & 0xFF)] ^ (crc >> 8);
	55	+ --size;
	56	+ }
	57	+
	58	+ return ~crc;
	59	+}
	1	+/*
	2	+ * Branch/Call/Jump (BCJ) filter decoders
	3	+ *
	4	+ * Authors: Lasse Collin <lasse.collin@tukaani.org>
	5	+ * Igor Pavlov <http://7-zip.org/>
	6	+ *
	7	+ * This file has been put into the public domain.
	8	+ * You can do whatever you want with this file.
	9	+ */
	10	+
	11	+#include "xz_private.h"
	12	+
	13	+/*
	14	+ * The rest of the file is inside this ifdef. It makes things a little more
	15	+ * convenient when building without support for any BCJ filters.
	16	+ */
	17	+#ifdef XZ_DEC_BCJ
	18	+
	19	+struct xz_dec_bcj {
	20	+ /* Type of the BCJ filter being used */
	21	+ enum {
	22	+ BCJ_X86 = 4, /* x86 or x86-64 */
	23	+ BCJ_POWERPC = 5, /* Big endian only */
	24	+ BCJ_IA64 = 6, /* Big or little endian */
	25	+ BCJ_ARM = 7, /* Little endian only */
	26	+ BCJ_ARMTHUMB = 8, /* Little endian only */
	27	+ BCJ_SPARC = 9 /* Big or little endian */
	28	+ } type;
	29	+
	30	+ /*
	31	+ * Return value of the next filter in the chain. We need to preserve
	32	+ * this information across calls, because we must not call the next
	33	+ * filter anymore once it has returned XZ_STREAM_END.
	34	+ */
	35	+ enum xz_ret ret;
	36	+
	37	+ /* True if we are operating in single-call mode. */
	38	+ bool single_call;
	39	+
	40	+ /*
	41	+ * Absolute position relative to the beginning of the uncompressed
	42	+ * data (in a single .xz Block). We care only about the lowest 32
	43	+ * bits so this doesn't need to be uint64_t even with big files.
	44	+ */
	45	+ uint32_t pos;
	46	+
	47	+ /* x86 filter state */
	48	+ uint32_t x86_prev_mask;
	49	+
	50	+ /* Temporary space to hold the variables from struct xz_buf */
	51	+ uint8_t *out;
	52	+ size_t out_pos;
	53	+ size_t out_size;
	54	+
	55	+ struct {
	56	+ /* Amount of already filtered data in the beginning of buf */
	57	+ size_t filtered;
	58	+
	59	+ /* Total amount of data currently stored in buf */
	60	+ size_t size;
	61	+
	62	+ /*
	63	+ * Buffer to hold a mix of filtered and unfiltered data. This
	64	+ * needs to be big enough to hold Alignment + 2 * Look-ahead:
	65	+ *
	66	+ * Type Alignment Look-ahead
	67	+ * x86 1 4
	68	+ * PowerPC 4 0
	69	+ * IA-64 16 0
	70	+ * ARM 4 0
	71	+ * ARM-Thumb 2 2
	72	+ * SPARC 4 0
	73	+ */
	74	+ uint8_t buf[16];
	75	+ } temp;
	76	+};
	77	+
	78	+#ifdef XZ_DEC_X86
	79	+/*
	80	+ * This is used to test the most significant byte of a memory address
	81	+ * in an x86 instruction.
	82	+ */
	83	+static inline int bcj_x86_test_msbyte(uint8_t b)
	84	+{
	85	+ return b == 0x00 \|\| b == 0xFF;
	86	+}
	87	+
	88	+static size_t bcj_x86(struct xz_dec_bcj s, uint8_t buf, size_t size)
	89	+{
	90	+ static const bool mask_to_allowed_status[8]
	91	+ = { true, true, true, false, true, false, false, false };
	92	+
	93	+ static const uint8_t mask_to_bit_num[8] = { 0, 1, 2, 2, 3, 3, 3, 3 };
	94	+
	95	+ size_t i;
	96	+ size_t prev_pos = (size_t)-1;
	97	+ uint32_t prev_mask = s->x86_prev_mask;
	98	+ uint32_t src;
	99	+ uint32_t dest;
	100	+ uint32_t j;
	101	+ uint8_t b;
	102	+
	103	+ if (size <= 4)
	104	+ return 0;
	105	+
	106	+ size -= 4;
	107	+ for (i = 0; i < size; ++i) {
	108	+ if ((buf[i] & 0xFE) != 0xE8)
	109	+ continue;
	110	+
	111	+ prev_pos = i - prev_pos;
	112	+ if (prev_pos > 3) {
	113	+ prev_mask = 0;
	114	+ } else {
	115	+ prev_mask = (prev_mask << (prev_pos - 1)) & 7;
	116	+ if (prev_mask != 0) {
	117	+ b = buf[i + 4 - mask_to_bit_num[prev_mask]];
	118	+ if (!mask_to_allowed_status[prev_mask]
	119	+ \|\| bcj_x86_test_msbyte(b)) {
	120	+ prev_pos = i;
	121	+ prev_mask = (prev_mask << 1) \| 1;
	122	+ continue;
	123	+ }
	124	+ }
	125	+ }
	126	+
	127	+ prev_pos = i;
	128	+
	129	+ if (bcj_x86_test_msbyte(buf[i + 4])) {
	130	+ src = get_unaligned_le32(buf + i + 1);
	131	+ while (true) {
	132	+ dest = src - (s->pos + (uint32_t)i + 5);
	133	+ if (prev_mask == 0)
	134	+ break;
	135	+
	136	+ j = mask_to_bit_num[prev_mask] * 8;
	137	+ b = (uint8_t)(dest >> (24 - j));
	138	+ if (!bcj_x86_test_msbyte(b))
	139	+ break;
	140	+
	141	+ src = dest ^ (((uint32_t)1 << (32 - j)) - 1);
	142	+ }
	143	+
	144	+ dest &= 0x01FFFFFF;
	145	+ dest \|= (uint32_t)0 - (dest & 0x01000000);
	146	+ put_unaligned_le32(dest, buf + i + 1);
	147	+ i += 4;
	148	+ } else {
	149	+ prev_mask = (prev_mask << 1) \| 1;
	150	+ }
	151	+ }
	152	+
	153	+ prev_pos = i - prev_pos;
	154	+ s->x86_prev_mask = prev_pos > 3 ? 0 : prev_mask << (prev_pos - 1);
	155	+ return i;
	156	+}
	157	+#endif
	158	+
	159	+#ifdef XZ_DEC_POWERPC
	160	+static size_t bcj_powerpc(struct xz_dec_bcj s, uint8_t buf, size_t size)
	161	+{
	162	+ size_t i;
	163	+ uint32_t instr;
	164	+
	165	+ for (i = 0; i + 4 <= size; i += 4) {
	166	+ instr = get_unaligned_be32(buf + i);
	167	+ if ((instr & 0xFC000003) == 0x48000001) {
	168	+ instr &= 0x03FFFFFC;
	169	+ instr -= s->pos + (uint32_t)i;
	170	+ instr &= 0x03FFFFFC;
	171	+ instr \|= 0x48000001;
	172	+ put_unaligned_be32(instr, buf + i);
	173	+ }
	174	+ }
	175	+
	176	+ return i;
	177	+}
	178	+#endif
	179	+
	180	+#ifdef XZ_DEC_IA64
	181	+static size_t bcj_ia64(struct xz_dec_bcj s, uint8_t buf, size_t size)
	182	+{
	183	+ static const uint8_t branch_table[32] = {
	184	+ 0, 0, 0, 0, 0, 0, 0, 0,
	185	+ 0, 0, 0, 0, 0, 0, 0, 0,
	186	+ 4, 4, 6, 6, 0, 0, 7, 7,
	187	+ 4, 4, 0, 0, 4, 4, 0, 0
	188	+ };
	189	+
	190	+ /*
	191	+ * The local variables take a little bit stack space, but it's less
	192	+ * than what LZMA2 decoder takes, so it doesn't make sense to reduce
	193	+ * stack usage here without doing that for the LZMA2 decoder too.
	194	+ */
	195	+
	196	+ /* Loop counters */
	197	+ size_t i;
	198	+ size_t j;
	199	+
	200	+ /* Instruction slot (0, 1, or 2) in the 128-bit instruction word */
	201	+ uint32_t slot;
	202	+
	203	+ /* Bitwise offset of the instruction indicated by slot */
	204	+ uint32_t bit_pos;
	205	+
	206	+ /* bit_pos split into byte and bit parts */
	207	+ uint32_t byte_pos;
	208	+ uint32_t bit_res;
	209	+
	210	+ /* Address part of an instruction */
	211	+ uint32_t addr;
	212	+
	213	+ /* Mask used to detect which instructions to convert */
	214	+ uint32_t mask;
	215	+
	216	+ /* 41-bit instruction stored somewhere in the lowest 48 bits */
	217	+ uint64_t instr;
	218	+
	219	+ /* Instruction normalized with bit_res for easier manipulation */
	220	+ uint64_t norm;
	221	+
	222	+ for (i = 0; i + 16 <= size; i += 16) {
	223	+ mask = branch_table[buf[i] & 0x1F];
	224	+ for (slot = 0, bit_pos = 5; slot < 3; ++slot, bit_pos += 41) {
	225	+ if (((mask >> slot) & 1) == 0)
	226	+ continue;
	227	+
	228	+ byte_pos = bit_pos >> 3;
	229	+ bit_res = bit_pos & 7;
	230	+ instr = 0;
	231	+ for (j = 0; j < 6; ++j)
	232	+ instr \|= (uint64_t)(buf[i + j + byte_pos])
	233	+ << (8 * j);
	234	+
	235	+ norm = instr >> bit_res;
	236	+
	237	+ if (((norm >> 37) & 0x0F) == 0x05
	238	+ && ((norm >> 9) & 0x07) == 0) {
	239	+ addr = (norm >> 13) & 0x0FFFFF;
	240	+ addr \|= ((uint32_t)(norm >> 36) & 1) << 20;
	241	+ addr <<= 4;
	242	+ addr -= s->pos + (uint32_t)i;
	243	+ addr >>= 4;
	244	+
	245	+ norm &= ~((uint64_t)0x8FFFFF << 13);
	246	+ norm \|= (uint64_t)(addr & 0x0FFFFF) << 13;
	247	+ norm \|= (uint64_t)(addr & 0x100000)
	248	+ << (36 - 20);
	249	+
	250	+ instr &= (1 << bit_res) - 1;
	251	+ instr \|= norm << bit_res;
	252	+
	253	+ for (j = 0; j < 6; j++)
	254	+ buf[i + j + byte_pos]
	255	+ = (uint8_t)(instr >> (8 * j));
	256	+ }
	257	+ }
	258	+ }
	259	+
	260	+ return i;
	261	+}
	262	+#endif
	263	+
	264	+#ifdef XZ_DEC_ARM
	265	+static size_t bcj_arm(struct xz_dec_bcj s, uint8_t buf, size_t size)
	266	+{
	267	+ size_t i;
	268	+ uint32_t addr;
	269	+
	270	+ for (i = 0; i + 4 <= size; i += 4) {
	271	+ if (buf[i + 3] == 0xEB) {
	272	+ addr = (uint32_t)buf[i] \| ((uint32_t)buf[i + 1] << 8)
	273	+ \| ((uint32_t)buf[i + 2] << 16);
	274	+ addr <<= 2;
	275	+ addr -= s->pos + (uint32_t)i + 8;
	276	+ addr >>= 2;
	277	+ buf[i] = (uint8_t)addr;
	278	+ buf[i + 1] = (uint8_t)(addr >> 8);
	279	+ buf[i + 2] = (uint8_t)(addr >> 16);
	280	+ }
	281	+ }
	282	+
	283	+ return i;
	284	+}
	285	+#endif
	286	+
	287	+#ifdef XZ_DEC_ARMTHUMB
	288	+static size_t bcj_armthumb(struct xz_dec_bcj s, uint8_t buf, size_t size)
	289	+{
	290	+ size_t i;
	291	+ uint32_t addr;
	292	+
	293	+ for (i = 0; i + 4 <= size; i += 2) {
	294	+ if ((buf[i + 1] & 0xF8) == 0xF0
	295	+ && (buf[i + 3] & 0xF8) == 0xF8) {
	296	+ addr = (((uint32_t)buf[i + 1] & 0x07) << 19)
	297	+ \| ((uint32_t)buf[i] << 11)
	298	+ \| (((uint32_t)buf[i + 3] & 0x07) << 8)
	299	+ \| (uint32_t)buf[i + 2];
	300	+ addr <<= 1;
	301	+ addr -= s->pos + (uint32_t)i + 4;
	302	+ addr >>= 1;
	303	+ buf[i + 1] = (uint8_t)(0xF0 \| ((addr >> 19) & 0x07));
	304	+ buf[i] = (uint8_t)(addr >> 11);
	305	+ buf[i + 3] = (uint8_t)(0xF8 \| ((addr >> 8) & 0x07));
	306	+ buf[i + 2] = (uint8_t)addr;
	307	+ i += 2;
	308	+ }
	309	+ }
	310	+
	311	+ return i;
	312	+}
	313	+#endif
	314	+
	315	+#ifdef XZ_DEC_SPARC
	316	+static size_t bcj_sparc(struct xz_dec_bcj s, uint8_t buf, size_t size)
	317	+{
	318	+ size_t i;
	319	+ uint32_t instr;
	320	+
	321	+ for (i = 0; i + 4 <= size; i += 4) {
	322	+ instr = get_unaligned_be32(buf + i);
	323	+ if ((instr >> 22) == 0x100 \|\| (instr >> 22) == 0x1FF) {
	324	+ instr <<= 2;
	325	+ instr -= s->pos + (uint32_t)i;
	326	+ instr >>= 2;
	327	+ instr = ((uint32_t)0x40000000 - (instr & 0x400000))
	328	+ \| 0x40000000 \| (instr & 0x3FFFFF);
	329	+ put_unaligned_be32(instr, buf + i);
	330	+ }
	331	+ }
	332	+
	333	+ return i;
	334	+}
	335	+#endif
	336	+
	337	+/*
	338	+ * Apply the selected BCJ filter. Update *pos and s->pos to match the amount
	339	+ * of data that got filtered.
	340	+ *
	341	+ * NOTE: This is implemented as a switch statement to avoid using function
	342	+ * pointers, which could be problematic in the kernel boot code, which must
	343	+ * avoid pointers to static data (at least on x86).
	344	+ */
	345	+static void bcj_apply(struct xz_dec_bcj *s,
	346	+ uint8_t buf, size_t pos, size_t size)
	347	+{
	348	+ size_t filtered;
	349	+
	350	+ buf += *pos;
	351	+ size -= *pos;
	352	+
	353	+ switch (s->type) {
	354	+#ifdef XZ_DEC_X86
	355	+ case BCJ_X86:
	356	+ filtered = bcj_x86(s, buf, size);
	357	+ break;
	358	+#endif
	359	+#ifdef XZ_DEC_POWERPC
	360	+ case BCJ_POWERPC:
	361	+ filtered = bcj_powerpc(s, buf, size);
	362	+ break;
	363	+#endif
	364	+#ifdef XZ_DEC_IA64
	365	+ case BCJ_IA64:
	366	+ filtered = bcj_ia64(s, buf, size);
	367	+ break;
	368	+#endif
	369	+#ifdef XZ_DEC_ARM
	370	+ case BCJ_ARM:
	371	+ filtered = bcj_arm(s, buf, size);
	372	+ break;
	373	+#endif
	374	+#ifdef XZ_DEC_ARMTHUMB
	375	+ case BCJ_ARMTHUMB:
	376	+ filtered = bcj_armthumb(s, buf, size);
	377	+ break;
	378	+#endif
	379	+#ifdef XZ_DEC_SPARC
	380	+ case BCJ_SPARC:
	381	+ filtered = bcj_sparc(s, buf, size);
	382	+ break;
	383	+#endif
	384	+ default:
	385	+ /* Never reached but silence compiler warnings. */
	386	+ filtered = 0;
	387	+ break;
	388	+ }
	389	+
	390	+ *pos += filtered;
	391	+ s->pos += filtered;
	392	+}
	393	+
	394	+/*
	395	+ * Flush pending filtered data from temp to the output buffer.
	396	+ * Move the remaining mixture of possibly filtered and unfiltered
	397	+ * data to the beginning of temp.
	398	+ */
	399	+static void bcj_flush(struct xz_dec_bcj s, struct xz_buf b)
	400	+{
	401	+ size_t copy_size;
	402	+
	403	+ copy_size = min_t(size_t, s->temp.filtered, b->out_size - b->out_pos);
	404	+ memcpy(b->out + b->out_pos, s->temp.buf, copy_size);
	405	+ b->out_pos += copy_size;
	406	+
	407	+ s->temp.filtered -= copy_size;
	408	+ s->temp.size -= copy_size;
	409	+ memmove(s->temp.buf, s->temp.buf + copy_size, s->temp.size);
	410	+}
	411	+
	412	+/*
	413	+ * The BCJ filter functions are primitive in sense that they process the
	414	+ * data in chunks of 1-16 bytes. To hide this issue, this function does
	415	+ * some buffering.
	416	+ */
	417	+XZ_EXTERN enum xz_ret xz_dec_bcj_run(struct xz_dec_bcj *s,
	418	+ struct xz_dec_lzma2 *lzma2,
	419	+ struct xz_buf *b)
	420	+{
	421	+ size_t out_start;
	422	+
	423	+ /*
	424	+ * Flush pending already filtered data to the output buffer. Return
	425	+ * immediatelly if we couldn't flush everything, or if the next
	426	+ * filter in the chain had already returned XZ_STREAM_END.
	427	+ */
	428	+ if (s->temp.filtered > 0) {
	429	+ bcj_flush(s, b);
	430	+ if (s->temp.filtered > 0)
	431	+ return XZ_OK;
	432	+
	433	+ if (s->ret == XZ_STREAM_END)
	434	+ return XZ_STREAM_END;
	435	+ }
	436	+
	437	+ /*
	438	+ * If we have more output space than what is currently pending in
	439	+ * temp, copy the unfiltered data from temp to the output buffer
	440	+ * and try to fill the output buffer by decoding more data from the
	441	+ * next filter in the chain. Apply the BCJ filter on the new data
	442	+ * in the output buffer. If everything cannot be filtered, copy it
	443	+ * to temp and rewind the output buffer position accordingly.
	444	+ */
	445	+ if (s->temp.size < b->out_size - b->out_pos) {
	446	+ out_start = b->out_pos;
	447	+ memcpy(b->out + b->out_pos, s->temp.buf, s->temp.size);
	448	+ b->out_pos += s->temp.size;
	449	+
	450	+ s->ret = xz_dec_lzma2_run(lzma2, b);
	451	+ if (s->ret != XZ_STREAM_END
	452	+ && (s->ret != XZ_OK \|\| s->single_call))
	453	+ return s->ret;
	454	+
	455	+ bcj_apply(s, b->out, &out_start, b->out_pos);
	456	+
	457	+ /*
	458	+ * As an exception, if the next filter returned XZ_STREAM_END,
	459	+ * we can do that too, since the last few bytes that remain
	460	+ * unfiltered are meant to remain unfiltered.
	461	+ */
	462	+ if (s->ret == XZ_STREAM_END)
	463	+ return XZ_STREAM_END;
	464	+
	465	+ s->temp.size = b->out_pos - out_start;
	466	+ b->out_pos -= s->temp.size;
	467	+ memcpy(s->temp.buf, b->out + b->out_pos, s->temp.size);
	468	+ }
	469	+
	470	+ /*
	471	+ * If we have unfiltered data in temp, try to fill by decoding more
	472	+ * data from the next filter. Apply the BCJ filter on temp. Then we
	473	+ * hopefully can fill the actual output buffer by copying filtered
	474	+ * data from temp. A mix of filtered and unfiltered data may be left
	475	+ * in temp; it will be taken care on the next call to this function.
	476	+ */
	477	+ if (s->temp.size > 0) {
	478	+ /* Make b->out{,_pos,_size} temporarily point to s->temp. */
	479	+ s->out = b->out;
	480	+ s->out_pos = b->out_pos;
	481	+ s->out_size = b->out_size;
	482	+ b->out = s->temp.buf;
	483	+ b->out_pos = s->temp.size;
	484	+ b->out_size = sizeof(s->temp.buf);
	485	+
	486	+ s->ret = xz_dec_lzma2_run(lzma2, b);
	487	+
	488	+ s->temp.size = b->out_pos;
	489	+ b->out = s->out;
	490	+ b->out_pos = s->out_pos;
	491	+ b->out_size = s->out_size;
	492	+
	493	+ if (s->ret != XZ_OK && s->ret != XZ_STREAM_END)
	494	+ return s->ret;
	495	+
	496	+ bcj_apply(s, s->temp.buf, &s->temp.filtered, s->temp.size);
	497	+
	498	+ /*
	499	+ * If the next filter returned XZ_STREAM_END, we mark that
	500	+ * everything is filtered, since the last unfiltered bytes
	501	+ * of the stream are meant to be left as is.
	502	+ */
	503	+ if (s->ret == XZ_STREAM_END)
	504	+ s->temp.filtered = s->temp.size;
	505	+
	506	+ bcj_flush(s, b);
	507	+ if (s->temp.filtered > 0)
	508	+ return XZ_OK;
	509	+ }
	510	+
	511	+ return s->ret;
	512	+}
	513	+
	514	+XZ_EXTERN struct xz_dec_bcj *xz_dec_bcj_create(bool single_call)
	515	+{
	516	+ struct xz_dec_bcj s = kmalloc(sizeof(s), GFP_KERNEL);
	517	+ if (s != NULL)
	518	+ s->single_call = single_call;
	519	+
	520	+ return s;
	521	+}
	522	+
	523	+XZ_EXTERN enum xz_ret xz_dec_bcj_reset(struct xz_dec_bcj *s, uint8_t id)
	524	+{
	525	+ switch (id) {
	526	+#ifdef XZ_DEC_X86
	527	+ case BCJ_X86:
	528	+#endif
	529	+#ifdef XZ_DEC_POWERPC
	530	+ case BCJ_POWERPC:
	531	+#endif
	532	+#ifdef XZ_DEC_IA64
	533	+ case BCJ_IA64:
	534	+#endif
	535	+#ifdef XZ_DEC_ARM
	536	+ case BCJ_ARM:
	537	+#endif
	538	+#ifdef XZ_DEC_ARMTHUMB
	539	+ case BCJ_ARMTHUMB:
	540	+#endif
	541	+#ifdef XZ_DEC_SPARC
	542	+ case BCJ_SPARC:
	543	+#endif
	544	+ break;
	545	+
	546	+ default:
	547	+ /* Unsupported Filter ID */
	548	+ return XZ_OPTIONS_ERROR;
	549	+ }
	550	+
	551	+ s->type = id;
	552	+ s->ret = XZ_OK;
	553	+ s->pos = 0;
	554	+ s->x86_prev_mask = 0;
	555	+ s->temp.filtered = 0;
	556	+ s->temp.size = 0;
	557	+
	558	+ return XZ_OK;
	559	+}
	560	+
	561	+#endif
	1	+/*
	2	+ * XZ decoder module information
	3	+ *
	4	+ * Author: Lasse Collin <lasse.collin@tukaani.org>
	5	+ *
	6	+ * This file has been put into the public domain.
	7	+ * You can do whatever you want with this file.
	8	+ */
	9	+
	10	+#include <linux/module.h>
	11	+#include <linux/xz.h>
	12	+
	13	+EXPORT_SYMBOL(xz_dec_init);
	14	+EXPORT_SYMBOL(xz_dec_reset);
	15	+EXPORT_SYMBOL(xz_dec_run);
	16	+EXPORT_SYMBOL(xz_dec_end);
	17	+
	18	+MODULE_DESCRIPTION("XZ decompressor");
	19	+MODULE_VERSION("1.0");
	20	+MODULE_AUTHOR("Lasse Collin <lasse.collin@tukaani.org> and Igor Pavlov");
	21	+
	22	+/*
	23	+ * This code is in the public domain, but in Linux it's simplest to just
	24	+ * say it's GPL and consider the authors as the copyright holders.
	25	+ */
	26	+MODULE_LICENSE("GPL");
	1	+/*
	2	+ * XZ decoder tester
	3	+ *
	4	+ * Author: Lasse Collin <lasse.collin@tukaani.org>
	5	+ *
	6	+ * This file has been put into the public domain.
	7	+ * You can do whatever you want with this file.
	8	+ */
	9	+
	10	+#include <linux/kernel.h>
	11	+#include <linux/module.h>
	12	+#include <linux/fs.h>
	13	+#include <linux/uaccess.h>
	14	+#include <linux/crc32.h>
	15	+#include <linux/xz.h>
	16	+
	17	+/* Maximum supported dictionary size */
	18	+#define DICT_MAX (1 << 20)
	19	+
	20	+/* Device name to pass to register_chrdev(). */
	21	+#define DEVICE_NAME "xz_dec_test"
	22	+
	23	+/* Dynamically allocated device major number */
	24	+static int device_major;
	25	+
	26	+/*
	27	+ * We reuse the same decoder state, and thus can decode only one
	28	+ * file at a time.
	29	+ */
	30	+static bool device_is_open;
	31	+
	32	+/* XZ decoder state */
	33	+static struct xz_dec *state;
	34	+
	35	+/*
	36	+ * Return value of xz_dec_run(). We need to avoid calling xz_dec_run() after
	37	+ * it has returned XZ_STREAM_END, so we make this static.
	38	+ */
	39	+static enum xz_ret ret;
	40	+
	41	+/*
	42	+ * Input and output buffers. The input buffer is used as a temporary safe
	43	+ * place for the data coming from the userspace.
	44	+ */
	45	+static uint8_t buffer_in[1024];
	46	+static uint8_t buffer_out[1024];
	47	+
	48	+/*
	49	+ * Structure to pass the input and output buffers to the XZ decoder.
	50	+ * A few of the fields are never modified so we initialize them here.
	51	+ */
	52	+static struct xz_buf buffers = {
	53	+ .in = buffer_in,
	54	+ .out = buffer_out,
	55	+ .out_size = sizeof(buffer_out)
	56	+};
	57	+
	58	+/*
	59	+ * CRC32 of uncompressed data. This is used to give the user a simple way
	60	+ * to check that the decoder produces correct output.
	61	+ */
	62	+static uint32_t crc;
	63	+
	64	+static int xz_dec_test_open(struct inode i, struct file f)
	65	+{
	66	+ if (device_is_open)
	67	+ return -EBUSY;
	68	+
	69	+ device_is_open = true;
	70	+
	71	+ xz_dec_reset(state);
	72	+ ret = XZ_OK;
	73	+ crc = 0xFFFFFFFF;
	74	+
	75	+ buffers.in_pos = 0;
	76	+ buffers.in_size = 0;
	77	+ buffers.out_pos = 0;
	78	+
	79	+ printk(KERN_INFO DEVICE_NAME ": opened\n");
	80	+ return 0;
	81	+}
	82	+
	83	+static int xz_dec_test_release(struct inode i, struct file f)
	84	+{
	85	+ device_is_open = false;
	86	+
	87	+ if (ret == XZ_OK)
	88	+ printk(KERN_INFO DEVICE_NAME ": input was truncated\n");
	89	+
	90	+ printk(KERN_INFO DEVICE_NAME ": closed\n");
	91	+ return 0;
	92	+}
	93	+
	94	+/*
	95	+ * Decode the data given to us from the userspace. CRC32 of the uncompressed
	96	+ * data is calculated and is printed at the end of successful decoding. The
	97	+ * uncompressed data isn't stored anywhere for further use.
	98	+ *
	99	+ * The .xz file must have exactly one Stream and no Stream Padding. The data
	100	+ * after the first Stream is considered to be garbage.
	101	+ */
	102	+static ssize_t xz_dec_test_write(struct file file, const char __user buf,
	103	+ size_t size, loff_t *pos)
	104	+{
	105	+ size_t remaining;
	106	+
	107	+ if (ret != XZ_OK) {
	108	+ if (size > 0)
	109	+ printk(KERN_INFO DEVICE_NAME ": %zu bytes of "
	110	+ "garbage at the end of the file\n",
	111	+ size);
	112	+
	113	+ return -ENOSPC;
	114	+ }
	115	+
	116	+ printk(KERN_INFO DEVICE_NAME ": decoding %zu bytes of input\n",
	117	+ size);
	118	+
	119	+ remaining = size;
	120	+ while ((remaining > 0 \|\| buffers.out_pos == buffers.out_size)
	121	+ && ret == XZ_OK) {
	122	+ if (buffers.in_pos == buffers.in_size) {
	123	+ buffers.in_pos = 0;
	124	+ buffers.in_size = min(remaining, sizeof(buffer_in));
	125	+ if (copy_from_user(buffer_in, buf, buffers.in_size))
	126	+ return -EFAULT;
	127	+
	128	+ buf += buffers.in_size;
	129	+ remaining -= buffers.in_size;
	130	+ }
	131	+
	132	+ buffers.out_pos = 0;
	133	+ ret = xz_dec_run(state, &buffers);
	134	+ crc = crc32(crc, buffer_out, buffers.out_pos);
	135	+ }
	136	+
	137	+ switch (ret) {
	138	+ case XZ_OK:
	139	+ printk(KERN_INFO DEVICE_NAME ": XZ_OK\n");
	140	+ return size;
	141	+
	142	+ case XZ_STREAM_END:
	143	+ printk(KERN_INFO DEVICE_NAME ": XZ_STREAM_END, "
	144	+ "CRC32 = 0x%08X\n", ~crc);
	145	+ return size - remaining - (buffers.in_size - buffers.in_pos);
	146	+
	147	+ case XZ_MEMLIMIT_ERROR:
	148	+ printk(KERN_INFO DEVICE_NAME ": XZ_MEMLIMIT_ERROR\n");
	149	+ break;
	150	+
	151	+ case XZ_FORMAT_ERROR:
	152	+ printk(KERN_INFO DEVICE_NAME ": XZ_FORMAT_ERROR\n");
	153	+ break;
	154	+
	155	+ case XZ_OPTIONS_ERROR:
	156	+ printk(KERN_INFO DEVICE_NAME ": XZ_OPTIONS_ERROR\n");
	157	+ break;
	158	+
	159	+ case XZ_DATA_ERROR:
	160	+ printk(KERN_INFO DEVICE_NAME ": XZ_DATA_ERROR\n");
	161	+ break;
	162	+
	163	+ case XZ_BUF_ERROR:
	164	+ printk(KERN_INFO DEVICE_NAME ": XZ_BUF_ERROR\n");
	165	+ break;
	166	+
	167	+ default:
	168	+ printk(KERN_INFO DEVICE_NAME ": Bug detected!\n");
	169	+ break;
	170	+ }
	171	+
	172	+ return -EIO;
	173	+}
	174	+
	175	+/* Allocate the XZ decoder state and register the character device. */
	176	+static int __init xz_dec_test_init(void)
	177	+{
	178	+ static const struct file_operations fileops = {
	179	+ .owner = THIS_MODULE,
	180	+ .open = &xz_dec_test_open,
	181	+ .release = &xz_dec_test_release,
	182	+ .write = &xz_dec_test_write
	183	+ };
	184	+
	185	+ state = xz_dec_init(XZ_PREALLOC, DICT_MAX);
	186	+ if (state == NULL)
	187	+ return -ENOMEM;
	188	+
	189	+ device_major = register_chrdev(0, DEVICE_NAME, &fileops);
	190	+ if (device_major < 0) {
	191	+ xz_dec_end(state);
	192	+ return device_major;
	193	+ }
	194	+
	195	+ printk(KERN_INFO DEVICE_NAME ": module loaded\n");
	196	+ printk(KERN_INFO DEVICE_NAME ": Create a device node with "
	197	+ "'mknod " DEVICE_NAME " c %d 0' and write .xz files "
	198	+ "to it.\n", device_major);
	199	+ return 0;
	200	+}
	201	+
	202	+static void __exit xz_dec_test_exit(void)
	203	+{
	204	+ unregister_chrdev(device_major, DEVICE_NAME);
	205	+ xz_dec_end(state);
	206	+ printk(KERN_INFO DEVICE_NAME ": module unloaded\n");
	207	+}
	208	+
	209	+module_init(xz_dec_test_init);
	210	+module_exit(xz_dec_test_exit);
	211	+
	212	+MODULE_DESCRIPTION("XZ decompressor tester");
	213	+MODULE_VERSION("1.0");
	214	+MODULE_AUTHOR("Lasse Collin <lasse.collin@tukaani.org>");
	215	+
	216	+/*
	217	+ * This code is in the public domain, but in Linux it's simplest to just
	218	+ * say it's GPL and consider the authors as the copyright holders.
	219	+ */
	220	+MODULE_LICENSE("GPL");