summaryrefslogtreecommitdiff
path: root/lib
diff options
context:
space:
mode:
authoralk3pInjection <webmaster@raspii.tech>2021-08-30 16:43:38 +0800
committeralk3pInjection <webmaster@raspii.tech>2021-08-30 16:43:38 +0800
commitcbe033a53bfe49d980774e59025e3b2af91778b7 (patch)
tree558535f91276162e0be70d07b34ed2e6577e38ad /lib
parentfdd43c66dd9e77283aa8f7e52a881be44d622441 (diff)
parentd44371841a2f1728a3f36839fd4b7e872d0927d3 (diff)
Merge tag 'v1.9.3' into lineage-18.1HEADlineage-18.1
Change-Id: Iad56c1b17a32f9f356a4c1ff9557f0e79addf481
Diffstat (limited to 'lib')
-rw-r--r--lib/Makefile96
-rw-r--r--lib/README.md84
-rw-r--r--lib/dll/liblz4.def62
-rw-r--r--lib/liblz4-dll.rc.in35
-rw-r--r--lib/lz4.c1452
-rw-r--r--lib/lz4.h599
-rw-r--r--lib/lz4frame.c593
-rw-r--r--lib/lz4frame.h188
-rw-r--r--lib/lz4hc.c689
-rw-r--r--lib/lz4hc.h286
-rw-r--r--lib/xxhash.c614
-rw-r--r--lib/xxhash.h221
12 files changed, 3136 insertions, 1783 deletions
diff --git a/lib/Makefile b/lib/Makefile
index 88d9b4f..c12949b 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -28,7 +28,7 @@
# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# You can contact the author at :
-# - LZ4 source repository : https://github.com/Cyan4973/lz4
+# - LZ4 source repository : https://github.com/lz4/lz4
# - LZ4 forum froup : https://groups.google.com/forum/#!forum/lz4c
# ################################################################
@@ -45,7 +45,6 @@ LIBVER := $(shell echo $(LIBVER_SCRIPT))
BUILD_SHARED:=yes
BUILD_STATIC:=yes
-OS ?= $(shell uname)
CPPFLAGS+= -DXXH_NAMESPACE=LZ4_
CFLAGS ?= -O3
DEBUGFLAGS:= -Wall -Wextra -Wcast-qual -Wcast-align -Wshadow \
@@ -56,10 +55,11 @@ FLAGS = $(CPPFLAGS) $(CFLAGS) $(LDFLAGS)
SRCFILES := $(sort $(wildcard *.c))
+include ../Makefile.inc
# OS X linker doesn't support -soname, and use different extension
# see : https://developer.apple.com/library/mac/documentation/DeveloperTools/Conceptual/DynamicLibraries/100-Articles/DynamicLibraryDesignGuidelines.html
-ifeq ($(OS), Darwin)
+ifeq ($(TARGET_OS), Darwin)
SHARED_EXT = dylib
SHARED_EXT_MAJOR = $(LIBVER_MAJOR).$(SHARED_EXT)
SHARED_EXT_VER = $(LIBVER).$(SHARED_EXT)
@@ -71,8 +71,6 @@ else
SHARED_EXT_VER = $(SHARED_EXT).$(LIBVER)
endif
-LIBLZ4 = liblz4.$(SHARED_EXT_VER)
-
.PHONY: default
default: lib-release
@@ -86,12 +84,6 @@ all: lib
all32: CFLAGS+=-m32
all32: all
-ifeq ($(V), 1)
-Q =
-else
-Q = @
-endif
-
liblz4.a: $(SRCFILES)
ifeq ($(BUILD_STATIC),yes) # can be disabled on command line
@echo compiling static library
@@ -99,32 +91,50 @@ ifeq ($(BUILD_STATIC),yes) # can be disabled on command line
$(Q)$(AR) rcs $@ *.o
endif
+ifeq ($(WINBASED),yes)
+liblz4-dll.rc: liblz4-dll.rc.in
+ @echo creating library resource
+ $(Q)sed -e 's|@LIBLZ4@|$(LIBLZ4)|' \
+ -e 's|@LIBVER_MAJOR@|$(LIBVER_MAJOR)|g' \
+ -e 's|@LIBVER_MINOR@|$(LIBVER_MINOR)|g' \
+ -e 's|@LIBVER_PATCH@|$(LIBVER_PATCH)|g' \
+ $< >$@
+
+liblz4-dll.o: liblz4-dll.rc
+ $(WINDRES) -i liblz4-dll.rc -o liblz4-dll.o
+
+$(LIBLZ4): $(SRCFILES) liblz4-dll.o
+else
$(LIBLZ4): $(SRCFILES)
+endif
ifeq ($(BUILD_SHARED),yes) # can be disabled on command line
@echo compiling dynamic library $(LIBVER)
-ifneq (,$(filter Windows%,$(OS)))
- $(Q)$(CC) $(FLAGS) -DLZ4_DLL_EXPORT=1 -shared $^ -o dll\$@.dll
- dlltool -D dll\liblz4.dll -d dll\liblz4.def -l dll\liblz4.lib
-else
+ ifeq ($(WINBASED),yes)
+ $(Q)$(CC) $(FLAGS) -DLZ4_DLL_EXPORT=1 -shared $^ -o dll/$@.dll -Wl,--out-implib,dll/$(LIBLZ4_EXP)
+ else
$(Q)$(CC) $(FLAGS) -shared $^ -fPIC -fvisibility=hidden $(SONAME_FLAGS) -o $@
@echo creating versioned links
- $(Q)ln -sf $@ liblz4.$(SHARED_EXT_MAJOR)
- $(Q)ln -sf $@ liblz4.$(SHARED_EXT)
-endif
+ $(Q)$(LN_SF) $@ liblz4.$(SHARED_EXT_MAJOR)
+ $(Q)$(LN_SF) $@ liblz4.$(SHARED_EXT)
+ endif
endif
+ifeq (,$(filter MINGW%,$(TARGET_OS)))
liblz4: $(LIBLZ4)
+endif
clean:
- $(Q)$(RM) core *.o liblz4.pc dll/liblz4.dll dll/liblz4.lib
+ifeq ($(WINBASED),yes)
+ $(Q)$(RM) *.rc
+endif
+ $(Q)$(RM) core *.o liblz4.pc dll/$(LIBLZ4).dll dll/$(LIBLZ4_EXP)
$(Q)$(RM) *.a *.$(SHARED_EXT) *.$(SHARED_EXT_MAJOR) *.$(SHARED_EXT_VER)
@echo Cleaning library completed
-
#-----------------------------------------------------------------------------
# make install is validated only for Linux, OSX, BSD, Hurd and Solaris targets
#-----------------------------------------------------------------------------
-ifneq (,$(filter $(shell uname),Linux Darwin GNU/kFreeBSD GNU OpenBSD FreeBSD NetBSD DragonFly SunOS Haiku MidnightBSD))
+ifeq ($(POSIX_ENV),Yes)
.PHONY: listL120
listL120: # extract lines >= 120 characters in *.{c,h}, by Takayuki Matsuoka (note : $$, for Makefile compatibility)
@@ -138,27 +148,20 @@ PREFIX ?= /usr/local
prefix ?= $(PREFIX)
EXEC_PREFIX ?= $(prefix)
exec_prefix ?= $(EXEC_PREFIX)
+BINDIR ?= $(exec_prefix)/bin
+bindir ?= $(BINDIR)
LIBDIR ?= $(exec_prefix)/lib
libdir ?= $(LIBDIR)
INCLUDEDIR ?= $(prefix)/include
includedir ?= $(INCLUDEDIR)
-ifneq (,$(filter $(OS),OpenBSD FreeBSD NetBSD DragonFly))
+ ifneq (,$(filter $(TARGET_OS),OpenBSD FreeBSD NetBSD DragonFly MidnightBSD))
PKGCONFIGDIR ?= $(prefix)/libdata/pkgconfig
-else
+ else
PKGCONFIGDIR ?= $(libdir)/pkgconfig
-endif
+ endif
pkgconfigdir ?= $(PKGCONFIGDIR)
-ifneq (,$(filter $(OS),SunOS))
-INSTALL ?= ginstall
-else
-INSTALL ?= install
-endif
-
-INSTALL_PROGRAM ?= $(INSTALL)
-INSTALL_DATA ?= $(INSTALL) -m 644
-
liblz4.pc: liblz4.pc.in Makefile
@echo creating pkgconfig
$(Q)sed -e 's|@PREFIX@|$(prefix)|' \
@@ -168,18 +171,26 @@ liblz4.pc: liblz4.pc.in Makefile
$< >$@
install: lib liblz4.pc
- $(Q)$(INSTALL) -d -m 755 $(DESTDIR)$(pkgconfigdir)/ $(DESTDIR)$(includedir)/ $(DESTDIR)$(libdir)/
+ $(Q)$(INSTALL_DIR) $(DESTDIR)$(pkgconfigdir)/ $(DESTDIR)$(includedir)/ $(DESTDIR)$(libdir)/ $(DESTDIR)$(bindir)/
$(Q)$(INSTALL_DATA) liblz4.pc $(DESTDIR)$(pkgconfigdir)/
@echo Installing libraries
-ifeq ($(BUILD_STATIC),yes)
+ ifeq ($(BUILD_STATIC),yes)
$(Q)$(INSTALL_DATA) liblz4.a $(DESTDIR)$(libdir)/liblz4.a
$(Q)$(INSTALL_DATA) lz4frame_static.h $(DESTDIR)$(includedir)/lz4frame_static.h
-endif
-ifeq ($(BUILD_SHARED),yes)
+ endif
+ ifeq ($(BUILD_SHARED),yes)
+# Traditionnally, one installs the DLLs in the bin directory as programs
+# search them first in their directory. This allows to not pollute system
+# directories (like c:/windows/system32), nor modify the PATH variable.
+ ifeq ($(WINBASED),yes)
+ $(Q)$(INSTALL_PROGRAM) dll/$(LIBLZ4).dll $(DESTDIR)$(bindir)
+ $(Q)$(INSTALL_PROGRAM) dll/$(LIBLZ4_EXP) $(DESTDIR)$(libdir)
+ else
$(Q)$(INSTALL_PROGRAM) liblz4.$(SHARED_EXT_VER) $(DESTDIR)$(libdir)
- $(Q)ln -sf liblz4.$(SHARED_EXT_VER) $(DESTDIR)$(libdir)/liblz4.$(SHARED_EXT_MAJOR)
- $(Q)ln -sf liblz4.$(SHARED_EXT_VER) $(DESTDIR)$(libdir)/liblz4.$(SHARED_EXT)
-endif
+ $(Q)$(LN_SF) liblz4.$(SHARED_EXT_VER) $(DESTDIR)$(libdir)/liblz4.$(SHARED_EXT_MAJOR)
+ $(Q)$(LN_SF) liblz4.$(SHARED_EXT_VER) $(DESTDIR)$(libdir)/liblz4.$(SHARED_EXT)
+ endif
+ endif
@echo Installing headers in $(includedir)
$(Q)$(INSTALL_DATA) lz4.h $(DESTDIR)$(includedir)/lz4.h
$(Q)$(INSTALL_DATA) lz4hc.h $(DESTDIR)$(includedir)/lz4hc.h
@@ -188,9 +199,14 @@ endif
uninstall:
$(Q)$(RM) $(DESTDIR)$(pkgconfigdir)/liblz4.pc
+ ifeq (WINBASED,1)
+ $(Q)$(RM) $(DESTDIR)$(bindir)/$(LIBLZ4).dll
+ $(Q)$(RM) $(DESTDIR)$(libdir)/$(LIBLZ4_EXP)
+ else
$(Q)$(RM) $(DESTDIR)$(libdir)/liblz4.$(SHARED_EXT)
$(Q)$(RM) $(DESTDIR)$(libdir)/liblz4.$(SHARED_EXT_MAJOR)
$(Q)$(RM) $(DESTDIR)$(libdir)/liblz4.$(SHARED_EXT_VER)
+ endif
$(Q)$(RM) $(DESTDIR)$(libdir)/liblz4.a
$(Q)$(RM) $(DESTDIR)$(includedir)/lz4.h
$(Q)$(RM) $(DESTDIR)$(includedir)/lz4hc.h
diff --git a/lib/README.md b/lib/README.md
index 7082fe3..e2af868 100644
--- a/lib/README.md
+++ b/lib/README.md
@@ -7,8 +7,8 @@ not all of them are necessary.
#### Minimal LZ4 build
The minimum required is **`lz4.c`** and **`lz4.h`**,
-which provides the fast compression and decompression algorithm.
-They generate and decode data using [LZ4 block format].
+which provides the fast compression and decompression algorithms.
+They generate and decode data using the [LZ4 block format].
#### High Compression variant
@@ -16,13 +16,14 @@ They generate and decode data using [LZ4 block format].
For more compression ratio at the cost of compression speed,
the High Compression variant called **lz4hc** is available.
Add files **`lz4hc.c`** and **`lz4hc.h`**.
-The variant still depends on regular `lib/lz4.*` source files.
+This variant also compresses data using the [LZ4 block format],
+and depends on regular `lib/lz4.*` source files.
-#### Frame variant, for interoperability
+#### Frame support, for interoperability
In order to produce compressed data compatible with `lz4` command line utility,
-it's necessary to encode lz4-compressed blocks using the [official interoperable frame format].
+it's necessary to use the [official interoperable frame format].
This format is generated and decoded automatically by the **lz4frame** library.
Its public API is described in `lib/lz4frame.h`.
In order to work properly, lz4frame needs all other modules present in `/lib`,
@@ -32,15 +33,78 @@ So it's necessary to include all `*.c` and `*.h` files present in `/lib`.
#### Advanced / Experimental API
-A complex API defined in `lz4frame_static.h` contains definitions
-which are not guaranteed to remain stable in future versions.
-As a consequence, it must be used with static linking ***only***.
+Definitions which are not guaranteed to remain stable in future versions,
+are protected behind macros, such as `LZ4_STATIC_LINKING_ONLY`.
+As the name strongly implies, these definitions should only be invoked
+in the context of static linking ***only***.
+Otherwise, dependent application may fail on API or ABI break in the future.
+The associated symbols are also not exposed by the dynamic library by default.
+Should they be nonetheless needed, it's possible to force their publication
+by using build macros `LZ4_PUBLISH_STATIC_FUNCTIONS`
+and `LZ4F_PUBLISH_STATIC_FUNCTIONS`.
+
+
+#### Build macros
+
+The following build macro can be selected to adjust source code behavior at compilation time :
+
+- `LZ4_FAST_DEC_LOOP` : this triggers a speed optimized decompression loop, more powerful on modern cpus.
+ This loop works great on `x86`, `x64` and `aarch64` cpus, and is automatically enabled for them.
+ It's also possible to enable or disable it manually, by passing `LZ4_FAST_DEC_LOOP=1` or `0` to the preprocessor.
+ For example, with `gcc` : `-DLZ4_FAST_DEC_LOOP=1`,
+ and with `make` : `CPPFLAGS+=-DLZ4_FAST_DEC_LOOP=1 make lz4`.
+
+- `LZ4_DISTANCE_MAX` : control the maximum offset that the compressor will allow.
+ Set to 65535 by default, which is the maximum value supported by lz4 format.
+ Reducing maximum distance will reduce opportunities for LZ4 to find matches,
+ hence will produce a worse compression ratio.
+ However, a smaller max distance can allow compatibility with specific decoders using limited memory budget.
+ This build macro only influences the compressed output of the compressor.
+
+- `LZ4_DISABLE_DEPRECATE_WARNINGS` : invoking a deprecated function will make the compiler generate a warning.
+ This is meant to invite users to update their source code.
+ Should this be a problem, it's generally possible to make the compiler ignore these warnings,
+ for example with `-Wno-deprecated-declarations` on `gcc`,
+ or `_CRT_SECURE_NO_WARNINGS` for Visual Studio.
+ This build macro offers another project-specific method
+ by defining `LZ4_DISABLE_DEPRECATE_WARNINGS` before including the LZ4 header files.
+
+- `LZ4_USER_MEMORY_FUNCTIONS` : replace calls to <stdlib>'s `malloc`, `calloc` and `free`
+ by user-defined functions, which must be called `LZ4_malloc()`, `LZ4_calloc()` and `LZ4_free()`.
+ User functions must be available at link time.
+
+- `LZ4_FORCE_SW_BITCOUNT` : by default, the compression algorithm tries to determine lengths
+ by using bitcount instructions, generally implemented as fast single instructions in many cpus.
+ In case the target cpus doesn't support it, or compiler intrinsic doesn't work, or feature bad performance,
+ it's possible to use an optimized software path instead.
+ This is achieved by setting this build macros .
+ In most cases, it's not expected to be necessary,
+ but it can be legitimately considered for less common platforms.
+
+- `LZ4_ALIGN_TEST` : alignment test ensures that the memory area
+ passed as argument to become a compression state is suitably aligned.
+ This test can be disabled if it proves flaky, by setting this value to 0.
+
+
+#### Amalgamation
+
+lz4 source code can be amalgamated into a single file.
+One can combine all source code into `lz4_all.c` by using following command:
+```
+cat lz4.c lz4hc.c lz4frame.c > lz4_all.c
+```
+(`cat` file order is important) then compile `lz4_all.c`.
+All `*.h` files present in `/lib` remain necessary to compile `lz4_all.c`.
#### Windows : using MinGW+MSYS to create DLL
DLL can be created using MinGW+MSYS with the `make liblz4` command.
This command creates `dll\liblz4.dll` and the import library `dll\liblz4.lib`.
+To override the `dlltool` command when cross-compiling on Linux, just set the `DLLTOOL` variable. Example of cross compilation on Linux with mingw-w64 64 bits:
+```
+make BUILD_STATIC=no CC=x86_64-w64-mingw32-gcc DLLTOOL=x86_64-w64-mingw32-dlltool OS=Windows_NT
+```
The import library is only required with Visual C++.
The header files `lz4.h`, `lz4hc.h`, `lz4frame.h` and the dynamic library
`dll\liblz4.dll` are required to compile a project using gcc/MinGW.
@@ -48,14 +112,14 @@ The dynamic library has to be added to linking options.
It means that if a project that uses LZ4 consists of a single `test-dll.c`
file it should be linked with `dll\liblz4.dll`. For example:
```
- gcc $(CFLAGS) -Iinclude/ test-dll.c -o test-dll dll\liblz4.dll
+ $(CC) $(CFLAGS) -Iinclude/ test-dll.c -o test-dll dll\liblz4.dll
```
The compiled executable will require LZ4 DLL which is available at `dll\liblz4.dll`.
#### Miscellaneous
-Other files present in the directory are not source code. There are :
+Other files present in the directory are not source code. They are :
- `LICENSE` : contains the BSD license text
- `Makefile` : `make` script to compile and install lz4 library (static and dynamic)
diff --git a/lib/dll/liblz4.def b/lib/dll/liblz4.def
deleted file mode 100644
index 0ace223..0000000
--- a/lib/dll/liblz4.def
+++ /dev/null
@@ -1,62 +0,0 @@
-LIBRARY liblz4.dll
-EXPORTS
- LZ4F_compressBegin
- LZ4F_compressBound
- LZ4F_compressEnd
- LZ4F_compressFrame
- LZ4F_compressFrameBound
- LZ4F_compressUpdate
- LZ4F_createCompressionContext
- LZ4F_createDecompressionContext
- LZ4F_decompress
- LZ4F_flush
- LZ4F_freeCompressionContext
- LZ4F_freeDecompressionContext
- LZ4F_getErrorName
- LZ4F_getFrameInfo
- LZ4F_getVersion
- LZ4F_isError
- LZ4_compress
- LZ4_compressBound
- LZ4_compressHC
- LZ4_compressHC_continue
- LZ4_compressHC_limitedOutput
- LZ4_compressHC_limitedOutput_continue
- LZ4_compressHC_limitedOutput_withStateHC
- LZ4_compressHC_withStateHC
- LZ4_compress_HC
- LZ4_compress_HC_continue
- LZ4_compress_HC_extStateHC
- LZ4_compress_continue
- LZ4_compress_default
- LZ4_compress_destSize
- LZ4_compress_fast
- LZ4_compress_fast_continue
- LZ4_compress_fast_extState
- LZ4_compress_limitedOutput
- LZ4_compress_limitedOutput_continue
- LZ4_compress_limitedOutput_withState
- LZ4_compress_withState
- LZ4_createStream
- LZ4_createStreamDecode
- LZ4_createStreamHC
- LZ4_decompress_fast
- LZ4_decompress_fast_continue
- LZ4_decompress_fast_usingDict
- LZ4_decompress_safe
- LZ4_decompress_safe_continue
- LZ4_decompress_safe_partial
- LZ4_decompress_safe_usingDict
- LZ4_freeStream
- LZ4_freeStreamDecode
- LZ4_freeStreamHC
- LZ4_loadDict
- LZ4_loadDictHC
- LZ4_resetStream
- LZ4_resetStreamHC
- LZ4_saveDict
- LZ4_saveDictHC
- LZ4_setStreamDecode
- LZ4_sizeofState
- LZ4_sizeofStateHC
- LZ4_versionNumber
diff --git a/lib/liblz4-dll.rc.in b/lib/liblz4-dll.rc.in
new file mode 100644
index 0000000..bf9adf5
--- /dev/null
+++ b/lib/liblz4-dll.rc.in
@@ -0,0 +1,35 @@
+#include <windows.h>
+
+// DLL version information.
+1 VERSIONINFO
+FILEVERSION @LIBVER_MAJOR@,@LIBVER_MINOR@,@LIBVER_PATCH@,0
+PRODUCTVERSION @LIBVER_MAJOR@,@LIBVER_MINOR@,@LIBVER_PATCH@,0
+FILEFLAGSMASK VS_FFI_FILEFLAGSMASK
+#ifdef _DEBUG
+ FILEFLAGS VS_FF_DEBUG | VS_FF_PRERELEASE
+#else
+ FILEFLAGS 0
+#endif
+FILEOS VOS_NT_WINDOWS32
+FILETYPE VFT_DLL
+FILESUBTYPE VFT2_UNKNOWN
+BEGIN
+ BLOCK "StringFileInfo"
+ BEGIN
+ BLOCK "040904B0"
+ BEGIN
+ VALUE "CompanyName", "Yann Collet"
+ VALUE "FileDescription", "Extremely fast compression"
+ VALUE "FileVersion", "@LIBVER_MAJOR@.@LIBVER_MINOR@.@LIBVER_PATCH@.0"
+ VALUE "InternalName", "@LIBLZ4@"
+ VALUE "LegalCopyright", "Copyright (C) 2013-2016, Yann Collet"
+ VALUE "OriginalFilename", "@LIBLZ4@.dll"
+ VALUE "ProductName", "LZ4"
+ VALUE "ProductVersion", "@LIBVER_MAJOR@.@LIBVER_MINOR@.@LIBVER_PATCH@.0"
+ END
+ END
+ BLOCK "VarFileInfo"
+ BEGIN
+ VALUE "Translation", 0x0409, 1200
+ END
+END
diff --git a/lib/lz4.c b/lib/lz4.c
index 4046102..9f5e9bf 100644
--- a/lib/lz4.c
+++ b/lib/lz4.c
@@ -32,7 +32,6 @@
- LZ4 source repository : https://github.com/lz4/lz4
*/
-
/*-************************************
* Tuning parameters
**************************************/
@@ -46,10 +45,16 @@
#endif
/*
- * ACCELERATION_DEFAULT :
+ * LZ4_ACCELERATION_DEFAULT :
* Select "acceleration" for LZ4_compress_fast() when parameter value <= 0
*/
-#define ACCELERATION_DEFAULT 1
+#define LZ4_ACCELERATION_DEFAULT 1
+/*
+ * LZ4_ACCELERATION_MAX :
+ * Any "acceleration" value higher than this threshold
+ * get treated as LZ4_ACCELERATION_MAX instead (fix #876)
+ */
+#define LZ4_ACCELERATION_MAX 65537
/*-************************************
@@ -83,6 +88,7 @@
* Define this parameter if your target system or compiler does not support hardware bit count
*/
#if defined(_MSC_VER) && defined(_WIN32_WCE) /* Visual Studio for WinCE doesn't support Hardware bit count */
+# undef LZ4_FORCE_SW_BITCOUNT /* avoid double def */
# define LZ4_FORCE_SW_BITCOUNT
#endif
@@ -91,8 +97,23 @@
/*-************************************
* Dependency
**************************************/
+/*
+ * LZ4_SRC_INCLUDED:
+ * Amalgamation flag, whether lz4.c is included
+ */
+#ifndef LZ4_SRC_INCLUDED
+# define LZ4_SRC_INCLUDED 1
+#endif
+
+#ifndef LZ4_STATIC_LINKING_ONLY
#define LZ4_STATIC_LINKING_ONLY
+#endif
+
+#ifndef LZ4_DISABLE_DEPRECATE_WARNINGS
#define LZ4_DISABLE_DEPRECATE_WARNINGS /* due to LZ4_decompress_safe_withPrefix64k */
+#endif
+
+#define LZ4_STATIC_LINKING_ONLY /* LZ4_DISTANCE_MAX */
#include "lz4.h"
/* see also "memory routines" below */
@@ -100,10 +121,9 @@
/*-************************************
* Compiler Options
**************************************/
-#ifdef _MSC_VER /* Visual Studio */
-# include <intrin.h>
-# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
-# pragma warning(disable : 4293) /* disable: C4293: too large shift (32-bits) */
+#if defined(_MSC_VER) && (_MSC_VER >= 1400) /* Visual Studio 2005+ */
+# include <intrin.h> /* only present in VS2005+ */
+# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
#endif /* _MSC_VER */
#ifndef LZ4_FORCE_INLINE
@@ -122,8 +142,8 @@
# endif /* _MSC_VER */
#endif /* LZ4_FORCE_INLINE */
-/* LZ4_FORCE_O2_GCC_PPC64LE and LZ4_FORCE_O2_INLINE_GCC_PPC64LE
- * Gcc on ppc64le generates an unrolled SIMDized loop for LZ4_wildCopy,
+/* LZ4_FORCE_O2 and LZ4_FORCE_INLINE
+ * gcc on ppc64le generates an unrolled SIMDized loop for LZ4_wildCopy8,
* together with a simple 8-byte copy loop as a fall-back path.
* However, this optimization hurts the decompression speed by >30%,
* because the execution does not go to the optimized loop
@@ -131,17 +151,17 @@
* before going to the fall-back path become useless overhead.
* This optimization happens only with the -O3 flag, and -O2 generates
* a simple 8-byte copy loop.
- * With gcc on ppc64le, all of the LZ4_decompress_* and LZ4_wildCopy
+ * With gcc on ppc64le, all of the LZ4_decompress_* and LZ4_wildCopy8
* functions are annotated with __attribute__((optimize("O2"))),
- * and also LZ4_wildCopy is forcibly inlined, so that the O2 attribute
- * of LZ4_wildCopy does not affect the compression speed.
+ * and also LZ4_wildCopy8 is forcibly inlined, so that the O2 attribute
+ * of LZ4_wildCopy8 does not affect the compression speed.
*/
-#if defined(__PPC64__) && defined(__LITTLE_ENDIAN__) && defined(__GNUC__)
-# define LZ4_FORCE_O2_GCC_PPC64LE __attribute__((optimize("O2")))
-# define LZ4_FORCE_O2_INLINE_GCC_PPC64LE __attribute__((optimize("O2"))) LZ4_FORCE_INLINE
+#if defined(__PPC64__) && defined(__LITTLE_ENDIAN__) && defined(__GNUC__) && !defined(__clang__)
+# define LZ4_FORCE_O2 __attribute__((optimize("O2")))
+# undef LZ4_FORCE_INLINE
+# define LZ4_FORCE_INLINE static __inline __attribute__((optimize("O2"),always_inline))
#else
-# define LZ4_FORCE_O2_GCC_PPC64LE
-# define LZ4_FORCE_O2_INLINE_GCC_PPC64LE static
+# define LZ4_FORCE_O2
#endif
#if (defined(__GNUC__) && (__GNUC__ >= 3)) || (defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 800)) || defined(__clang__)
@@ -157,21 +177,100 @@
#define unlikely(expr) expect((expr) != 0, 0)
#endif
+/* Should the alignment test prove unreliable, for some reason,
+ * it can be disabled by setting LZ4_ALIGN_TEST to 0 */
+#ifndef LZ4_ALIGN_TEST /* can be externally provided */
+# define LZ4_ALIGN_TEST 1
+#endif
+
/*-************************************
* Memory routines
**************************************/
-#include <stdlib.h> /* malloc, calloc, free */
-#define ALLOC(s) malloc(s)
-#define ALLOC_AND_ZERO(s) calloc(1,s)
-#define FREEMEM(p) free(p)
+#ifdef LZ4_USER_MEMORY_FUNCTIONS
+/* memory management functions can be customized by user project.
+ * Below functions must exist somewhere in the Project
+ * and be available at link time */
+void* LZ4_malloc(size_t s);
+void* LZ4_calloc(size_t n, size_t s);
+void LZ4_free(void* p);
+# define ALLOC(s) LZ4_malloc(s)
+# define ALLOC_AND_ZERO(s) LZ4_calloc(1,s)
+# define FREEMEM(p) LZ4_free(p)
+#else
+# include <stdlib.h> /* malloc, calloc, free */
+# define ALLOC(s) malloc(s)
+# define ALLOC_AND_ZERO(s) calloc(1,s)
+# define FREEMEM(p) free(p)
+#endif
+
#include <string.h> /* memset, memcpy */
#define MEM_INIT(p,v,s) memset((p),(v),(s))
/*-************************************
-* Basic Types
+* Common Constants
**************************************/
+#define MINMATCH 4
+
+#define WILDCOPYLENGTH 8
+#define LASTLITERALS 5 /* see ../doc/lz4_Block_format.md#parsing-restrictions */
+#define MFLIMIT 12 /* see ../doc/lz4_Block_format.md#parsing-restrictions */
+#define MATCH_SAFEGUARD_DISTANCE ((2*WILDCOPYLENGTH) - MINMATCH) /* ensure it's possible to write 2 x wildcopyLength without overflowing output buffer */
+#define FASTLOOP_SAFE_DISTANCE 64
+static const int LZ4_minLength = (MFLIMIT+1);
+
+#define KB *(1 <<10)
+#define MB *(1 <<20)
+#define GB *(1U<<30)
+
+#define LZ4_DISTANCE_ABSOLUTE_MAX 65535
+#if (LZ4_DISTANCE_MAX > LZ4_DISTANCE_ABSOLUTE_MAX) /* max supported by LZ4 format */
+# error "LZ4_DISTANCE_MAX is too big : must be <= 65535"
+#endif
+
+#define ML_BITS 4
+#define ML_MASK ((1U<<ML_BITS)-1)
+#define RUN_BITS (8-ML_BITS)
+#define RUN_MASK ((1U<<RUN_BITS)-1)
+
+
+/*-************************************
+* Error detection
+**************************************/
+#if defined(LZ4_DEBUG) && (LZ4_DEBUG>=1)
+# include <assert.h>
+#else
+# ifndef assert
+# define assert(condition) ((void)0)
+# endif
+#endif
+
+#define LZ4_STATIC_ASSERT(c) { enum { LZ4_static_assert = 1/(int)(!!(c)) }; } /* use after variable declarations */
+
+#if defined(LZ4_DEBUG) && (LZ4_DEBUG>=2)
+# include <stdio.h>
+ static int g_debuglog_enable = 1;
+# define DEBUGLOG(l, ...) { \
+ if ((g_debuglog_enable) && (l<=LZ4_DEBUG)) { \
+ fprintf(stderr, __FILE__ ": "); \
+ fprintf(stderr, __VA_ARGS__); \
+ fprintf(stderr, " \n"); \
+ } }
+#else
+# define DEBUGLOG(l, ...) {} /* disabled */
+#endif
+
+static int LZ4_isAligned(const void* ptr, size_t alignment)
+{
+ return ((size_t)ptr & (alignment -1)) == 0;
+}
+
+
+/*-************************************
+* Types
+**************************************/
+#include <limits.h>
#if defined(__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# include <stdint.h>
typedef uint8_t BYTE;
@@ -181,6 +280,9 @@
typedef uint64_t U64;
typedef uintptr_t uptrval;
#else
+# if UINT_MAX != 4294967295UL
+# error "LZ4 code (when not C++ or C99) assumes that sizeof(int) == 4"
+# endif
typedef unsigned char BYTE;
typedef unsigned short U16;
typedef unsigned int U32;
@@ -195,9 +297,31 @@
typedef size_t reg_t; /* 32-bits in x32 mode */
#endif
+typedef enum {
+ notLimited = 0,
+ limitedOutput = 1,
+ fillOutput = 2
+} limitedOutput_directive;
+
+
/*-************************************
* Reading and writing into memory
**************************************/
+
+/**
+ * LZ4 relies on memcpy with a constant size being inlined. In freestanding
+ * environments, the compiler can't assume the implementation of memcpy() is
+ * standard compliant, so it can't apply its specialized memcpy() inlining
+ * logic. When possible, use __builtin_memcpy() to tell the compiler to analyze
+ * memcpy() as if it were standard compliant, so it can inline it in freestanding
+ * environments. This is needed when decompressing the Linux Kernel, for example.
+ */
+#if defined(__GNUC__) && (__GNUC__ >= 4)
+#define LZ4_memcpy(dst, src, size) __builtin_memcpy(dst, src, size)
+#else
+#define LZ4_memcpy(dst, src, size) memcpy(dst, src, size)
+#endif
+
static unsigned LZ4_isLittleEndian(void)
{
const union { U32 u; BYTE c[4]; } one = { 1 }; /* don't use static : performance detrimental */
@@ -228,31 +352,31 @@ static reg_t LZ4_read_ARCH(const void* ptr) { return ((const unalign*)ptr)->uArc
static void LZ4_write16(void* memPtr, U16 value) { ((unalign*)memPtr)->u16 = value; }
static void LZ4_write32(void* memPtr, U32 value) { ((unalign*)memPtr)->u32 = value; }
-#else /* safe and portable access through memcpy() */
+#else /* safe and portable access using memcpy() */
static U16 LZ4_read16(const void* memPtr)
{
- U16 val; memcpy(&val, memPtr, sizeof(val)); return val;
+ U16 val; LZ4_memcpy(&val, memPtr, sizeof(val)); return val;
}
static U32 LZ4_read32(const void* memPtr)
{
- U32 val; memcpy(&val, memPtr, sizeof(val)); return val;
+ U32 val; LZ4_memcpy(&val, memPtr, sizeof(val)); return val;
}
static reg_t LZ4_read_ARCH(const void* memPtr)
{
- reg_t val; memcpy(&val, memPtr, sizeof(val)); return val;
+ reg_t val; LZ4_memcpy(&val, memPtr, sizeof(val)); return val;
}
static void LZ4_write16(void* memPtr, U16 value)
{
- memcpy(memPtr, &value, sizeof(value));
+ LZ4_memcpy(memPtr, &value, sizeof(value));
}
static void LZ4_write32(void* memPtr, U32 value)
{
- memcpy(memPtr, &value, sizeof(value));
+ LZ4_memcpy(memPtr, &value, sizeof(value));
}
#endif /* LZ4_FORCE_MEMORY_ACCESS */
@@ -280,65 +404,106 @@ static void LZ4_writeLE16(void* memPtr, U16 value)
}
/* customized variant of memcpy, which can overwrite up to 8 bytes beyond dstEnd */
-LZ4_FORCE_O2_INLINE_GCC_PPC64LE
-void LZ4_wildCopy(void* dstPtr, const void* srcPtr, void* dstEnd)
+LZ4_FORCE_INLINE
+void LZ4_wildCopy8(void* dstPtr, const void* srcPtr, void* dstEnd)
{
BYTE* d = (BYTE*)dstPtr;
const BYTE* s = (const BYTE*)srcPtr;
BYTE* const e = (BYTE*)dstEnd;
- do { memcpy(d,s,8); d+=8; s+=8; } while (d<e);
+ do { LZ4_memcpy(d,s,8); d+=8; s+=8; } while (d<e);
}
+static const unsigned inc32table[8] = {0, 1, 2, 1, 0, 4, 4, 4};
+static const int dec64table[8] = {0, 0, 0, -1, -4, 1, 2, 3};
-/*-************************************
-* Common Constants
-**************************************/
-#define MINMATCH 4
-#define WILDCOPYLENGTH 8
-#define LASTLITERALS 5 /* see ../doc/lz4_Block_format.md#parsing-restrictions */
-#define MFLIMIT 12 /* see ../doc/lz4_Block_format.md#parsing-restrictions */
-#define MATCH_SAFEGUARD_DISTANCE ((2*WILDCOPYLENGTH) - MINMATCH) /* ensure it's possible to write 2 x wildcopyLength without overflowing output buffer */
-static const int LZ4_minLength = (MFLIMIT+1);
-
-#define KB *(1 <<10)
-#define MB *(1 <<20)
-#define GB *(1U<<30)
-
-#define MAXD_LOG 16
-#define MAX_DISTANCE ((1 << MAXD_LOG) - 1)
+#ifndef LZ4_FAST_DEC_LOOP
+# if defined __i386__ || defined _M_IX86 || defined __x86_64__ || defined _M_X64
+# define LZ4_FAST_DEC_LOOP 1
+# elif defined(__aarch64__) && !defined(__clang__)
+ /* On aarch64, we disable this optimization for clang because on certain
+ * mobile chipsets, performance is reduced with clang. For information
+ * refer to https://github.com/lz4/lz4/pull/707 */
+# define LZ4_FAST_DEC_LOOP 1
+# else
+# define LZ4_FAST_DEC_LOOP 0
+# endif
+#endif
-#define ML_BITS 4
-#define ML_MASK ((1U<<ML_BITS)-1)
-#define RUN_BITS (8-ML_BITS)
-#define RUN_MASK ((1U<<RUN_BITS)-1)
+#if LZ4_FAST_DEC_LOOP
+
+LZ4_FORCE_INLINE void
+LZ4_memcpy_using_offset_base(BYTE* dstPtr, const BYTE* srcPtr, BYTE* dstEnd, const size_t offset)
+{
+ assert(srcPtr + offset == dstPtr);
+ if (offset < 8) {
+ LZ4_write32(dstPtr, 0); /* silence an msan warning when offset==0 */
+ dstPtr[0] = srcPtr[0];
+ dstPtr[1] = srcPtr[1];
+ dstPtr[2] = srcPtr[2];
+ dstPtr[3] = srcPtr[3];
+ srcPtr += inc32table[offset];
+ LZ4_memcpy(dstPtr+4, srcPtr, 4);
+ srcPtr -= dec64table[offset];
+ dstPtr += 8;
+ } else {
+ LZ4_memcpy(dstPtr, srcPtr, 8);
+ dstPtr += 8;
+ srcPtr += 8;
+ }
+ LZ4_wildCopy8(dstPtr, srcPtr, dstEnd);
+}
-/*-************************************
-* Error detection
-**************************************/
-#if defined(LZ4_DEBUG) && (LZ4_DEBUG>=1)
-# include <assert.h>
-#else
-# ifndef assert
-# define assert(condition) ((void)0)
-# endif
-#endif
+/* customized variant of memcpy, which can overwrite up to 32 bytes beyond dstEnd
+ * this version copies two times 16 bytes (instead of one time 32 bytes)
+ * because it must be compatible with offsets >= 16. */
+LZ4_FORCE_INLINE void
+LZ4_wildCopy32(void* dstPtr, const void* srcPtr, void* dstEnd)
+{
+ BYTE* d = (BYTE*)dstPtr;
+ const BYTE* s = (const BYTE*)srcPtr;
+ BYTE* const e = (BYTE*)dstEnd;
-#define LZ4_STATIC_ASSERT(c) { enum { LZ4_static_assert = 1/(int)(!!(c)) }; } /* use after variable declarations */
+ do { LZ4_memcpy(d,s,16); LZ4_memcpy(d+16,s+16,16); d+=32; s+=32; } while (d<e);
+}
+
+/* LZ4_memcpy_using_offset() presumes :
+ * - dstEnd >= dstPtr + MINMATCH
+ * - there is at least 8 bytes available to write after dstEnd */
+LZ4_FORCE_INLINE void
+LZ4_memcpy_using_offset(BYTE* dstPtr, const BYTE* srcPtr, BYTE* dstEnd, const size_t offset)
+{
+ BYTE v[8];
+
+ assert(dstEnd >= dstPtr + MINMATCH);
+
+ switch(offset) {
+ case 1:
+ MEM_INIT(v, *srcPtr, 8);
+ break;
+ case 2:
+ LZ4_memcpy(v, srcPtr, 2);
+ LZ4_memcpy(&v[2], srcPtr, 2);
+ LZ4_memcpy(&v[4], v, 4);
+ break;
+ case 4:
+ LZ4_memcpy(v, srcPtr, 4);
+ LZ4_memcpy(&v[4], srcPtr, 4);
+ break;
+ default:
+ LZ4_memcpy_using_offset_base(dstPtr, srcPtr, dstEnd, offset);
+ return;
+ }
-#if defined(LZ4_DEBUG) && (LZ4_DEBUG>=2)
-# include <stdio.h>
-static int g_debuglog_enable = 1;
-# define DEBUGLOG(l, ...) { \
- if ((g_debuglog_enable) && (l<=LZ4_DEBUG)) { \
- fprintf(stderr, __FILE__ ": "); \
- fprintf(stderr, __VA_ARGS__); \
- fprintf(stderr, " \n"); \
- } }
-#else
-# define DEBUGLOG(l, ...) {} /* disabled */
+ LZ4_memcpy(dstPtr, v, 8);
+ dstPtr += 8;
+ while (dstPtr < dstEnd) {
+ LZ4_memcpy(dstPtr, v, 8);
+ dstPtr += 8;
+ }
+}
#endif
@@ -347,75 +512,92 @@ static int g_debuglog_enable = 1;
**************************************/
static unsigned LZ4_NbCommonBytes (reg_t val)
{
+ assert(val != 0);
if (LZ4_isLittleEndian()) {
- if (sizeof(val)==8) {
-# if defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ if (sizeof(val) == 8) {
+# if defined(_MSC_VER) && (_MSC_VER >= 1800) && defined(_M_AMD64) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ /* x64 CPUS without BMI support interpret `TZCNT` as `REP BSF` */
+ return (unsigned)_tzcnt_u64(val) >> 3;
+# elif defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
unsigned long r = 0;
- _BitScanForward64( &r, (U64)val );
- return (int)(r>>3);
-# elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
- return (__builtin_ctzll((U64)val) >> 3);
+ _BitScanForward64(&r, (U64)val);
+ return (unsigned)r >> 3;
+# elif (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+ ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+ !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (unsigned)__builtin_ctzll((U64)val) >> 3;
# else
- static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2,
- 0, 3, 1, 3, 1, 4, 2, 7,
- 0, 2, 3, 6, 1, 5, 3, 5,
- 1, 3, 4, 4, 2, 5, 6, 7,
- 7, 0, 1, 2, 3, 3, 4, 6,
- 2, 6, 5, 5, 3, 4, 5, 6,
- 7, 1, 2, 4, 6, 4, 4, 5,
- 7, 2, 6, 5, 7, 6, 7, 7 };
- return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
+ const U64 m = 0x0101010101010101ULL;
+ val ^= val - 1;
+ return (unsigned)(((U64)((val & (m - 1)) * m)) >> 56);
# endif
} else /* 32 bits */ {
-# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+# if defined(_MSC_VER) && (_MSC_VER >= 1400) && !defined(LZ4_FORCE_SW_BITCOUNT)
unsigned long r;
- _BitScanForward( &r, (U32)val );
- return (int)(r>>3);
-# elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
- return (__builtin_ctz((U32)val) >> 3);
+ _BitScanForward(&r, (U32)val);
+ return (unsigned)r >> 3;
+# elif (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+ ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+ !defined(__TINYC__) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (unsigned)__builtin_ctz((U32)val) >> 3;
# else
- static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0,
- 3, 2, 2, 1, 3, 2, 0, 1,
- 3, 3, 1, 2, 2, 2, 2, 0,
- 3, 1, 2, 0, 1, 0, 1, 1 };
- return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
+ const U32 m = 0x01010101;
+ return (unsigned)((((val - 1) ^ val) & (m - 1)) * m) >> 24;
# endif
}
} else /* Big Endian CPU */ {
- if (sizeof(val)==8) { /* 64-bits */
-# if defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
- unsigned long r = 0;
- _BitScanReverse64( &r, val );
- return (unsigned)(r>>3);
-# elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
- return (__builtin_clzll((U64)val) >> 3);
+ if (sizeof(val)==8) {
+# if (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+ ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+ !defined(__TINYC__) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (unsigned)__builtin_clzll((U64)val) >> 3;
# else
+#if 1
+ /* this method is probably faster,
+ * but adds a 128 bytes lookup table */
+ static const unsigned char ctz7_tab[128] = {
+ 7, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ 5, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ 6, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ 5, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0,
+ };
+ U64 const mask = 0x0101010101010101ULL;
+ U64 const t = (((val >> 8) - mask) | val) & mask;
+ return ctz7_tab[(t * 0x0080402010080402ULL) >> 57];
+#else
+ /* this method doesn't consume memory space like the previous one,
+ * but it contains several branches,
+ * that may end up slowing execution */
static const U32 by32 = sizeof(val)*4; /* 32 on 64 bits (goal), 16 on 32 bits.
- Just to avoid some static analyzer complaining about shift by 32 on 32-bits target.
- Note that this code path is never triggered in 32-bits mode. */
+ Just to avoid some static analyzer complaining about shift by 32 on 32-bits target.
+ Note that this code path is never triggered in 32-bits mode. */
unsigned r;
if (!(val>>by32)) { r=4; } else { r=0; val>>=by32; }
if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
r += (!val);
return r;
+#endif
# endif
} else /* 32 bits */ {
-# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
- unsigned long r = 0;
- _BitScanReverse( &r, (unsigned long)val );
- return (unsigned)(r>>3);
-# elif (defined(__clang__) || (defined(__GNUC__) && (__GNUC__>=3))) && !defined(LZ4_FORCE_SW_BITCOUNT)
- return (__builtin_clz((U32)val) >> 3);
+# if (defined(__clang__) || (defined(__GNUC__) && ((__GNUC__ > 3) || \
+ ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4))))) && \
+ !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (unsigned)__builtin_clz((U32)val) >> 3;
# else
- unsigned r;
- if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
- r += (!val);
- return r;
+ val >>= 8;
+ val = ((((val + 0x00FFFF00) | 0x00FFFFFF) + val) |
+ (val + 0x00FF0000)) >> 24;
+ return (unsigned)val ^ 3;
# endif
}
}
}
+
#define STEPSIZE sizeof(reg_t)
LZ4_FORCE_INLINE
unsigned LZ4_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* pInLimit)
@@ -455,7 +637,6 @@ static const U32 LZ4_skipTrigger = 6; /* Increase this value ==> compression ru
/*-************************************
* Local Structures and types
**************************************/
-typedef enum { notLimited = 0, limitedOutput = 1, fillOutput = 2 } limitedOutput_directive;
typedef enum { clearedTable = 0, byPtr, byU32, byU16 } tableType_t;
/**
@@ -491,7 +672,7 @@ typedef enum { noDictIssue = 0, dictSmall } dictIssue_directive;
int LZ4_versionNumber (void) { return LZ4_VERSION_NUMBER; }
const char* LZ4_versionString(void) { return LZ4_VERSION_STRING; }
int LZ4_compressBound(int isize) { return LZ4_COMPRESSBOUND(isize); }
-int LZ4_sizeofState() { return LZ4_STREAMSIZE; }
+int LZ4_sizeofState(void) { return LZ4_STREAMSIZE; }
/*-************************************
@@ -501,9 +682,11 @@ int LZ4_sizeofState() { return LZ4_STREAMSIZE; }
extern "C" {
#endif
-int LZ4_compress_forceExtDict (LZ4_stream_t* LZ4_stream, const char* source, char* dest, int inputSize);
+int LZ4_compress_forceExtDict (LZ4_stream_t* LZ4_dict, const char* source, char* dest, int srcSize);
-int LZ4_decompress_safe_forceExtDict(const char* in, char* out, int inSize, int outSize, const void* dict, size_t dictSize);
+int LZ4_decompress_safe_forceExtDict(const char* source, char* dest,
+ int compressedSize, int maxOutputSize,
+ const void* dictStart, size_t dictSize);
#if defined (__cplusplus)
}
@@ -512,7 +695,7 @@ int LZ4_decompress_safe_forceExtDict(const char* in, char* out, int inSize, int
/*-******************************
* Compression functions
********************************/
-static U32 LZ4_hash4(U32 sequence, tableType_t const tableType)
+LZ4_FORCE_INLINE U32 LZ4_hash4(U32 sequence, tableType_t const tableType)
{
if (tableType == byU16)
return ((sequence * 2654435761U) >> ((MINMATCH*8)-(LZ4_HASHLOG+1)));
@@ -520,15 +703,16 @@ static U32 LZ4_hash4(U32 sequence, tableType_t const tableType)
return ((sequence * 2654435761U) >> ((MINMATCH*8)-LZ4_HASHLOG));
}
-static U32 LZ4_hash5(U64 sequence, tableType_t const tableType)
+LZ4_FORCE_INLINE U32 LZ4_hash5(U64 sequence, tableType_t const tableType)
{
- static const U64 prime5bytes = 889523592379ULL;
- static const U64 prime8bytes = 11400714785074694791ULL;
const U32 hashLog = (tableType == byU16) ? LZ4_HASHLOG+1 : LZ4_HASHLOG;
- if (LZ4_isLittleEndian())
+ if (LZ4_isLittleEndian()) {
+ const U64 prime5bytes = 889523592379ULL;
return (U32)(((sequence << 24) * prime5bytes) >> (64 - hashLog));
- else
+ } else {
+ const U64 prime8bytes = 11400714785074694791ULL;
return (U32)(((sequence >> 24) * prime8bytes) >> (64 - hashLog));
+ }
}
LZ4_FORCE_INLINE U32 LZ4_hashPosition(const void* const p, tableType_t const tableType)
@@ -537,7 +721,19 @@ LZ4_FORCE_INLINE U32 LZ4_hashPosition(const void* const p, tableType_t const tab
return LZ4_hash4(LZ4_read32(p), tableType);
}
-static void LZ4_putIndexOnHash(U32 idx, U32 h, void* tableBase, tableType_t const tableType)
+LZ4_FORCE_INLINE void LZ4_clearHash(U32 h, void* tableBase, tableType_t const tableType)
+{
+ switch (tableType)
+ {
+ default: /* fallthrough */
+ case clearedTable: { /* illegal! */ assert(0); return; }
+ case byPtr: { const BYTE** hashTable = (const BYTE**)tableBase; hashTable[h] = NULL; return; }
+ case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = 0; return; }
+ case byU16: { U16* hashTable = (U16*) tableBase; hashTable[h] = 0; return; }
+ }
+}
+
+LZ4_FORCE_INLINE void LZ4_putIndexOnHash(U32 idx, U32 h, void* tableBase, tableType_t const tableType)
{
switch (tableType)
{
@@ -549,7 +745,7 @@ static void LZ4_putIndexOnHash(U32 idx, U32 h, void* tableBase, tableType_t cons
}
}
-static void LZ4_putPositionOnHash(const BYTE* p, U32 h,
+LZ4_FORCE_INLINE void LZ4_putPositionOnHash(const BYTE* p, U32 h,
void* tableBase, tableType_t const tableType,
const BYTE* srcBase)
{
@@ -574,7 +770,7 @@ LZ4_FORCE_INLINE void LZ4_putPosition(const BYTE* p, void* tableBase, tableType_
* Assumption 1 : only valid if tableType == byU32 or byU16.
* Assumption 2 : h is presumed valid (within limits of hash table)
*/
-static U32 LZ4_getIndexOnHash(U32 h, const void* tableBase, tableType_t tableType)
+LZ4_FORCE_INLINE U32 LZ4_getIndexOnHash(U32 h, const void* tableBase, tableType_t tableType)
{
LZ4_STATIC_ASSERT(LZ4_MEMORY_USAGE > 2);
if (tableType == byU32) {
@@ -597,39 +793,41 @@ static const BYTE* LZ4_getPositionOnHash(U32 h, const void* tableBase, tableType
{ const U16* const hashTable = (const U16*) tableBase; return hashTable[h] + srcBase; } /* default, to ensure a return */
}
-LZ4_FORCE_INLINE const BYTE* LZ4_getPosition(const BYTE* p,
- const void* tableBase, tableType_t tableType,
- const BYTE* srcBase)
+LZ4_FORCE_INLINE const BYTE*
+LZ4_getPosition(const BYTE* p,
+ const void* tableBase, tableType_t tableType,
+ const BYTE* srcBase)
{
U32 const h = LZ4_hashPosition(p, tableType);
return LZ4_getPositionOnHash(h, tableBase, tableType, srcBase);
}
-LZ4_FORCE_INLINE void LZ4_prepareTable(
- LZ4_stream_t_internal* const cctx,
- const int inputSize,
- const tableType_t tableType) {
+LZ4_FORCE_INLINE void
+LZ4_prepareTable(LZ4_stream_t_internal* const cctx,
+ const int inputSize,
+ const tableType_t tableType) {
/* If the table hasn't been used, it's guaranteed to be zeroed out, and is
* therefore safe to use no matter what mode we're in. Otherwise, we figure
* out if it's safe to leave as is or whether it needs to be reset.
*/
- if (cctx->tableType != clearedTable) {
- if (cctx->tableType != tableType
- || (tableType == byU16 && cctx->currentOffset + inputSize >= 0xFFFFU)
- || (tableType == byU32 && cctx->currentOffset > 1 GB)
+ if ((tableType_t)cctx->tableType != clearedTable) {
+ assert(inputSize >= 0);
+ if ((tableType_t)cctx->tableType != tableType
+ || ((tableType == byU16) && cctx->currentOffset + (unsigned)inputSize >= 0xFFFFU)
+ || ((tableType == byU32) && cctx->currentOffset > 1 GB)
|| tableType == byPtr
|| inputSize >= 4 KB)
{
DEBUGLOG(4, "LZ4_prepareTable: Resetting table in %p", cctx);
MEM_INIT(cctx->hashTable, 0, LZ4_HASHTABLESIZE);
cctx->currentOffset = 0;
- cctx->tableType = clearedTable;
+ cctx->tableType = (U32)clearedTable;
} else {
DEBUGLOG(4, "LZ4_prepareTable: Re-use hash table (no reset)");
}
}
- /* Adding a gap, so all previous entries are > MAX_DISTANCE back, is faster
+ /* Adding a gap, so all previous entries are > LZ4_DISTANCE_MAX back, is faster
* than compressing without a gap. However, compressing with
* currentOffset == 0 is faster still, so we preserve that case.
*/
@@ -645,20 +843,25 @@ LZ4_FORCE_INLINE void LZ4_prepareTable(
}
/** LZ4_compress_generic() :
- inlined, to ensure branches are decided at compilation time */
-LZ4_FORCE_INLINE int LZ4_compress_generic(
+ * inlined, to ensure branches are decided at compilation time.
+ * Presumed already validated at this stage:
+ * - source != NULL
+ * - inputSize > 0
+ */
+LZ4_FORCE_INLINE int LZ4_compress_generic_validated(
LZ4_stream_t_internal* const cctx,
const char* const source,
char* const dest,
const int inputSize,
- int *inputConsumed, /* only written when outputLimited == fillOutput */
+ int *inputConsumed, /* only written when outputDirective == fillOutput */
const int maxOutputSize,
- const limitedOutput_directive outputLimited,
+ const limitedOutput_directive outputDirective,
const tableType_t tableType,
const dict_directive dictDirective,
const dictIssue_directive dictIssue,
- const U32 acceleration)
+ const int acceleration)
{
+ int result;
const BYTE* ip = (const BYTE*) source;
U32 const startIndex = cctx->currentOffset;
@@ -674,7 +877,7 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
int const maybe_extMem = (dictDirective == usingExtDict) || (dictDirective == usingDictCtx);
U32 const prefixIdxLimit = startIndex - dictSize; /* used when dictDirective == dictSmall */
- const BYTE* const dictEnd = dictionary + dictSize;
+ const BYTE* const dictEnd = dictionary ? dictionary + dictSize : dictionary;
const BYTE* anchor = (const BYTE*) source;
const BYTE* const iend = ip + inputSize;
const BYTE* const mflimitPlusOne = iend - MFLIMIT + 1;
@@ -682,7 +885,7 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
/* the dictCtx currentOffset is indexed on the start of the dictionary,
* while a dictionary in the current context precedes the currentOffset */
- const BYTE* dictBase = (dictDirective == usingDictCtx) ?
+ const BYTE* dictBase = !dictionary ? NULL : (dictDirective == usingDictCtx) ?
dictionary + dictSize - dictCtx->currentOffset :
dictionary + dictSize - startIndex;
@@ -692,11 +895,12 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
U32 offset = 0;
U32 forwardH;
- DEBUGLOG(5, "LZ4_compress_generic: srcSize=%i, tableType=%u", inputSize, tableType);
- /* Init conditions */
- if (outputLimited == fillOutput && maxOutputSize < 1) return 0; /* Impossible to store anything */
- if ((U32)inputSize > (U32)LZ4_MAX_INPUT_SIZE) return 0; /* Unsupported inputSize, too large (or negative) */
- if ((tableType == byU16) && (inputSize>=LZ4_64Klimit)) return 0; /* Size too large (not within 64K limit) */
+ DEBUGLOG(5, "LZ4_compress_generic_validated: srcSize=%i, tableType=%u", inputSize, tableType);
+ assert(ip != NULL);
+ /* If init conditions are not met, we don't have to mark stream
+ * as having dirty context, since no action was taken yet */
+ if (outputDirective == fillOutput && maxOutputSize < 1) { return 0; } /* Impossible to store anything */
+ if ((tableType == byU16) && (inputSize>=LZ4_64Klimit)) { return 0; } /* Size too large (not within 64K limit) */
if (tableType==byPtr) assert(dictDirective==noDict); /* only supported use case with byPtr */
assert(acceleration >= 1);
@@ -712,7 +916,7 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
cctx->dictSize += (U32)inputSize;
}
cctx->currentOffset += (U32)inputSize;
- cctx->tableType = (U16)tableType;
+ cctx->tableType = (U32)tableType;
if (inputSize<LZ4_minLength) goto _last_literals; /* Input too small, no compression (all literals) */
@@ -724,12 +928,13 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
for ( ; ; ) {
const BYTE* match;
BYTE* token;
+ const BYTE* filledIp;
/* Find a match */
if (tableType == byPtr) {
const BYTE* forwardIp = ip;
- unsigned step = 1;
- unsigned searchMatchNb = acceleration << LZ4_skipTrigger;
+ int step = 1;
+ int searchMatchNb = acceleration << LZ4_skipTrigger;
do {
U32 const h = forwardH;
ip = forwardIp;
@@ -743,14 +948,14 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
forwardH = LZ4_hashPosition(forwardIp, tableType);
LZ4_putPositionOnHash(ip, h, cctx->hashTable, tableType, base);
- } while ( (match+MAX_DISTANCE < ip)
+ } while ( (match+LZ4_DISTANCE_MAX < ip)
|| (LZ4_read32(match) != LZ4_read32(ip)) );
} else { /* byU32, byU16 */
const BYTE* forwardIp = ip;
- unsigned step = 1;
- unsigned searchMatchNb = acceleration << LZ4_skipTrigger;
+ int step = 1;
+ int searchMatchNb = acceleration << LZ4_skipTrigger;
do {
U32 const h = forwardH;
U32 const current = (U32)(forwardIp - base);
@@ -792,10 +997,14 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
forwardH = LZ4_hashPosition(forwardIp, tableType);
LZ4_putIndexOnHash(current, h, cctx->hashTable, tableType);
- if ((dictIssue == dictSmall) && (matchIndex < prefixIdxLimit)) continue; /* match outside of valid area */
+ DEBUGLOG(7, "candidate at pos=%u (offset=%u \n", matchIndex, current - matchIndex);
+ if ((dictIssue == dictSmall) && (matchIndex < prefixIdxLimit)) { continue; } /* match outside of valid area */
assert(matchIndex < current);
- if ((tableType != byU16) && (matchIndex+MAX_DISTANCE < current)) continue; /* too far */
- if (tableType == byU16) assert((current - matchIndex) <= MAX_DISTANCE); /* too_far presumed impossible with byU16 */
+ if ( ((tableType != byU16) || (LZ4_DISTANCE_MAX < LZ4_DISTANCE_ABSOLUTE_MAX))
+ && (matchIndex+LZ4_DISTANCE_MAX < current)) {
+ continue;
+ } /* too far */
+ assert((current - matchIndex) <= LZ4_DISTANCE_MAX); /* match now expected within distance */
if (LZ4_read32(match) == LZ4_read32(ip)) {
if (maybe_extMem) offset = current - matchIndex;
@@ -806,21 +1015,23 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
}
/* Catch up */
+ filledIp = ip;
while (((ip>anchor) & (match > lowLimit)) && (unlikely(ip[-1]==match[-1]))) { ip--; match--; }
/* Encode Literals */
{ unsigned const litLength = (unsigned)(ip - anchor);
token = op++;
- if ((outputLimited == limitedOutput) && /* Check output buffer overflow */
- (unlikely(op + litLength + (2 + 1 + LASTLITERALS) + (litLength/255) > olimit)))
- return 0;
- if ((outputLimited == fillOutput) &&
+ if ((outputDirective == limitedOutput) && /* Check output buffer overflow */
+ (unlikely(op + litLength + (2 + 1 + LASTLITERALS) + (litLength/255) > olimit)) ) {
+ return 0; /* cannot compress within `dst` budget. Stored indexes in hash table are nonetheless fine */
+ }
+ if ((outputDirective == fillOutput) &&
(unlikely(op + (litLength+240)/255 /* litlen */ + litLength /* literals */ + 2 /* offset */ + 1 /* token */ + MFLIMIT - MINMATCH /* min last literals so last match is <= end - MFLIMIT */ > olimit))) {
op--;
goto _last_literals;
}
if (litLength >= RUN_MASK) {
- int len = (int)litLength-RUN_MASK;
+ int len = (int)(litLength - RUN_MASK);
*token = (RUN_MASK<<ML_BITS);
for(; len >= 255 ; len-=255) *op++ = 255;
*op++ = (BYTE)len;
@@ -828,7 +1039,7 @@ LZ4_FORCE_INLINE int LZ4_compress_generic(
else *token = (BYTE)(litLength<<ML_BITS);
/* Copy Literals */
- LZ4_wildCopy(op, anchor, op+litLength);
+ LZ4_wildCopy8(op, anchor, op+litLength);
op+=litLength;
DEBUGLOG(6, "seq.start:%i, literals=%u, match.start:%i",
(int)(anchor-(const BYTE*)source), litLength, (int)(ip-(const BYTE*)source));
@@ -843,7 +1054,7 @@ _next_match:
* - token and *token : position to write 4-bits for match length; higher 4-bits for literal length supposed already written
*/
- if ((outputLimited == fillOutput) &&
+ if ((outputDirective == fillOutput) &&
(op + 2 /* offset */ + 1 /* token */ + MFLIMIT - MINMATCH /* min last literals so last match is <= end - MFLIMIT */ > olimit)) {
/* the match was too close to the end, rewind and go to last literals */
op = token;
@@ -853,11 +1064,11 @@ _next_match:
/* Encode Offset */
if (maybe_extMem) { /* static test */
DEBUGLOG(6, " with offset=%u (ext if > %i)", offset, (int)(ip - (const BYTE*)source));
- assert(offset <= MAX_DISTANCE && offset > 0);
+ assert(offset <= LZ4_DISTANCE_MAX && offset > 0);
LZ4_writeLE16(op, (U16)offset); op+=2;
} else {
DEBUGLOG(6, " with offset=%u (same segment)", (U32)(ip - match));
- assert(ip-match <= MAX_DISTANCE);
+ assert(ip-match <= LZ4_DISTANCE_MAX);
LZ4_writeLE16(op, (U16)(ip - match)); op+=2;
}
@@ -870,7 +1081,7 @@ _next_match:
assert(dictEnd > match);
if (limit > matchlimit) limit = matchlimit;
matchCode = LZ4_count(ip+MINMATCH, match+MINMATCH, limit);
- ip += MINMATCH + matchCode;
+ ip += (size_t)matchCode + MINMATCH;
if (ip==limit) {
unsigned const more = LZ4_count(limit, (const BYTE*)source, matchlimit);
matchCode += more;
@@ -879,19 +1090,34 @@ _next_match:
DEBUGLOG(6, " with matchLength=%u starting in extDict", matchCode+MINMATCH);
} else {
matchCode = LZ4_count(ip+MINMATCH, match+MINMATCH, matchlimit);
- ip += MINMATCH + matchCode;
+ ip += (size_t)matchCode + MINMATCH;
DEBUGLOG(6, " with matchLength=%u", matchCode+MINMATCH);
}
- if ((outputLimited) && /* Check output buffer overflow */
- (unlikely(op + (1 + LASTLITERALS) + (matchCode>>8) > olimit)) ) {
- if (outputLimited == limitedOutput)
- return 0;
- if (outputLimited == fillOutput) {
+ if ((outputDirective) && /* Check output buffer overflow */
+ (unlikely(op + (1 + LASTLITERALS) + (matchCode+240)/255 > olimit)) ) {
+ if (outputDirective == fillOutput) {
/* Match description too long : reduce it */
- U32 newMatchCode = 15 /* in token */ - 1 /* to avoid needing a zero byte */ + ((U32)(olimit - op) - 2 - 1 - LASTLITERALS) * 255;
+ U32 newMatchCode = 15 /* in token */ - 1 /* to avoid needing a zero byte */ + ((U32)(olimit - op) - 1 - LASTLITERALS) * 255;
ip -= matchCode - newMatchCode;
+ assert(newMatchCode < matchCode);
matchCode = newMatchCode;
+ if (unlikely(ip <= filledIp)) {
+ /* We have already filled up to filledIp so if ip ends up less than filledIp
+ * we have positions in the hash table beyond the current position. This is
+ * a problem if we reuse the hash table. So we have to remove these positions
+ * from the hash table.
+ */
+ const BYTE* ptr;
+ DEBUGLOG(5, "Clearing %u positions", (U32)(filledIp - ip));
+ for (ptr = ip; ptr <= filledIp; ++ptr) {
+ U32 const h = LZ4_hashPosition(ptr, tableType);
+ LZ4_clearHash(h, cctx->hashTable, tableType);
+ }
+ }
+ } else {
+ assert(outputDirective == limitedOutput);
+ return 0; /* cannot compress within `dst` budget. Stored indexes in hash table are nonetheless fine */
}
}
if (matchCode >= ML_MASK) {
@@ -908,6 +1134,8 @@ _next_match:
} else
*token += (BYTE)(matchCode);
}
+ /* Ensure we have enough space for the last literals. */
+ assert(!(outputDirective == fillOutput && op + 1 + LASTLITERALS > olimit));
anchor = ip;
@@ -922,7 +1150,7 @@ _next_match:
match = LZ4_getPosition(ip, cctx->hashTable, tableType, base);
LZ4_putPosition(ip, cctx->hashTable, tableType, base);
- if ( (match+MAX_DISTANCE >= ip)
+ if ( (match+LZ4_DISTANCE_MAX >= ip)
&& (LZ4_read32(match) == LZ4_read32(ip)) )
{ token=op++; *token=0; goto _next_match; }
@@ -957,7 +1185,7 @@ _next_match:
LZ4_putIndexOnHash(current, h, cctx->hashTable, tableType);
assert(matchIndex < current);
if ( ((dictIssue==dictSmall) ? (matchIndex >= prefixIdxLimit) : 1)
- && ((tableType==byU16) ? 1 : (matchIndex+MAX_DISTANCE >= current))
+ && (((tableType==byU16) && (LZ4_DISTANCE_MAX == LZ4_DISTANCE_ABSOLUTE_MAX)) ? 1 : (matchIndex+LZ4_DISTANCE_MAX >= current))
&& (LZ4_read32(match) == LZ4_read32(ip)) ) {
token=op++;
*token=0;
@@ -976,16 +1204,19 @@ _next_match:
_last_literals:
/* Encode Last Literals */
{ size_t lastRun = (size_t)(iend - anchor);
- if ( (outputLimited) && /* Check output buffer overflow */
+ if ( (outputDirective) && /* Check output buffer overflow */
(op + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > olimit)) {
- if (outputLimited == fillOutput) {
+ if (outputDirective == fillOutput) {
/* adapt lastRun to fill 'dst' */
- lastRun = (olimit-op) - 1;
- lastRun -= (lastRun+240)/255;
+ assert(olimit >= op);
+ lastRun = (size_t)(olimit-op) - 1/*token*/;
+ lastRun -= (lastRun + 256 - RUN_MASK) / 256; /*additional length tokens*/
+ } else {
+ assert(outputDirective == limitedOutput);
+ return 0; /* cannot compress within `dst` budget. Stored indexes in hash table are nonetheless fine */
}
- if (outputLimited == limitedOutput)
- return 0;
}
+ DEBUGLOG(6, "Final literal run : %i literals", (int)lastRun);
if (lastRun >= RUN_MASK) {
size_t accumulator = lastRun - RUN_MASK;
*op++ = RUN_MASK << ML_BITS;
@@ -994,36 +1225,80 @@ _last_literals:
} else {
*op++ = (BYTE)(lastRun<<ML_BITS);
}
- memcpy(op, anchor, lastRun);
+ LZ4_memcpy(op, anchor, lastRun);
ip = anchor + lastRun;
op += lastRun;
}
- if (outputLimited == fillOutput) {
+ if (outputDirective == fillOutput) {
*inputConsumed = (int) (((const char*)ip)-source);
}
- DEBUGLOG(5, "LZ4_compress_generic: compressed %i bytes into %i bytes", inputSize, (int)(((char*)op) - dest));
- return (int)(((char*)op) - dest);
+ result = (int)(((char*)op) - dest);
+ assert(result > 0);
+ DEBUGLOG(5, "LZ4_compress_generic: compressed %i bytes into %i bytes", inputSize, result);
+ return result;
+}
+
+/** LZ4_compress_generic() :
+ * inlined, to ensure branches are decided at compilation time;
+ * takes care of src == (NULL, 0)
+ * and forward the rest to LZ4_compress_generic_validated */
+LZ4_FORCE_INLINE int LZ4_compress_generic(
+ LZ4_stream_t_internal* const cctx,
+ const char* const src,
+ char* const dst,
+ const int srcSize,
+ int *inputConsumed, /* only written when outputDirective == fillOutput */
+ const int dstCapacity,
+ const limitedOutput_directive outputDirective,
+ const tableType_t tableType,
+ const dict_directive dictDirective,
+ const dictIssue_directive dictIssue,
+ const int acceleration)
+{
+ DEBUGLOG(5, "LZ4_compress_generic: srcSize=%i, dstCapacity=%i",
+ srcSize, dstCapacity);
+
+ if ((U32)srcSize > (U32)LZ4_MAX_INPUT_SIZE) { return 0; } /* Unsupported srcSize, too large (or negative) */
+ if (srcSize == 0) { /* src == NULL supported if srcSize == 0 */
+ if (outputDirective != notLimited && dstCapacity <= 0) return 0; /* no output, can't write anything */
+ DEBUGLOG(5, "Generating an empty block");
+ assert(outputDirective == notLimited || dstCapacity >= 1);
+ assert(dst != NULL);
+ dst[0] = 0;
+ if (outputDirective == fillOutput) {
+ assert (inputConsumed != NULL);
+ *inputConsumed = 0;
+ }
+ return 1;
+ }
+ assert(src != NULL);
+
+ return LZ4_compress_generic_validated(cctx, src, dst, srcSize,
+ inputConsumed, /* only written into if outputDirective == fillOutput */
+ dstCapacity, outputDirective,
+ tableType, dictDirective, dictIssue, acceleration);
}
int LZ4_compress_fast_extState(void* state, const char* source, char* dest, int inputSize, int maxOutputSize, int acceleration)
{
- LZ4_stream_t_internal* ctx = &((LZ4_stream_t*)state)->internal_donotuse;
- if (acceleration < 1) acceleration = ACCELERATION_DEFAULT;
- LZ4_resetStream((LZ4_stream_t*)state);
+ LZ4_stream_t_internal* const ctx = & LZ4_initStream(state, sizeof(LZ4_stream_t)) -> internal_donotuse;
+ assert(ctx != NULL);
+ if (acceleration < 1) acceleration = LZ4_ACCELERATION_DEFAULT;
+ if (acceleration > LZ4_ACCELERATION_MAX) acceleration = LZ4_ACCELERATION_MAX;
if (maxOutputSize >= LZ4_compressBound(inputSize)) {
if (inputSize < LZ4_64Klimit) {
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, 0, notLimited, byU16, noDict, noDictIssue, acceleration);
} else {
- const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)source > MAX_DISTANCE)) ? byPtr : byU32;
+ const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)source > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, 0, notLimited, tableType, noDict, noDictIssue, acceleration);
}
} else {
- if (inputSize < LZ4_64Klimit) {;
+ if (inputSize < LZ4_64Klimit) {
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, byU16, noDict, noDictIssue, acceleration);
} else {
- const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)source > MAX_DISTANCE)) ? byPtr : byU32;
+ const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)source > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, noDict, noDictIssue, acceleration);
}
}
@@ -1041,7 +1316,8 @@ int LZ4_compress_fast_extState(void* state, const char* source, char* dest, int
int LZ4_compress_fast_extState_fastReset(void* state, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration)
{
LZ4_stream_t_internal* ctx = &((LZ4_stream_t*)state)->internal_donotuse;
- if (acceleration < 1) acceleration = ACCELERATION_DEFAULT;
+ if (acceleration < 1) acceleration = LZ4_ACCELERATION_DEFAULT;
+ if (acceleration > LZ4_ACCELERATION_MAX) acceleration = LZ4_ACCELERATION_MAX;
if (dstCapacity >= LZ4_compressBound(srcSize)) {
if (srcSize < LZ4_64Klimit) {
@@ -1053,7 +1329,7 @@ int LZ4_compress_fast_extState_fastReset(void* state, const char* src, char* dst
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, 0, notLimited, tableType, noDict, noDictIssue, acceleration);
}
} else {
- const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)src > MAX_DISTANCE)) ? byPtr : byU32;
+ const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)src > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
LZ4_prepareTable(ctx, srcSize, tableType);
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, 0, notLimited, tableType, noDict, noDictIssue, acceleration);
}
@@ -1067,7 +1343,7 @@ int LZ4_compress_fast_extState_fastReset(void* state, const char* src, char* dst
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, dstCapacity, limitedOutput, tableType, noDict, noDictIssue, acceleration);
}
} else {
- const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)src > MAX_DISTANCE)) ? byPtr : byU32;
+ const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)src > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
LZ4_prepareTable(ctx, srcSize, tableType);
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, dstCapacity, limitedOutput, tableType, noDict, noDictIssue, acceleration);
}
@@ -1094,23 +1370,9 @@ int LZ4_compress_fast(const char* source, char* dest, int inputSize, int maxOutp
}
-int LZ4_compress_default(const char* source, char* dest, int inputSize, int maxOutputSize)
-{
- return LZ4_compress_fast(source, dest, inputSize, maxOutputSize, 1);
-}
-
-
-/* hidden debug function */
-/* strangely enough, gcc generates faster code when this function is uncommented, even if unused */
-int LZ4_compress_fast_force(const char* source, char* dest, int inputSize, int maxOutputSize, int acceleration)
+int LZ4_compress_default(const char* src, char* dst, int srcSize, int maxOutputSize)
{
- LZ4_stream_t ctx;
- LZ4_resetStream(&ctx);
-
- if (inputSize < LZ4_64Klimit)
- return LZ4_compress_generic(&ctx.internal_donotuse, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, byU16, noDict, noDictIssue, acceleration);
- else
- return LZ4_compress_generic(&ctx.internal_donotuse, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, sizeof(void*)==8 ? byU32 : byPtr, noDict, noDictIssue, acceleration);
+ return LZ4_compress_fast(src, dst, srcSize, maxOutputSize, 1);
}
@@ -1119,7 +1381,8 @@ int LZ4_compress_fast_force(const char* source, char* dest, int inputSize, int m
* _continue() call without resetting it. */
static int LZ4_compress_destSize_extState (LZ4_stream_t* state, const char* src, char* dst, int* srcSizePtr, int targetDstSize)
{
- LZ4_resetStream(state);
+ void* const s = LZ4_initStream(state, sizeof (*state));
+ assert(s != NULL); (void)s;
if (targetDstSize >= LZ4_compressBound(*srcSizePtr)) { /* compression success is guaranteed */
return LZ4_compress_fast_extState(state, src, dst, *srcSizePtr, targetDstSize, 1);
@@ -1127,8 +1390,8 @@ static int LZ4_compress_destSize_extState (LZ4_stream_t* state, const char* src,
if (*srcSizePtr < LZ4_64Klimit) {
return LZ4_compress_generic(&state->internal_donotuse, src, dst, *srcSizePtr, srcSizePtr, targetDstSize, fillOutput, byU16, noDict, noDictIssue, 1);
} else {
- tableType_t const tableType = ((sizeof(void*)==4) && ((uptrval)src > MAX_DISTANCE)) ? byPtr : byU32;
- return LZ4_compress_generic(&state->internal_donotuse, src, dst, *srcSizePtr, srcSizePtr, targetDstSize, fillOutput, tableType, noDict, noDictIssue, 1);
+ tableType_t const addrMode = ((sizeof(void*)==4) && ((uptrval)src > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
+ return LZ4_compress_generic(&state->internal_donotuse, src, dst, *srcSizePtr, srcSizePtr, targetDstSize, fillOutput, addrMode, noDict, noDictIssue, 1);
} }
}
@@ -1159,18 +1422,40 @@ int LZ4_compress_destSize(const char* src, char* dst, int* srcSizePtr, int targe
LZ4_stream_t* LZ4_createStream(void)
{
- LZ4_stream_t* lz4s = (LZ4_stream_t*)ALLOC(sizeof(LZ4_stream_t));
+ LZ4_stream_t* const lz4s = (LZ4_stream_t*)ALLOC(sizeof(LZ4_stream_t));
LZ4_STATIC_ASSERT(LZ4_STREAMSIZE >= sizeof(LZ4_stream_t_internal)); /* A compilation error here means LZ4_STREAMSIZE is not large enough */
DEBUGLOG(4, "LZ4_createStream %p", lz4s);
if (lz4s == NULL) return NULL;
- LZ4_resetStream(lz4s);
+ LZ4_initStream(lz4s, sizeof(*lz4s));
return lz4s;
}
+static size_t LZ4_stream_t_alignment(void)
+{
+#if LZ4_ALIGN_TEST
+ typedef struct { char c; LZ4_stream_t t; } t_a;
+ return sizeof(t_a) - sizeof(LZ4_stream_t);
+#else
+ return 1; /* effectively disabled */
+#endif
+}
+
+LZ4_stream_t* LZ4_initStream (void* buffer, size_t size)
+{
+ DEBUGLOG(5, "LZ4_initStream");
+ if (buffer == NULL) { return NULL; }
+ if (size < sizeof(LZ4_stream_t)) { return NULL; }
+ if (!LZ4_isAligned(buffer, LZ4_stream_t_alignment())) return NULL;
+ MEM_INIT(buffer, 0, sizeof(LZ4_stream_t_internal));
+ return (LZ4_stream_t*)buffer;
+}
+
+/* resetStream is now deprecated,
+ * prefer initStream() which is more general */
void LZ4_resetStream (LZ4_stream_t* LZ4_stream)
{
DEBUGLOG(5, "LZ4_resetStream (ctx:%p)", LZ4_stream);
- MEM_INIT(LZ4_stream, 0, sizeof(LZ4_stream_t));
+ MEM_INIT(LZ4_stream, 0, sizeof(LZ4_stream_t_internal));
}
void LZ4_resetStream_fast(LZ4_stream_t* ctx) {
@@ -1209,46 +1494,58 @@ int LZ4_loadDict (LZ4_stream_t* LZ4_dict, const char* dictionary, int dictSize)
* there are only valid offsets in the window, which allows an optimization
* in LZ4_compress_fast_continue() where it uses noDictIssue even when the
* dictionary isn't a full 64k. */
-
- if ((dictEnd - p) > 64 KB) p = dictEnd - 64 KB;
- base = dictEnd - 64 KB - dict->currentOffset;
- dict->dictionary = p;
- dict->dictSize = (U32)(dictEnd - p);
dict->currentOffset += 64 KB;
- dict->tableType = tableType;
if (dictSize < (int)HASH_UNIT) {
return 0;
}
+ if ((dictEnd - p) > 64 KB) p = dictEnd - 64 KB;
+ base = dictEnd - dict->currentOffset;
+ dict->dictionary = p;
+ dict->dictSize = (U32)(dictEnd - p);
+ dict->tableType = (U32)tableType;
+
while (p <= dictEnd-HASH_UNIT) {
LZ4_putPosition(p, dict->hashTable, tableType, base);
p+=3;
}
- return dict->dictSize;
+ return (int)dict->dictSize;
}
-void LZ4_attach_dictionary(LZ4_stream_t *working_stream, const LZ4_stream_t *dictionary_stream) {
- if (dictionary_stream != NULL) {
+void LZ4_attach_dictionary(LZ4_stream_t* workingStream, const LZ4_stream_t* dictionaryStream) {
+ const LZ4_stream_t_internal* dictCtx = dictionaryStream == NULL ? NULL :
+ &(dictionaryStream->internal_donotuse);
+
+ DEBUGLOG(4, "LZ4_attach_dictionary (%p, %p, size %u)",
+ workingStream, dictionaryStream,
+ dictCtx != NULL ? dictCtx->dictSize : 0);
+
+ if (dictCtx != NULL) {
/* If the current offset is zero, we will never look in the
* external dictionary context, since there is no value a table
* entry can take that indicate a miss. In that case, we need
* to bump the offset to something non-zero.
*/
- if (working_stream->internal_donotuse.currentOffset == 0) {
- working_stream->internal_donotuse.currentOffset = 64 KB;
+ if (workingStream->internal_donotuse.currentOffset == 0) {
+ workingStream->internal_donotuse.currentOffset = 64 KB;
+ }
+
+ /* Don't actually attach an empty dictionary.
+ */
+ if (dictCtx->dictSize == 0) {
+ dictCtx = NULL;
}
- working_stream->internal_donotuse.dictCtx = &(dictionary_stream->internal_donotuse);
- } else {
- working_stream->internal_donotuse.dictCtx = NULL;
}
+ workingStream->internal_donotuse.dictCtx = dictCtx;
}
static void LZ4_renormDictT(LZ4_stream_t_internal* LZ4_dict, int nextSize)
{
- if (LZ4_dict->currentOffset + nextSize > 0x80000000) { /* potential ptrdiff_t overflow (32-bits mode) */
+ assert(nextSize >= 0);
+ if (LZ4_dict->currentOffset + (unsigned)nextSize > 0x80000000) { /* potential ptrdiff_t overflow (32-bits mode) */
/* rescale hash table */
U32 const delta = LZ4_dict->currentOffset - 64 KB;
const BYTE* dictEnd = LZ4_dict->dictionary + LZ4_dict->dictSize;
@@ -1265,7 +1562,10 @@ static void LZ4_renormDictT(LZ4_stream_t_internal* LZ4_dict, int nextSize)
}
-int LZ4_compress_fast_continue (LZ4_stream_t* LZ4_stream, const char* source, char* dest, int inputSize, int maxOutputSize, int acceleration)
+int LZ4_compress_fast_continue (LZ4_stream_t* LZ4_stream,
+ const char* source, char* dest,
+ int inputSize, int maxOutputSize,
+ int acceleration)
{
const tableType_t tableType = byU32;
LZ4_stream_t_internal* streamPtr = &LZ4_stream->internal_donotuse;
@@ -1273,12 +1573,12 @@ int LZ4_compress_fast_continue (LZ4_stream_t* LZ4_stream, const char* source, ch
DEBUGLOG(5, "LZ4_compress_fast_continue (inputSize=%i)", inputSize);
- if (streamPtr->initCheck) return 0; /* Uninitialized structure detected */
LZ4_renormDictT(streamPtr, inputSize); /* avoid index overflow */
- if (acceleration < 1) acceleration = ACCELERATION_DEFAULT;
+ if (acceleration < 1) acceleration = LZ4_ACCELERATION_DEFAULT;
+ if (acceleration > LZ4_ACCELERATION_MAX) acceleration = LZ4_ACCELERATION_MAX;
/* invalidate tiny dictionaries */
- if ( (streamPtr->dictSize-1 < 4) /* intentional underflow */
+ if ( (streamPtr->dictSize-1 < 4-1) /* intentional underflow */
&& (dictEnd != (const BYTE*)source) ) {
DEBUGLOG(5, "LZ4_compress_fast_continue: dictSize(%u) at addr:%p is too small", streamPtr->dictSize, streamPtr->dictionary);
streamPtr->dictSize = 0;
@@ -1318,7 +1618,7 @@ int LZ4_compress_fast_continue (LZ4_stream_t* LZ4_stream, const char* source, ch
* cost to copy the dictionary's tables into the active context,
* so that the compression loop is only looking into one table.
*/
- memcpy(streamPtr, streamPtr->dictCtx, sizeof(LZ4_stream_t));
+ LZ4_memcpy(streamPtr, streamPtr->dictCtx, sizeof(*streamPtr));
result = LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, usingExtDict, noDictIssue, acceleration);
} else {
result = LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, usingDictCtx, noDictIssue, acceleration);
@@ -1370,10 +1670,12 @@ int LZ4_saveDict (LZ4_stream_t* LZ4_dict, char* safeBuffer, int dictSize)
LZ4_stream_t_internal* const dict = &LZ4_dict->internal_donotuse;
const BYTE* const previousDictEnd = dict->dictionary + dict->dictSize;
- if ((U32)dictSize > 64 KB) dictSize = 64 KB; /* useless to define a dictionary > 64 KB */
- if ((U32)dictSize > dict->dictSize) dictSize = dict->dictSize;
+ if ((U32)dictSize > 64 KB) { dictSize = 64 KB; } /* useless to define a dictionary > 64 KB */
+ if ((U32)dictSize > dict->dictSize) { dictSize = (int)dict->dictSize; }
- memmove(safeBuffer, previousDictEnd - dictSize, dictSize);
+ if (safeBuffer == NULL) assert(dictSize == 0);
+ if (dictSize > 0)
+ memmove(safeBuffer, previousDictEnd - dictSize, dictSize);
dict->dictionary = (const BYTE*)safeBuffer;
dict->dictSize = (U32)dictSize;
@@ -1393,6 +1695,39 @@ typedef enum { decode_full_block = 0, partial_decode = 1 } earlyEnd_directive;
#undef MIN
#define MIN(a,b) ( (a) < (b) ? (a) : (b) )
+/* Read the variable-length literal or match length.
+ *
+ * ip - pointer to use as input.
+ * lencheck - end ip. Return an error if ip advances >= lencheck.
+ * loop_check - check ip >= lencheck in body of loop. Returns loop_error if so.
+ * initial_check - check ip >= lencheck before start of loop. Returns initial_error if so.
+ * error (output) - error code. Should be set to 0 before call.
+ */
+typedef enum { loop_error = -2, initial_error = -1, ok = 0 } variable_length_error;
+LZ4_FORCE_INLINE unsigned
+read_variable_length(const BYTE**ip, const BYTE* lencheck,
+ int loop_check, int initial_check,
+ variable_length_error* error)
+{
+ U32 length = 0;
+ U32 s;
+ if (initial_check && unlikely((*ip) >= lencheck)) { /* overflow detection */
+ *error = initial_error;
+ return length;
+ }
+ do {
+ s = **ip;
+ (*ip)++;
+ length += s;
+ if (loop_check && unlikely((*ip) >= lencheck)) { /* overflow detection */
+ *error = loop_error;
+ return length;
+ }
+ } while (s==255);
+
+ return length;
+}
+
/*! LZ4_decompress_generic() :
* This generic decompression function covers all use cases.
* It shall be instantiated several times, using different sets of directives.
@@ -1414,240 +1749,425 @@ LZ4_decompress_generic(
const size_t dictSize /* note : = 0 if noDict */
)
{
- const BYTE* ip = (const BYTE*) src;
- const BYTE* const iend = ip + srcSize;
+ if (src == NULL) { return -1; }
- BYTE* op = (BYTE*) dst;
- BYTE* const oend = op + outputSize;
- BYTE* cpy;
+ { const BYTE* ip = (const BYTE*) src;
+ const BYTE* const iend = ip + srcSize;
- const BYTE* const dictEnd = (const BYTE*)dictStart + dictSize;
- const unsigned inc32table[8] = {0, 1, 2, 1, 0, 4, 4, 4};
- const int dec64table[8] = {0, 0, 0, -1, -4, 1, 2, 3};
+ BYTE* op = (BYTE*) dst;
+ BYTE* const oend = op + outputSize;
+ BYTE* cpy;
- const int safeDecode = (endOnInput==endOnInputSize);
- const int checkOffset = ((safeDecode) && (dictSize < (int)(64 KB)));
+ const BYTE* const dictEnd = (dictStart == NULL) ? NULL : dictStart + dictSize;
- /* Set up the "end" pointers for the shortcut. */
- const BYTE* const shortiend = iend - (endOnInput ? 14 : 8) /*maxLL*/ - 2 /*offset*/;
- const BYTE* const shortoend = oend - (endOnInput ? 14 : 8) /*maxLL*/ - 18 /*maxML*/;
+ const int safeDecode = (endOnInput==endOnInputSize);
+ const int checkOffset = ((safeDecode) && (dictSize < (int)(64 KB)));
- DEBUGLOG(5, "LZ4_decompress_generic (srcSize:%i, dstSize:%i)", srcSize, outputSize);
- /* Special cases */
- assert(lowPrefix <= op);
- assert(src != NULL);
- if ((endOnInput) && (unlikely(outputSize==0))) return ((srcSize==1) && (*ip==0)) ? 0 : -1; /* Empty output buffer */
- if ((!endOnInput) && (unlikely(outputSize==0))) return (*ip==0 ? 1 : -1);
- if ((endOnInput) && unlikely(srcSize==0)) return -1;
+ /* Set up the "end" pointers for the shortcut. */
+ const BYTE* const shortiend = iend - (endOnInput ? 14 : 8) /*maxLL*/ - 2 /*offset*/;
+ const BYTE* const shortoend = oend - (endOnInput ? 14 : 8) /*maxLL*/ - 18 /*maxML*/;
- /* Main Loop : decode sequences */
- while (1) {
const BYTE* match;
size_t offset;
+ unsigned token;
+ size_t length;
- unsigned const token = *ip++;
- size_t length = token >> ML_BITS; /* literal length */
- assert(!endOnInput || ip <= iend); /* ip < iend before the increment */
+ DEBUGLOG(5, "LZ4_decompress_generic (srcSize:%i, dstSize:%i)", srcSize, outputSize);
- /* A two-stage shortcut for the most common case:
- * 1) If the literal length is 0..14, and there is enough space,
- * enter the shortcut and copy 16 bytes on behalf of the literals
- * (in the fast mode, only 8 bytes can be safely copied this way).
- * 2) Further if the match length is 4..18, copy 18 bytes in a similar
- * manner; but we ensure that there's enough space in the output for
- * those 18 bytes earlier, upon entering the shortcut (in other words,
- * there is a combined check for both stages).
- */
- if ( (endOnInput ? length != RUN_MASK : length <= 8)
- /* strictly "less than" on input, to re-enter the loop with at least one byte */
- && likely((endOnInput ? ip < shortiend : 1) & (op <= shortoend)) ) {
- /* Copy the literals */
- memcpy(op, ip, endOnInput ? 16 : 8);
- op += length; ip += length;
-
- /* The second stage: prepare for match copying, decode full info.
- * If it doesn't work out, the info won't be wasted. */
- length = token & ML_MASK; /* match length */
- offset = LZ4_readLE16(ip); ip += 2;
- match = op - offset;
- assert(match <= op); /* check overflow */
-
- /* Do not deal with overlapping matches. */
- if ( (length != ML_MASK)
- && (offset >= 8)
- && (dict==withPrefix64k || match >= lowPrefix) ) {
- /* Copy the match. */
- memcpy(op + 0, match + 0, 8);
- memcpy(op + 8, match + 8, 8);
- memcpy(op +16, match +16, 2);
- op += length + MINMATCH;
- /* Both stages worked, load the next token. */
- continue;
- }
-
- /* The second stage didn't work out, but the info is ready.
- * Propel it right to the point of match copying. */
- goto _copy_match;
+ /* Special cases */
+ assert(lowPrefix <= op);
+ if ((endOnInput) && (unlikely(outputSize==0))) {
+ /* Empty output buffer */
+ if (partialDecoding) return 0;
+ return ((srcSize==1) && (*ip==0)) ? 0 : -1;
}
-
- /* decode literal length */
- if (length == RUN_MASK) {
- unsigned s;
- if (unlikely(endOnInput ? ip >= iend-RUN_MASK : 0)) goto _output_error; /* overflow detection */
- do {
- s = *ip++;
- length += s;
- } while ( likely(endOnInput ? ip<iend-RUN_MASK : 1) & (s==255) );
- if ((safeDecode) && unlikely((uptrval)(op)+length<(uptrval)(op))) goto _output_error; /* overflow detection */
- if ((safeDecode) && unlikely((uptrval)(ip)+length<(uptrval)(ip))) goto _output_error; /* overflow detection */
+ if ((!endOnInput) && (unlikely(outputSize==0))) { return (*ip==0 ? 1 : -1); }
+ if ((endOnInput) && unlikely(srcSize==0)) { return -1; }
+
+ /* Currently the fast loop shows a regression on qualcomm arm chips. */
+#if LZ4_FAST_DEC_LOOP
+ if ((oend - op) < FASTLOOP_SAFE_DISTANCE) {
+ DEBUGLOG(6, "skip fast decode loop");
+ goto safe_decode;
}
- /* copy literals */
- cpy = op+length;
- LZ4_STATIC_ASSERT(MFLIMIT >= WILDCOPYLENGTH);
- if ( ((endOnInput) && ((cpy>oend-MFLIMIT) || (ip+length>iend-(2+1+LASTLITERALS))) )
- || ((!endOnInput) && (cpy>oend-WILDCOPYLENGTH)) )
- {
- if (partialDecoding) {
- if (cpy > oend) { cpy = oend; length = oend-op; } /* Partial decoding : stop in the middle of literal segment */
- if ((endOnInput) && (ip+length > iend)) goto _output_error; /* Error : read attempt beyond end of input buffer */
+ /* Fast loop : decode sequences as long as output < iend-FASTLOOP_SAFE_DISTANCE */
+ while (1) {
+ /* Main fastloop assertion: We can always wildcopy FASTLOOP_SAFE_DISTANCE */
+ assert(oend - op >= FASTLOOP_SAFE_DISTANCE);
+ if (endOnInput) { assert(ip < iend); }
+ token = *ip++;
+ length = token >> ML_BITS; /* literal length */
+
+ assert(!endOnInput || ip <= iend); /* ip < iend before the increment */
+
+ /* decode literal length */
+ if (length == RUN_MASK) {
+ variable_length_error error = ok;
+ length += read_variable_length(&ip, iend-RUN_MASK, (int)endOnInput, (int)endOnInput, &error);
+ if (error == initial_error) { goto _output_error; }
+ if ((safeDecode) && unlikely((uptrval)(op)+length<(uptrval)(op))) { goto _output_error; } /* overflow detection */
+ if ((safeDecode) && unlikely((uptrval)(ip)+length<(uptrval)(ip))) { goto _output_error; } /* overflow detection */
+
+ /* copy literals */
+ cpy = op+length;
+ LZ4_STATIC_ASSERT(MFLIMIT >= WILDCOPYLENGTH);
+ if (endOnInput) { /* LZ4_decompress_safe() */
+ if ((cpy>oend-32) || (ip+length>iend-32)) { goto safe_literal_copy; }
+ LZ4_wildCopy32(op, ip, cpy);
+ } else { /* LZ4_decompress_fast() */
+ if (cpy>oend-8) { goto safe_literal_copy; }
+ LZ4_wildCopy8(op, ip, cpy); /* LZ4_decompress_fast() cannot copy more than 8 bytes at a time :
+ * it doesn't know input length, and only relies on end-of-block properties */
+ }
+ ip += length; op = cpy;
} else {
- if ((!endOnInput) && (cpy != oend)) goto _output_error; /* Error : block decoding must stop exactly there */
- if ((endOnInput) && ((ip+length != iend) || (cpy > oend))) goto _output_error; /* Error : input must be consumed */
- }
- memcpy(op, ip, length);
- ip += length;
- op += length;
- if (!partialDecoding || (cpy == oend)) {
- /* Necessarily EOF, due to parsing restrictions */
- break;
+ cpy = op+length;
+ if (endOnInput) { /* LZ4_decompress_safe() */
+ DEBUGLOG(7, "copy %u bytes in a 16-bytes stripe", (unsigned)length);
+ /* We don't need to check oend, since we check it once for each loop below */
+ if (ip > iend-(16 + 1/*max lit + offset + nextToken*/)) { goto safe_literal_copy; }
+ /* Literals can only be 14, but hope compilers optimize if we copy by a register size */
+ LZ4_memcpy(op, ip, 16);
+ } else { /* LZ4_decompress_fast() */
+ /* LZ4_decompress_fast() cannot copy more than 8 bytes at a time :
+ * it doesn't know input length, and relies on end-of-block properties */
+ LZ4_memcpy(op, ip, 8);
+ if (length > 8) { LZ4_memcpy(op+8, ip+8, 8); }
+ }
+ ip += length; op = cpy;
}
- } else {
- LZ4_wildCopy(op, ip, cpy); /* may overwrite up to WILDCOPYLENGTH beyond cpy */
- ip += length; op = cpy;
- }
+ /* get offset */
+ offset = LZ4_readLE16(ip); ip+=2;
+ match = op - offset;
+ assert(match <= op);
+
+ /* get matchlength */
+ length = token & ML_MASK;
+
+ if (length == ML_MASK) {
+ variable_length_error error = ok;
+ if ((checkOffset) && (unlikely(match + dictSize < lowPrefix))) { goto _output_error; } /* Error : offset outside buffers */
+ length += read_variable_length(&ip, iend - LASTLITERALS + 1, (int)endOnInput, 0, &error);
+ if (error != ok) { goto _output_error; }
+ if ((safeDecode) && unlikely((uptrval)(op)+length<(uptrval)op)) { goto _output_error; } /* overflow detection */
+ length += MINMATCH;
+ if (op + length >= oend - FASTLOOP_SAFE_DISTANCE) {
+ goto safe_match_copy;
+ }
+ } else {
+ length += MINMATCH;
+ if (op + length >= oend - FASTLOOP_SAFE_DISTANCE) {
+ goto safe_match_copy;
+ }
- /* get offset */
- offset = LZ4_readLE16(ip); ip+=2;
- match = op - offset;
+ /* Fastpath check: Avoids a branch in LZ4_wildCopy32 if true */
+ if ((dict == withPrefix64k) || (match >= lowPrefix)) {
+ if (offset >= 8) {
+ assert(match >= lowPrefix);
+ assert(match <= op);
+ assert(op + 18 <= oend);
+
+ LZ4_memcpy(op, match, 8);
+ LZ4_memcpy(op+8, match+8, 8);
+ LZ4_memcpy(op+16, match+16, 2);
+ op += length;
+ continue;
+ } } }
+
+ if (checkOffset && (unlikely(match + dictSize < lowPrefix))) { goto _output_error; } /* Error : offset outside buffers */
+ /* match starting within external dictionary */
+ if ((dict==usingExtDict) && (match < lowPrefix)) {
+ if (unlikely(op+length > oend-LASTLITERALS)) {
+ if (partialDecoding) {
+ DEBUGLOG(7, "partialDecoding: dictionary match, close to dstEnd");
+ length = MIN(length, (size_t)(oend-op));
+ } else {
+ goto _output_error; /* end-of-block condition violated */
+ } }
+
+ if (length <= (size_t)(lowPrefix-match)) {
+ /* match fits entirely within external dictionary : just copy */
+ memmove(op, dictEnd - (lowPrefix-match), length);
+ op += length;
+ } else {
+ /* match stretches into both external dictionary and current block */
+ size_t const copySize = (size_t)(lowPrefix - match);
+ size_t const restSize = length - copySize;
+ LZ4_memcpy(op, dictEnd - copySize, copySize);
+ op += copySize;
+ if (restSize > (size_t)(op - lowPrefix)) { /* overlap copy */
+ BYTE* const endOfMatch = op + restSize;
+ const BYTE* copyFrom = lowPrefix;
+ while (op < endOfMatch) { *op++ = *copyFrom++; }
+ } else {
+ LZ4_memcpy(op, lowPrefix, restSize);
+ op += restSize;
+ } }
+ continue;
+ }
- /* get matchlength */
- length = token & ML_MASK;
+ /* copy match within block */
+ cpy = op + length;
-_copy_match:
- if ((checkOffset) && (unlikely(match + dictSize < lowPrefix))) goto _output_error; /* Error : offset outside buffers */
- if (!partialDecoding) {
- assert(oend > op);
- assert(oend - op >= 4);
- LZ4_write32(op, 0); /* silence an msan warning when offset==0; costs <1%; */
- } /* note : when partialDecoding, there is no guarantee that at least 4 bytes remain available in output buffer */
+ assert((op <= oend) && (oend-op >= 32));
+ if (unlikely(offset<16)) {
+ LZ4_memcpy_using_offset(op, match, cpy, offset);
+ } else {
+ LZ4_wildCopy32(op, match, cpy);
+ }
- if (length == ML_MASK) {
- unsigned s;
- do {
- s = *ip++;
- if ((endOnInput) && (ip > iend-LASTLITERALS)) goto _output_error;
- length += s;
- } while (s==255);
- if ((safeDecode) && unlikely((uptrval)(op)+length<(uptrval)op)) goto _output_error; /* overflow detection */
+ op = cpy; /* wildcopy correction */
}
- length += MINMATCH;
+ safe_decode:
+#endif
+
+ /* Main Loop : decode remaining sequences where output < FASTLOOP_SAFE_DISTANCE */
+ while (1) {
+ token = *ip++;
+ length = token >> ML_BITS; /* literal length */
+
+ assert(!endOnInput || ip <= iend); /* ip < iend before the increment */
+
+ /* A two-stage shortcut for the most common case:
+ * 1) If the literal length is 0..14, and there is enough space,
+ * enter the shortcut and copy 16 bytes on behalf of the literals
+ * (in the fast mode, only 8 bytes can be safely copied this way).
+ * 2) Further if the match length is 4..18, copy 18 bytes in a similar
+ * manner; but we ensure that there's enough space in the output for
+ * those 18 bytes earlier, upon entering the shortcut (in other words,
+ * there is a combined check for both stages).
+ */
+ if ( (endOnInput ? length != RUN_MASK : length <= 8)
+ /* strictly "less than" on input, to re-enter the loop with at least one byte */
+ && likely((endOnInput ? ip < shortiend : 1) & (op <= shortoend)) ) {
+ /* Copy the literals */
+ LZ4_memcpy(op, ip, endOnInput ? 16 : 8);
+ op += length; ip += length;
+
+ /* The second stage: prepare for match copying, decode full info.
+ * If it doesn't work out, the info won't be wasted. */
+ length = token & ML_MASK; /* match length */
+ offset = LZ4_readLE16(ip); ip += 2;
+ match = op - offset;
+ assert(match <= op); /* check overflow */
+
+ /* Do not deal with overlapping matches. */
+ if ( (length != ML_MASK)
+ && (offset >= 8)
+ && (dict==withPrefix64k || match >= lowPrefix) ) {
+ /* Copy the match. */
+ LZ4_memcpy(op + 0, match + 0, 8);
+ LZ4_memcpy(op + 8, match + 8, 8);
+ LZ4_memcpy(op +16, match +16, 2);
+ op += length + MINMATCH;
+ /* Both stages worked, load the next token. */
+ continue;
+ }
- /* match starting within external dictionary */
- if ((dict==usingExtDict) && (match < lowPrefix)) {
- if (unlikely(op+length > oend-LASTLITERALS)) {
- if (partialDecoding) length = MIN(length, (size_t)(oend-op));
- else goto _output_error; /* doesn't respect parsing restriction */
+ /* The second stage didn't work out, but the info is ready.
+ * Propel it right to the point of match copying. */
+ goto _copy_match;
}
- if (length <= (size_t)(lowPrefix-match)) {
- /* match fits entirely within external dictionary : just copy */
- memmove(op, dictEnd - (lowPrefix-match), length);
+ /* decode literal length */
+ if (length == RUN_MASK) {
+ variable_length_error error = ok;
+ length += read_variable_length(&ip, iend-RUN_MASK, (int)endOnInput, (int)endOnInput, &error);
+ if (error == initial_error) { goto _output_error; }
+ if ((safeDecode) && unlikely((uptrval)(op)+length<(uptrval)(op))) { goto _output_error; } /* overflow detection */
+ if ((safeDecode) && unlikely((uptrval)(ip)+length<(uptrval)(ip))) { goto _output_error; } /* overflow detection */
+ }
+
+ /* copy literals */
+ cpy = op+length;
+#if LZ4_FAST_DEC_LOOP
+ safe_literal_copy:
+#endif
+ LZ4_STATIC_ASSERT(MFLIMIT >= WILDCOPYLENGTH);
+ if ( ((endOnInput) && ((cpy>oend-MFLIMIT) || (ip+length>iend-(2+1+LASTLITERALS))) )
+ || ((!endOnInput) && (cpy>oend-WILDCOPYLENGTH)) )
+ {
+ /* We've either hit the input parsing restriction or the output parsing restriction.
+ * In the normal scenario, decoding a full block, it must be the last sequence,
+ * otherwise it's an error (invalid input or dimensions).
+ * In partialDecoding scenario, it's necessary to ensure there is no buffer overflow.
+ */
+ if (partialDecoding) {
+ /* Since we are partial decoding we may be in this block because of the output parsing
+ * restriction, which is not valid since the output buffer is allowed to be undersized.
+ */
+ assert(endOnInput);
+ DEBUGLOG(7, "partialDecoding: copying literals, close to input or output end")
+ DEBUGLOG(7, "partialDecoding: literal length = %u", (unsigned)length);
+ DEBUGLOG(7, "partialDecoding: remaining space in dstBuffer : %i", (int)(oend - op));
+ DEBUGLOG(7, "partialDecoding: remaining space in srcBuffer : %i", (int)(iend - ip));
+ /* Finishing in the middle of a literals segment,
+ * due to lack of input.
+ */
+ if (ip+length > iend) {
+ length = (size_t)(iend-ip);
+ cpy = op + length;
+ }
+ /* Finishing in the middle of a literals segment,
+ * due to lack of output space.
+ */
+ if (cpy > oend) {
+ cpy = oend;
+ assert(op<=oend);
+ length = (size_t)(oend-op);
+ }
+ } else {
+ /* We must be on the last sequence because of the parsing limitations so check
+ * that we exactly regenerate the original size (must be exact when !endOnInput).
+ */
+ if ((!endOnInput) && (cpy != oend)) { goto _output_error; }
+ /* We must be on the last sequence (or invalid) because of the parsing limitations
+ * so check that we exactly consume the input and don't overrun the output buffer.
+ */
+ if ((endOnInput) && ((ip+length != iend) || (cpy > oend))) {
+ DEBUGLOG(6, "should have been last run of literals")
+ DEBUGLOG(6, "ip(%p) + length(%i) = %p != iend (%p)", ip, (int)length, ip+length, iend);
+ DEBUGLOG(6, "or cpy(%p) > oend(%p)", cpy, oend);
+ goto _output_error;
+ }
+ }
+ memmove(op, ip, length); /* supports overlapping memory regions; only matters for in-place decompression scenarios */
+ ip += length;
op += length;
+ /* Necessarily EOF when !partialDecoding.
+ * When partialDecoding, it is EOF if we've either
+ * filled the output buffer or
+ * can't proceed with reading an offset for following match.
+ */
+ if (!partialDecoding || (cpy == oend) || (ip >= (iend-2))) {
+ break;
+ }
} else {
- /* match stretches into both external dictionary and current block */
- size_t const copySize = (size_t)(lowPrefix - match);
- size_t const restSize = length - copySize;
- memcpy(op, dictEnd - copySize, copySize);
- op += copySize;
- if (restSize > (size_t)(op - lowPrefix)) { /* overlap copy */
- BYTE* const endOfMatch = op + restSize;
- const BYTE* copyFrom = lowPrefix;
- while (op < endOfMatch) *op++ = *copyFrom++;
+ LZ4_wildCopy8(op, ip, cpy); /* may overwrite up to WILDCOPYLENGTH beyond cpy */
+ ip += length; op = cpy;
+ }
+
+ /* get offset */
+ offset = LZ4_readLE16(ip); ip+=2;
+ match = op - offset;
+
+ /* get matchlength */
+ length = token & ML_MASK;
+
+ _copy_match:
+ if (length == ML_MASK) {
+ variable_length_error error = ok;
+ length += read_variable_length(&ip, iend - LASTLITERALS + 1, (int)endOnInput, 0, &error);
+ if (error != ok) goto _output_error;
+ if ((safeDecode) && unlikely((uptrval)(op)+length<(uptrval)op)) goto _output_error; /* overflow detection */
+ }
+ length += MINMATCH;
+
+#if LZ4_FAST_DEC_LOOP
+ safe_match_copy:
+#endif
+ if ((checkOffset) && (unlikely(match + dictSize < lowPrefix))) goto _output_error; /* Error : offset outside buffers */
+ /* match starting within external dictionary */
+ if ((dict==usingExtDict) && (match < lowPrefix)) {
+ if (unlikely(op+length > oend-LASTLITERALS)) {
+ if (partialDecoding) length = MIN(length, (size_t)(oend-op));
+ else goto _output_error; /* doesn't respect parsing restriction */
+ }
+
+ if (length <= (size_t)(lowPrefix-match)) {
+ /* match fits entirely within external dictionary : just copy */
+ memmove(op, dictEnd - (lowPrefix-match), length);
+ op += length;
} else {
- memcpy(op, lowPrefix, restSize);
- op += restSize;
- } }
- continue;
- }
+ /* match stretches into both external dictionary and current block */
+ size_t const copySize = (size_t)(lowPrefix - match);
+ size_t const restSize = length - copySize;
+ LZ4_memcpy(op, dictEnd - copySize, copySize);
+ op += copySize;
+ if (restSize > (size_t)(op - lowPrefix)) { /* overlap copy */
+ BYTE* const endOfMatch = op + restSize;
+ const BYTE* copyFrom = lowPrefix;
+ while (op < endOfMatch) *op++ = *copyFrom++;
+ } else {
+ LZ4_memcpy(op, lowPrefix, restSize);
+ op += restSize;
+ } }
+ continue;
+ }
+ assert(match >= lowPrefix);
+
+ /* copy match within block */
+ cpy = op + length;
+
+ /* partialDecoding : may end anywhere within the block */
+ assert(op<=oend);
+ if (partialDecoding && (cpy > oend-MATCH_SAFEGUARD_DISTANCE)) {
+ size_t const mlen = MIN(length, (size_t)(oend-op));
+ const BYTE* const matchEnd = match + mlen;
+ BYTE* const copyEnd = op + mlen;
+ if (matchEnd > op) { /* overlap copy */
+ while (op < copyEnd) { *op++ = *match++; }
+ } else {
+ LZ4_memcpy(op, match, mlen);
+ }
+ op = copyEnd;
+ if (op == oend) { break; }
+ continue;
+ }
- /* copy match within block */
- cpy = op + length;
-
- /* partialDecoding : may not respect endBlock parsing restrictions */
- assert(op<=oend);
- if (partialDecoding && (cpy > oend-MATCH_SAFEGUARD_DISTANCE)) {
- size_t const mlen = MIN(length, (size_t)(oend-op));
- const BYTE* const matchEnd = match + mlen;
- BYTE* const copyEnd = op + mlen;
- if (matchEnd > op) { /* overlap copy */
- while (op < copyEnd) *op++ = *match++;
+ if (unlikely(offset<8)) {
+ LZ4_write32(op, 0); /* silence msan warning when offset==0 */
+ op[0] = match[0];
+ op[1] = match[1];
+ op[2] = match[2];
+ op[3] = match[3];
+ match += inc32table[offset];
+ LZ4_memcpy(op+4, match, 4);
+ match -= dec64table[offset];
} else {
- memcpy(op, match, mlen);
+ LZ4_memcpy(op, match, 8);
+ match += 8;
}
- op = copyEnd;
- if (op==oend) break;
- continue;
- }
-
- if (unlikely(offset<8)) {
- op[0] = match[0];
- op[1] = match[1];
- op[2] = match[2];
- op[3] = match[3];
- match += inc32table[offset];
- memcpy(op+4, match, 4);
- match -= dec64table[offset];
- } else {
- memcpy(op, match, 8);
- match += 8;
- }
- op += 8;
-
- if (unlikely(cpy > oend-MATCH_SAFEGUARD_DISTANCE)) {
- BYTE* const oCopyLimit = oend - (WILDCOPYLENGTH-1);
- if (cpy > oend-LASTLITERALS) goto _output_error; /* Error : last LASTLITERALS bytes must be literals (uncompressed) */
- if (op < oCopyLimit) {
- LZ4_wildCopy(op, match, oCopyLimit);
- match += oCopyLimit - op;
- op = oCopyLimit;
+ op += 8;
+
+ if (unlikely(cpy > oend-MATCH_SAFEGUARD_DISTANCE)) {
+ BYTE* const oCopyLimit = oend - (WILDCOPYLENGTH-1);
+ if (cpy > oend-LASTLITERALS) { goto _output_error; } /* Error : last LASTLITERALS bytes must be literals (uncompressed) */
+ if (op < oCopyLimit) {
+ LZ4_wildCopy8(op, match, oCopyLimit);
+ match += oCopyLimit - op;
+ op = oCopyLimit;
+ }
+ while (op < cpy) { *op++ = *match++; }
+ } else {
+ LZ4_memcpy(op, match, 8);
+ if (length > 16) { LZ4_wildCopy8(op+8, match+8, cpy); }
}
- while (op < cpy) *op++ = *match++;
- } else {
- memcpy(op, match, 8);
- if (length > 16) LZ4_wildCopy(op+8, match+8, cpy);
+ op = cpy; /* wildcopy correction */
}
- op = cpy; /* wildcopy correction */
- }
-
- /* end of decoding */
- if (endOnInput)
- return (int) (((char*)op)-dst); /* Nb of output bytes decoded */
- else
- return (int) (((const char*)ip)-src); /* Nb of input bytes read */
- /* Overflow error detected */
-_output_error:
- return (int) (-(((const char*)ip)-src))-1;
+ /* end of decoding */
+ if (endOnInput) {
+ DEBUGLOG(5, "decoded %i bytes", (int) (((char*)op)-dst));
+ return (int) (((char*)op)-dst); /* Nb of output bytes decoded */
+ } else {
+ return (int) (((const char*)ip)-src); /* Nb of input bytes read */
+ }
+
+ /* Overflow error detected */
+ _output_error:
+ return (int) (-(((const char*)ip)-src))-1;
+ }
}
/*===== Instantiate the API decoding functions. =====*/
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
int LZ4_decompress_safe(const char* source, char* dest, int compressedSize, int maxDecompressedSize)
{
return LZ4_decompress_generic(source, dest, compressedSize, maxDecompressedSize,
@@ -1655,7 +2175,7 @@ int LZ4_decompress_safe(const char* source, char* dest, int compressedSize, int
(BYTE*)dest, NULL, 0);
}
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
int LZ4_decompress_safe_partial(const char* src, char* dst, int compressedSize, int targetOutputSize, int dstCapacity)
{
dstCapacity = MIN(targetOutputSize, dstCapacity);
@@ -1664,7 +2184,7 @@ int LZ4_decompress_safe_partial(const char* src, char* dst, int compressedSize,
noDict, (BYTE*)dst, NULL, 0);
}
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
int LZ4_decompress_fast(const char* source, char* dest, int originalSize)
{
return LZ4_decompress_generic(source, dest, 0, originalSize,
@@ -1674,7 +2194,7 @@ int LZ4_decompress_fast(const char* source, char* dest, int originalSize)
/*===== Instantiate a few more decoding cases, used more than once. =====*/
-LZ4_FORCE_O2_GCC_PPC64LE /* Exported, an obsolete API function. */
+LZ4_FORCE_O2 /* Exported, an obsolete API function. */
int LZ4_decompress_safe_withPrefix64k(const char* source, char* dest, int compressedSize, int maxOutputSize)
{
return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize,
@@ -1690,7 +2210,7 @@ int LZ4_decompress_fast_withPrefix64k(const char* source, char* dest, int origin
return LZ4_decompress_fast(source, dest, originalSize);
}
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
static int LZ4_decompress_safe_withSmallPrefix(const char* source, char* dest, int compressedSize, int maxOutputSize,
size_t prefixSize)
{
@@ -1699,7 +2219,7 @@ static int LZ4_decompress_safe_withSmallPrefix(const char* source, char* dest, i
(BYTE*)dest-prefixSize, NULL, 0);
}
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
int LZ4_decompress_safe_forceExtDict(const char* source, char* dest,
int compressedSize, int maxOutputSize,
const void* dictStart, size_t dictSize)
@@ -1709,7 +2229,7 @@ int LZ4_decompress_safe_forceExtDict(const char* source, char* dest,
(BYTE*)dest, (const BYTE*)dictStart, dictSize);
}
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
static int LZ4_decompress_fast_extDict(const char* source, char* dest, int originalSize,
const void* dictStart, size_t dictSize)
{
@@ -1745,12 +2265,13 @@ int LZ4_decompress_fast_doubleDict(const char* source, char* dest, int originalS
LZ4_streamDecode_t* LZ4_createStreamDecode(void)
{
LZ4_streamDecode_t* lz4s = (LZ4_streamDecode_t*) ALLOC_AND_ZERO(sizeof(LZ4_streamDecode_t));
+ LZ4_STATIC_ASSERT(LZ4_STREAMDECODESIZE >= sizeof(LZ4_streamDecode_t_internal)); /* A compilation error here means LZ4_STREAMDECODESIZE is not large enough */
return lz4s;
}
int LZ4_freeStreamDecode (LZ4_streamDecode_t* LZ4_stream)
{
- if (!LZ4_stream) return 0; /* support free on NULL */
+ if (LZ4_stream == NULL) { return 0; } /* support free on NULL */
FREEMEM(LZ4_stream);
return 0;
}
@@ -1797,7 +2318,7 @@ int LZ4_decoderRingBufferSize(int maxBlockSize)
If it's not possible, save the relevant part of decoded data into a safe buffer,
and indicate where it stands using LZ4_setStreamDecode()
*/
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* source, char* dest, int compressedSize, int maxOutputSize)
{
LZ4_streamDecode_t_internal* lz4sd = &LZ4_streamDecode->internal_donotuse;
@@ -1808,7 +2329,7 @@ int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const ch
assert(lz4sd->extDictSize == 0);
result = LZ4_decompress_safe(source, dest, compressedSize, maxOutputSize);
if (result <= 0) return result;
- lz4sd->prefixSize = result;
+ lz4sd->prefixSize = (size_t)result;
lz4sd->prefixEnd = (BYTE*)dest + result;
} else if (lz4sd->prefixEnd == (BYTE*)dest) {
/* They're rolling the current segment. */
@@ -1821,7 +2342,7 @@ int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const ch
result = LZ4_decompress_safe_doubleDict(source, dest, compressedSize, maxOutputSize,
lz4sd->prefixSize, lz4sd->externalDict, lz4sd->extDictSize);
if (result <= 0) return result;
- lz4sd->prefixSize += result;
+ lz4sd->prefixSize += (size_t)result;
lz4sd->prefixEnd += result;
} else {
/* The buffer wraps around, or they're switching to another buffer. */
@@ -1830,24 +2351,25 @@ int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const ch
result = LZ4_decompress_safe_forceExtDict(source, dest, compressedSize, maxOutputSize,
lz4sd->externalDict, lz4sd->extDictSize);
if (result <= 0) return result;
- lz4sd->prefixSize = result;
+ lz4sd->prefixSize = (size_t)result;
lz4sd->prefixEnd = (BYTE*)dest + result;
}
return result;
}
-LZ4_FORCE_O2_GCC_PPC64LE
+LZ4_FORCE_O2
int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* source, char* dest, int originalSize)
{
LZ4_streamDecode_t_internal* lz4sd = &LZ4_streamDecode->internal_donotuse;
int result;
+ assert(originalSize >= 0);
if (lz4sd->prefixSize == 0) {
assert(lz4sd->extDictSize == 0);
result = LZ4_decompress_fast(source, dest, originalSize);
if (result <= 0) return result;
- lz4sd->prefixSize = originalSize;
+ lz4sd->prefixSize = (size_t)originalSize;
lz4sd->prefixEnd = (BYTE*)dest + originalSize;
} else if (lz4sd->prefixEnd == (BYTE*)dest) {
if (lz4sd->prefixSize >= 64 KB - 1 || lz4sd->extDictSize == 0)
@@ -1856,7 +2378,7 @@ int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const ch
result = LZ4_decompress_fast_doubleDict(source, dest, originalSize,
lz4sd->prefixSize, lz4sd->externalDict, lz4sd->extDictSize);
if (result <= 0) return result;
- lz4sd->prefixSize += originalSize;
+ lz4sd->prefixSize += (size_t)originalSize;
lz4sd->prefixEnd += originalSize;
} else {
lz4sd->extDictSize = lz4sd->prefixSize;
@@ -1864,7 +2386,7 @@ int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const ch
result = LZ4_decompress_fast_extDict(source, dest, originalSize,
lz4sd->externalDict, lz4sd->extDictSize);
if (result <= 0) return result;
- lz4sd->prefixSize = originalSize;
+ lz4sd->prefixSize = (size_t)originalSize;
lz4sd->prefixEnd = (BYTE*)dest + originalSize;
}
@@ -1884,18 +2406,22 @@ int LZ4_decompress_safe_usingDict(const char* source, char* dest, int compressed
if (dictSize==0)
return LZ4_decompress_safe(source, dest, compressedSize, maxOutputSize);
if (dictStart+dictSize == dest) {
- if (dictSize >= 64 KB - 1)
+ if (dictSize >= 64 KB - 1) {
return LZ4_decompress_safe_withPrefix64k(source, dest, compressedSize, maxOutputSize);
- return LZ4_decompress_safe_withSmallPrefix(source, dest, compressedSize, maxOutputSize, dictSize);
+ }
+ assert(dictSize >= 0);
+ return LZ4_decompress_safe_withSmallPrefix(source, dest, compressedSize, maxOutputSize, (size_t)dictSize);
}
- return LZ4_decompress_safe_forceExtDict(source, dest, compressedSize, maxOutputSize, dictStart, dictSize);
+ assert(dictSize >= 0);
+ return LZ4_decompress_safe_forceExtDict(source, dest, compressedSize, maxOutputSize, dictStart, (size_t)dictSize);
}
int LZ4_decompress_fast_usingDict(const char* source, char* dest, int originalSize, const char* dictStart, int dictSize)
{
if (dictSize==0 || dictStart+dictSize == dest)
return LZ4_decompress_fast(source, dest, originalSize);
- return LZ4_decompress_fast_extDict(source, dest, originalSize, dictStart, dictSize);
+ assert(dictSize >= 0);
+ return LZ4_decompress_fast_extDict(source, dest, originalSize, dictStart, (size_t)dictSize);
}
@@ -1907,9 +2433,9 @@ int LZ4_compress_limitedOutput(const char* source, char* dest, int inputSize, in
{
return LZ4_compress_default(source, dest, inputSize, maxOutputSize);
}
-int LZ4_compress(const char* source, char* dest, int inputSize)
+int LZ4_compress(const char* src, char* dest, int srcSize)
{
- return LZ4_compress_default(source, dest, inputSize, LZ4_compressBound(inputSize));
+ return LZ4_compress_default(src, dest, srcSize, LZ4_compressBound(srcSize));
}
int LZ4_compress_limitedOutput_withState (void* state, const char* src, char* dst, int srcSize, int dstSize)
{
@@ -1945,7 +2471,7 @@ int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize,
/* Obsolete Streaming functions */
-int LZ4_sizeofStreamState() { return LZ4_STREAMSIZE; }
+int LZ4_sizeofStreamState(void) { return LZ4_STREAMSIZE; }
int LZ4_resetStreamState(void* state, char* inputBuffer)
{
diff --git a/lib/lz4.h b/lib/lz4.h
index 059ef7c..7ab1e48 100644
--- a/lib/lz4.h
+++ b/lib/lz4.h
@@ -46,24 +46,31 @@ extern "C" {
/**
Introduction
- LZ4 is lossless compression algorithm, providing compression speed at 500 MB/s per core,
+ LZ4 is lossless compression algorithm, providing compression speed >500 MB/s per core,
scalable with multi-cores CPU. It features an extremely fast decoder, with speed in
multiple GB/s per core, typically reaching RAM speed limits on multi-core systems.
The LZ4 compression library provides in-memory compression and decompression functions.
+ It gives full buffer control to user.
Compression can be done in:
- a single step (described as Simple Functions)
- a single step, reusing a context (described in Advanced Functions)
- unbounded multiple steps (described as Streaming compression)
- lz4.h provides block compression functions. It gives full buffer control to user.
- Decompressing an lz4-compressed block also requires metadata (such as compressed size).
- Each application is free to encode such metadata in whichever way it wants.
+ lz4.h generates and decodes LZ4-compressed blocks (doc/lz4_Block_format.md).
+ Decompressing such a compressed block requires additional metadata.
+ Exact metadata depends on exact decompression function.
+ For the typical case of LZ4_decompress_safe(),
+ metadata includes block's compressed size, and maximum bound of decompressed size.
+ Each application is free to encode and pass such metadata in whichever way it wants.
- An additional format, called LZ4 frame specification (doc/lz4_Frame_format.md),
- take care of encoding standard metadata alongside LZ4-compressed blocks.
- Frame format is required for interoperability.
- It is delivered through a companion API, declared in lz4frame.h.
+ lz4.h only handle blocks, it can not generate Frames.
+
+ Blocks are different from Frames (doc/lz4_Frame_format.md).
+ Frames bundle both blocks and metadata in a specified manner.
+ Embedding metadata is required for compressed data to be self-contained and portable.
+ Frame format is delivered through a companion API, declared in lz4frame.h.
+ The `lz4` CLI can only manage frames.
*/
/*^***************************************************************
@@ -92,7 +99,7 @@ extern "C" {
/*------ Version ------*/
#define LZ4_VERSION_MAJOR 1 /* for breaking interface changes */
-#define LZ4_VERSION_MINOR 8 /* for new (non-breaking) interface capabilities */
+#define LZ4_VERSION_MINOR 9 /* for new (non-breaking) interface capabilities */
#define LZ4_VERSION_RELEASE 3 /* for tweaks, bug-fixes, or development */
#define LZ4_VERSION_NUMBER (LZ4_VERSION_MAJOR *100*100 + LZ4_VERSION_MINOR *100 + LZ4_VERSION_RELEASE)
@@ -103,7 +110,7 @@ extern "C" {
#define LZ4_VERSION_STRING LZ4_EXPAND_AND_QUOTE(LZ4_LIB_VERSION)
LZ4LIB_API int LZ4_versionNumber (void); /**< library version number; useful to check dll version */
-LZ4LIB_API const char* LZ4_versionString (void); /**< library version string; unseful to check dll version */
+LZ4LIB_API const char* LZ4_versionString (void); /**< library version string; useful to check dll version */
/*-************************************
@@ -112,40 +119,48 @@ LZ4LIB_API const char* LZ4_versionString (void); /**< library version string;
/*!
* LZ4_MEMORY_USAGE :
* Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.)
- * Increasing memory usage improves compression ratio
- * Reduced memory usage may improve speed, thanks to cache effect
+ * Increasing memory usage improves compression ratio.
+ * Reduced memory usage may improve speed, thanks to better cache locality.
* Default value is 14, for 16KB, which nicely fits into Intel x86 L1 cache
*/
#ifndef LZ4_MEMORY_USAGE
# define LZ4_MEMORY_USAGE 14
#endif
+
/*-************************************
* Simple Functions
**************************************/
/*! LZ4_compress_default() :
- Compresses 'srcSize' bytes from buffer 'src'
- into already allocated 'dst' buffer of size 'dstCapacity'.
- Compression is guaranteed to succeed if 'dstCapacity' >= LZ4_compressBound(srcSize).
- It also runs faster, so it's a recommended setting.
- If the function cannot compress 'src' into a more limited 'dst' budget,
- compression stops *immediately*, and the function result is zero.
- Note : as a consequence, 'dst' content is not valid.
- Note 2 : This function is protected against buffer overflow scenarios (never writes outside 'dst' buffer, nor read outside 'source' buffer).
- srcSize : max supported value is LZ4_MAX_INPUT_SIZE.
- dstCapacity : size of buffer 'dst' (which must be already allocated)
- return : the number of bytes written into buffer 'dst' (necessarily <= dstCapacity)
- or 0 if compression fails */
+ * Compresses 'srcSize' bytes from buffer 'src'
+ * into already allocated 'dst' buffer of size 'dstCapacity'.
+ * Compression is guaranteed to succeed if 'dstCapacity' >= LZ4_compressBound(srcSize).
+ * It also runs faster, so it's a recommended setting.
+ * If the function cannot compress 'src' into a more limited 'dst' budget,
+ * compression stops *immediately*, and the function result is zero.
+ * In which case, 'dst' content is undefined (invalid).
+ * srcSize : max supported value is LZ4_MAX_INPUT_SIZE.
+ * dstCapacity : size of buffer 'dst' (which must be already allocated)
+ * @return : the number of bytes written into buffer 'dst' (necessarily <= dstCapacity)
+ * or 0 if compression fails
+ * Note : This function is protected against buffer overflow scenarios (never writes outside 'dst' buffer, nor read outside 'source' buffer).
+ */
LZ4LIB_API int LZ4_compress_default(const char* src, char* dst, int srcSize, int dstCapacity);
/*! LZ4_decompress_safe() :
- compressedSize : is the exact complete size of the compressed block.
- dstCapacity : is the size of destination buffer, which must be already allocated.
- return : the number of bytes decompressed into destination buffer (necessarily <= dstCapacity)
- If destination buffer is not large enough, decoding will stop and output an error code (negative value).
- If the source stream is detected malformed, the function will stop decoding and return a negative result.
- This function is protected against malicious data packets.
-*/
+ * compressedSize : is the exact complete size of the compressed block.
+ * dstCapacity : is the size of destination buffer (which must be already allocated), presumed an upper bound of decompressed size.
+ * @return : the number of bytes decompressed into destination buffer (necessarily <= dstCapacity)
+ * If destination buffer is not large enough, decoding will stop and output an error code (negative value).
+ * If the source stream is detected malformed, the function will stop decoding and return a negative result.
+ * Note 1 : This function is protected against malicious data packets :
+ * it will never writes outside 'dst' buffer, nor read outside 'source' buffer,
+ * even if the compressed block is maliciously modified to order the decoder to do these actions.
+ * In such case, the decoder stops immediately, and considers the compressed block malformed.
+ * Note 2 : compressedSize and dstCapacity must be provided to the function, the compressed block does not contain them.
+ * The implementation is free to send / store / derive this information in whichever way is most beneficial.
+ * If there is a need for a different format which bundles together both compressed data and its metadata, consider looking at lz4frame.h instead.
+ */
LZ4LIB_API int LZ4_decompress_safe (const char* src, char* dst, int compressedSize, int dstCapacity);
@@ -155,8 +170,7 @@ LZ4LIB_API int LZ4_decompress_safe (const char* src, char* dst, int compressedSi
#define LZ4_MAX_INPUT_SIZE 0x7E000000 /* 2 113 929 216 bytes */
#define LZ4_COMPRESSBOUND(isize) ((unsigned)(isize) > (unsigned)LZ4_MAX_INPUT_SIZE ? 0 : (isize) + ((isize)/255) + 16)
-/*!
-LZ4_compressBound() :
+/*! LZ4_compressBound() :
Provides the maximum size that LZ4 compression may output in a "worst case" scenario (input data not compressible)
This function is primarily useful for memory allocation purposes (destination buffer size).
Macro LZ4_COMPRESSBOUND() is also provided for compilation-time evaluation (stack memory allocation for example).
@@ -167,24 +181,23 @@ LZ4_compressBound() :
*/
LZ4LIB_API int LZ4_compressBound(int inputSize);
-/*!
-LZ4_compress_fast() :
+/*! LZ4_compress_fast() :
Same as LZ4_compress_default(), but allows selection of "acceleration" factor.
The larger the acceleration value, the faster the algorithm, but also the lesser the compression.
It's a trade-off. It can be fine tuned, with each successive value providing roughly +~3% to speed.
An acceleration value of "1" is the same as regular LZ4_compress_default()
- Values <= 0 will be replaced by ACCELERATION_DEFAULT (currently == 1, see lz4.c).
+ Values <= 0 will be replaced by LZ4_ACCELERATION_DEFAULT (currently == 1, see lz4.c).
+ Values > LZ4_ACCELERATION_MAX will be replaced by LZ4_ACCELERATION_MAX (currently == 65537, see lz4.c).
*/
LZ4LIB_API int LZ4_compress_fast (const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
-/*!
-LZ4_compress_fast_extState() :
- Same compression function, just using an externally allocated memory space to store compression state.
- Use LZ4_sizeofState() to know how much memory must be allocated,
- and allocate it on 8-bytes boundaries (using malloc() typically).
- Then, provide this buffer as 'void* state' to compression function.
-*/
+/*! LZ4_compress_fast_extState() :
+ * Same as LZ4_compress_fast(), using an externally allocated memory space for its state.
+ * Use LZ4_sizeofState() to know how much memory must be allocated,
+ * and allocate it on 8-bytes boundaries (using `malloc()` typically).
+ * Then, provide this buffer as `void* state` to compression function.
+ */
LZ4LIB_API int LZ4_sizeofState(void);
LZ4LIB_API int LZ4_compress_fast_extState (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
@@ -200,54 +213,54 @@ LZ4LIB_API int LZ4_compress_fast_extState (void* state, const char* src, char* d
* New value is necessarily <= input value.
* @return : Nb bytes written into 'dst' (necessarily <= targetDestSize)
* or 0 if compression fails.
-*/
-LZ4LIB_API int LZ4_compress_destSize (const char* src, char* dst, int* srcSizePtr, int targetDstSize);
-
-
-/*! LZ4_decompress_fast() : **unsafe!**
- * This function used to be a bit faster than LZ4_decompress_safe(),
- * though situation has changed in recent versions,
- * and now `LZ4_decompress_safe()` can be as fast and sometimes faster than `LZ4_decompress_fast()`.
- * Moreover, LZ4_decompress_fast() is not protected vs malformed input, as it doesn't perform full validation of compressed data.
- * As a consequence, this function is no longer recommended, and may be deprecated in future versions.
- * It's only remaining specificity is that it can decompress data without knowing its compressed size.
*
- * originalSize : is the uncompressed size to regenerate.
- * `dst` must be already allocated, its size must be >= 'originalSize' bytes.
- * @return : number of bytes read from source buffer (== compressed size).
- * If the source stream is detected malformed, the function stops decoding and returns a negative result.
- * note : This function requires uncompressed originalSize to be known in advance.
- * The function never writes past the output buffer.
- * However, since it doesn't know its 'src' size, it may read past the intended input.
- * Also, because match offsets are not validated during decoding,
- * reads from 'src' may underflow.
- * Use this function in trusted environment **only**.
+ * Note : from v1.8.2 to v1.9.1, this function had a bug (fixed un v1.9.2+):
+ * the produced compressed content could, in specific circumstances,
+ * require to be decompressed into a destination buffer larger
+ * by at least 1 byte than the content to decompress.
+ * If an application uses `LZ4_compress_destSize()`,
+ * it's highly recommended to update liblz4 to v1.9.2 or better.
+ * If this can't be done or ensured,
+ * the receiving decompression function should provide
+ * a dstCapacity which is > decompressedSize, by at least 1 byte.
+ * See https://github.com/lz4/lz4/issues/859 for details
*/
-LZ4LIB_API int LZ4_decompress_fast (const char* src, char* dst, int originalSize);
+LZ4LIB_API int LZ4_compress_destSize (const char* src, char* dst, int* srcSizePtr, int targetDstSize);
+
/*! LZ4_decompress_safe_partial() :
* Decompress an LZ4 compressed block, of size 'srcSize' at position 'src',
* into destination buffer 'dst' of size 'dstCapacity'.
* Up to 'targetOutputSize' bytes will be decoded.
- * The function stops decoding on reaching this objective,
- * which can boost performance when only the beginning of a block is required.
+ * The function stops decoding on reaching this objective.
+ * This can be useful to boost performance
+ * whenever only the beginning of a block is required.
*
- * @return : the number of bytes decoded in `dst` (necessarily <= dstCapacity)
+ * @return : the number of bytes decoded in `dst` (necessarily <= targetOutputSize)
* If source stream is detected malformed, function returns a negative result.
*
- * Note : @return can be < targetOutputSize, if compressed block contains less data.
+ * Note 1 : @return can be < targetOutputSize, if compressed block contains less data.
*
- * Note 2 : this function features 2 parameters, targetOutputSize and dstCapacity,
- * and expects targetOutputSize <= dstCapacity.
- * It effectively stops decoding on reaching targetOutputSize,
+ * Note 2 : targetOutputSize must be <= dstCapacity
+ *
+ * Note 3 : this function effectively stops decoding on reaching targetOutputSize,
* so dstCapacity is kind of redundant.
- * This is because in a previous version of this function,
- * decoding operation would not "break" a sequence in the middle.
- * As a consequence, there was no guarantee that decoding would stop at exactly targetOutputSize,
+ * This is because in older versions of this function,
+ * decoding operation would still write complete sequences.
+ * Therefore, there was no guarantee that it would stop writing at exactly targetOutputSize,
* it could write more bytes, though only up to dstCapacity.
* Some "margin" used to be required for this operation to work properly.
- * This is no longer necessary.
- * The function nonetheless keeps its signature, in an effort to not break API.
+ * Thankfully, this is no longer necessary.
+ * The function nonetheless keeps the same signature, in an effort to preserve API compatibility.
+ *
+ * Note 4 : If srcSize is the exact size of the block,
+ * then targetOutputSize can be any value,
+ * including larger than the block's decompressed size.
+ * The function will, at most, generate block's decompressed size.
+ *
+ * Note 5 : If srcSize is _larger_ than block's compressed size,
+ * then targetOutputSize **MUST** be <= block's decompressed size.
+ * Otherwise, *silent corruption will occur*.
*/
LZ4LIB_API int LZ4_decompress_safe_partial (const char* src, char* dst, int srcSize, int targetOutputSize, int dstCapacity);
@@ -257,30 +270,49 @@ LZ4LIB_API int LZ4_decompress_safe_partial (const char* src, char* dst, int srcS
***********************************************/
typedef union LZ4_stream_u LZ4_stream_t; /* incomplete type (defined later) */
-/*! LZ4_createStream() and LZ4_freeStream() :
- * LZ4_createStream() will allocate and initialize an `LZ4_stream_t` structure.
- * LZ4_freeStream() releases its memory.
- */
LZ4LIB_API LZ4_stream_t* LZ4_createStream(void);
LZ4LIB_API int LZ4_freeStream (LZ4_stream_t* streamPtr);
-/*! LZ4_resetStream() :
- * An LZ4_stream_t structure can be allocated once and re-used multiple times.
- * Use this function to start compressing a new stream.
+/*! LZ4_resetStream_fast() : v1.9.0+
+ * Use this to prepare an LZ4_stream_t for a new chain of dependent blocks
+ * (e.g., LZ4_compress_fast_continue()).
+ *
+ * An LZ4_stream_t must be initialized once before usage.
+ * This is automatically done when created by LZ4_createStream().
+ * However, should the LZ4_stream_t be simply declared on stack (for example),
+ * it's necessary to initialize it first, using LZ4_initStream().
+ *
+ * After init, start any new stream with LZ4_resetStream_fast().
+ * A same LZ4_stream_t can be re-used multiple times consecutively
+ * and compress multiple streams,
+ * provided that it starts each new stream with LZ4_resetStream_fast().
+ *
+ * LZ4_resetStream_fast() is much faster than LZ4_initStream(),
+ * but is not compatible with memory regions containing garbage data.
+ *
+ * Note: it's only useful to call LZ4_resetStream_fast()
+ * in the context of streaming compression.
+ * The *extState* functions perform their own resets.
+ * Invoking LZ4_resetStream_fast() before is redundant, and even counterproductive.
*/
-LZ4LIB_API void LZ4_resetStream (LZ4_stream_t* streamPtr);
+LZ4LIB_API void LZ4_resetStream_fast (LZ4_stream_t* streamPtr);
/*! LZ4_loadDict() :
- * Use this function to load a static dictionary into LZ4_stream_t.
- * Any previous data will be forgotten, only 'dictionary' will remain in memory.
+ * Use this function to reference a static dictionary into LZ4_stream_t.
+ * The dictionary must remain available during compression.
+ * LZ4_loadDict() triggers a reset, so any previous data will be forgotten.
+ * The same dictionary will have to be loaded on decompression side for successful decoding.
+ * Dictionary are useful for better compression of small data (KB range).
+ * While LZ4 accept any input as dictionary,
+ * results are generally better when using Zstandard's Dictionary Builder.
* Loading a size of 0 is allowed, and is the same as reset.
- * @return : dictionary size, in bytes (necessarily <= 64 KB)
+ * @return : loaded dictionary size, in bytes (necessarily <= 64 KB)
*/
LZ4LIB_API int LZ4_loadDict (LZ4_stream_t* streamPtr, const char* dictionary, int dictSize);
/*! LZ4_compress_fast_continue() :
* Compress 'src' content using data from previously compressed blocks, for better compression ratio.
- * 'dst' buffer must be already allocated.
+ * 'dst' buffer must be already allocated.
* If dstCapacity >= LZ4_compressBound(srcSize), compression is guaranteed to succeed, and runs faster.
*
* @return : size of compressed block
@@ -288,10 +320,10 @@ LZ4LIB_API int LZ4_loadDict (LZ4_stream_t* streamPtr, const char* dictionary, in
*
* Note 1 : Each invocation to LZ4_compress_fast_continue() generates a new block.
* Each block has precise boundaries.
+ * Each block must be decompressed separately, calling LZ4_decompress_*() with relevant metadata.
* It's not possible to append blocks together and expect a single invocation of LZ4_decompress_*() to decompress them together.
- * Each block must be decompressed separately, calling LZ4_decompress_*() with associated metadata.
*
- * Note 2 : The previous 64KB of source data is __assumed__ to remain present, unmodified, at same address in memory!
+ * Note 2 : The previous 64KB of source data is __assumed__ to remain present, unmodified, at same address in memory !
*
* Note 3 : When input is structured as a double-buffer, each buffer can have any size, including < 64 KB.
* Make sure that buffers are separated, by at least one byte.
@@ -299,7 +331,7 @@ LZ4LIB_API int LZ4_loadDict (LZ4_stream_t* streamPtr, const char* dictionary, in
*
* Note 4 : If input buffer is a ring-buffer, it can have any size, including < 64 KB.
*
- * Note 5 : After an error, the stream status is invalid, it can only be reset or freed.
+ * Note 5 : After an error, the stream status is undefined (invalid), it can only be reset or freed.
*/
LZ4LIB_API int LZ4_compress_fast_continue (LZ4_stream_t* streamPtr, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
@@ -335,7 +367,7 @@ LZ4LIB_API int LZ4_freeStreamDecode (LZ4_streamDecode_t* LZ4_str
*/
LZ4LIB_API int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const char* dictionary, int dictSize);
-/*! LZ4_decoderRingBufferSize() : v1.8.2
+/*! LZ4_decoderRingBufferSize() : v1.8.2+
* Note : in a ring buffer scenario (optional),
* blocks are presumed decompressed next to each other
* up to the moment there is not enough remaining space for next block (remainingSize < maxBlockSize),
@@ -347,7 +379,7 @@ LZ4LIB_API int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const
* or 0 if there is an error (invalid maxBlockSize).
*/
LZ4LIB_API int LZ4_decoderRingBufferSize(int maxBlockSize);
-#define LZ4_DECODER_RING_BUFFER_SIZE(mbs) (65536 + 14 + (mbs)) /* for static allocation; mbs presumed valid */
+#define LZ4_DECODER_RING_BUFFER_SIZE(maxBlockSize) (65536 + 14 + (maxBlockSize)) /* for static allocation; maxBlockSize presumed valid */
/*! LZ4_decompress_*_continue() :
* These decoding functions allow decompression of consecutive blocks in "streaming" mode.
@@ -375,83 +407,72 @@ LZ4LIB_API int LZ4_decoderRingBufferSize(int maxBlockSize);
* then indicate where this data is saved using LZ4_setStreamDecode(), before decompressing next block.
*/
LZ4LIB_API int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* src, char* dst, int srcSize, int dstCapacity);
-LZ4LIB_API int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* src, char* dst, int originalSize);
/*! LZ4_decompress_*_usingDict() :
* These decoding functions work the same as
* a combination of LZ4_setStreamDecode() followed by LZ4_decompress_*_continue()
* They are stand-alone, and don't need an LZ4_streamDecode_t structure.
- * Dictionary is presumed stable : it must remain accessible and unmodified during next decompression.
+ * Dictionary is presumed stable : it must remain accessible and unmodified during decompression.
+ * Performance tip : Decompression speed can be substantially increased
+ * when dst == dictStart + dictSize.
*/
LZ4LIB_API int LZ4_decompress_safe_usingDict (const char* src, char* dst, int srcSize, int dstCapcity, const char* dictStart, int dictSize);
-LZ4LIB_API int LZ4_decompress_fast_usingDict (const char* src, char* dst, int originalSize, const char* dictStart, int dictSize);
+
+#endif /* LZ4_H_2983827168210 */
-/*^**********************************************
+/*^*************************************
* !!!!!! STATIC LINKING ONLY !!!!!!
- ***********************************************/
+ ***************************************/
-/*-************************************
- * Unstable declarations
- **************************************
- * Declarations in this section should be considered unstable.
- * Use at your own peril, etc., etc.
- * They may be removed in the future.
- * Their signatures may change.
- **************************************/
+/*-****************************************************************************
+ * Experimental section
+ *
+ * Symbols declared in this section must be considered unstable. Their
+ * signatures or semantics may change, or they may be removed altogether in the
+ * future. They are therefore only safe to depend on when the caller is
+ * statically linked against the library.
+ *
+ * To protect against unsafe usage, not only are the declarations guarded,
+ * the definitions are hidden by default
+ * when building LZ4 as a shared/dynamic library.
+ *
+ * In order to access these declarations,
+ * define LZ4_STATIC_LINKING_ONLY in your application
+ * before including LZ4's headers.
+ *
+ * In order to make their implementations accessible dynamically, you must
+ * define LZ4_PUBLISH_STATIC_FUNCTIONS when building the LZ4 library.
+ ******************************************************************************/
#ifdef LZ4_STATIC_LINKING_ONLY
-/*! LZ4_resetStream_fast() :
- * Use this, like LZ4_resetStream(), to prepare a context for a new chain of
- * calls to a streaming API (e.g., LZ4_compress_fast_continue()).
- *
- * Note:
- * Using this in advance of a non- streaming-compression function is redundant,
- * and potentially bad for performance, since they all perform their own custom
- * reset internally.
- *
- * Differences from LZ4_resetStream():
- * When an LZ4_stream_t is known to be in a internally coherent state,
- * it can often be prepared for a new compression with almost no work, only
- * sometimes falling back to the full, expensive reset that is always required
- * when the stream is in an indeterminate state (i.e., the reset performed by
- * LZ4_resetStream()).
- *
- * LZ4_streams are guaranteed to be in a valid state when:
- * - returned from LZ4_createStream()
- * - reset by LZ4_resetStream()
- * - memset(stream, 0, sizeof(LZ4_stream_t)), though this is discouraged
- * - the stream was in a valid state and was reset by LZ4_resetStream_fast()
- * - the stream was in a valid state and was then used in any compression call
- * that returned success
- * - the stream was in an indeterminate state and was used in a compression
- * call that fully reset the state (e.g., LZ4_compress_fast_extState()) and
- * that returned success
- *
- * When a stream isn't known to be in a valid state, it is not safe to pass to
- * any fastReset or streaming function. It must first be cleansed by the full
- * LZ4_resetStream().
- */
-LZ4LIB_API void LZ4_resetStream_fast (LZ4_stream_t* streamPtr);
+#ifndef LZ4_STATIC_3504398509
+#define LZ4_STATIC_3504398509
+
+#ifdef LZ4_PUBLISH_STATIC_FUNCTIONS
+#define LZ4LIB_STATIC_API LZ4LIB_API
+#else
+#define LZ4LIB_STATIC_API
+#endif
+
/*! LZ4_compress_fast_extState_fastReset() :
* A variant of LZ4_compress_fast_extState().
*
- * Using this variant avoids an expensive initialization step. It is only safe
- * to call if the state buffer is known to be correctly initialized already
- * (see above comment on LZ4_resetStream_fast() for a definition of "correctly
- * initialized"). From a high level, the difference is that this function
- * initializes the provided state with a call to something like
- * LZ4_resetStream_fast() while LZ4_compress_fast_extState() starts with a
- * call to LZ4_resetStream().
+ * Using this variant avoids an expensive initialization step.
+ * It is only safe to call if the state buffer is known to be correctly initialized already
+ * (see above comment on LZ4_resetStream_fast() for a definition of "correctly initialized").
+ * From a high level, the difference is that
+ * this function initializes the provided state with a call to something like LZ4_resetStream_fast()
+ * while LZ4_compress_fast_extState() starts with a call to LZ4_resetStream().
*/
-LZ4LIB_API int LZ4_compress_fast_extState_fastReset (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
+LZ4LIB_STATIC_API int LZ4_compress_fast_extState_fastReset (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration);
/*! LZ4_attach_dictionary() :
- * This is an experimental API that allows for the efficient use of a
- * static dictionary many times.
+ * This is an experimental API that allows
+ * efficient use of a static dictionary many times.
*
* Rather than re-loading the dictionary buffer into a working context before
* each compression, or copying a pre-loaded dictionary's LZ4_stream_t into a
@@ -462,8 +483,8 @@ LZ4LIB_API int LZ4_compress_fast_extState_fastReset (void* state, const char* sr
* Currently, only streams which have been prepared by LZ4_loadDict() should
* be expected to work.
*
- * Alternatively, the provided dictionary stream pointer may be NULL, in which
- * case any existing dictionary stream is unset.
+ * Alternatively, the provided dictionaryStream may be NULL,
+ * in which case any existing dictionary stream is unset.
*
* If a dictionary is provided, it replaces any pre-existing stream history.
* The dictionary contents are the only history that can be referenced and
@@ -475,89 +496,163 @@ LZ4LIB_API int LZ4_compress_fast_extState_fastReset (void* state, const char* sr
* stream (and source buffer) must remain in-place / accessible / unchanged
* through the completion of the first compression call on the stream.
*/
-LZ4LIB_API void LZ4_attach_dictionary(LZ4_stream_t *working_stream, const LZ4_stream_t *dictionary_stream);
+LZ4LIB_STATIC_API void LZ4_attach_dictionary(LZ4_stream_t* workingStream, const LZ4_stream_t* dictionaryStream);
+
+
+/*! In-place compression and decompression
+ *
+ * It's possible to have input and output sharing the same buffer,
+ * for highly contrained memory environments.
+ * In both cases, it requires input to lay at the end of the buffer,
+ * and decompression to start at beginning of the buffer.
+ * Buffer size must feature some margin, hence be larger than final size.
+ *
+ * |<------------------------buffer--------------------------------->|
+ * |<-----------compressed data--------->|
+ * |<-----------decompressed size------------------>|
+ * |<----margin---->|
+ *
+ * This technique is more useful for decompression,
+ * since decompressed size is typically larger,
+ * and margin is short.
+ *
+ * In-place decompression will work inside any buffer
+ * which size is >= LZ4_DECOMPRESS_INPLACE_BUFFER_SIZE(decompressedSize).
+ * This presumes that decompressedSize > compressedSize.
+ * Otherwise, it means compression actually expanded data,
+ * and it would be more efficient to store such data with a flag indicating it's not compressed.
+ * This can happen when data is not compressible (already compressed, or encrypted).
+ *
+ * For in-place compression, margin is larger, as it must be able to cope with both
+ * history preservation, requiring input data to remain unmodified up to LZ4_DISTANCE_MAX,
+ * and data expansion, which can happen when input is not compressible.
+ * As a consequence, buffer size requirements are much higher,
+ * and memory savings offered by in-place compression are more limited.
+ *
+ * There are ways to limit this cost for compression :
+ * - Reduce history size, by modifying LZ4_DISTANCE_MAX.
+ * Note that it is a compile-time constant, so all compressions will apply this limit.
+ * Lower values will reduce compression ratio, except when input_size < LZ4_DISTANCE_MAX,
+ * so it's a reasonable trick when inputs are known to be small.
+ * - Require the compressor to deliver a "maximum compressed size".
+ * This is the `dstCapacity` parameter in `LZ4_compress*()`.
+ * When this size is < LZ4_COMPRESSBOUND(inputSize), then compression can fail,
+ * in which case, the return code will be 0 (zero).
+ * The caller must be ready for these cases to happen,
+ * and typically design a backup scheme to send data uncompressed.
+ * The combination of both techniques can significantly reduce
+ * the amount of margin required for in-place compression.
+ *
+ * In-place compression can work in any buffer
+ * which size is >= (maxCompressedSize)
+ * with maxCompressedSize == LZ4_COMPRESSBOUND(srcSize) for guaranteed compression success.
+ * LZ4_COMPRESS_INPLACE_BUFFER_SIZE() depends on both maxCompressedSize and LZ4_DISTANCE_MAX,
+ * so it's possible to reduce memory requirements by playing with them.
+ */
+
+#define LZ4_DECOMPRESS_INPLACE_MARGIN(compressedSize) (((compressedSize) >> 8) + 32)
+#define LZ4_DECOMPRESS_INPLACE_BUFFER_SIZE(decompressedSize) ((decompressedSize) + LZ4_DECOMPRESS_INPLACE_MARGIN(decompressedSize)) /**< note: presumes that compressedSize < decompressedSize. note2: margin is overestimated a bit, since it could use compressedSize instead */
+#ifndef LZ4_DISTANCE_MAX /* history window size; can be user-defined at compile time */
+# define LZ4_DISTANCE_MAX 65535 /* set to maximum value by default */
#endif
-/*-************************************
- * Private definitions
- **************************************
- * Do not use these definitions.
- * They are exposed to allow static allocation of `LZ4_stream_t` and `LZ4_streamDecode_t`.
- * Using these definitions will expose code to API and/or ABI break in future versions of the library.
- **************************************/
+#define LZ4_COMPRESS_INPLACE_MARGIN (LZ4_DISTANCE_MAX + 32) /* LZ4_DISTANCE_MAX can be safely replaced by srcSize when it's smaller */
+#define LZ4_COMPRESS_INPLACE_BUFFER_SIZE(maxCompressedSize) ((maxCompressedSize) + LZ4_COMPRESS_INPLACE_MARGIN) /**< maxCompressedSize is generally LZ4_COMPRESSBOUND(inputSize), but can be set to any lower value, with the risk that compression can fail (return code 0(zero)) */
+
+#endif /* LZ4_STATIC_3504398509 */
+#endif /* LZ4_STATIC_LINKING_ONLY */
+
+
+
+#ifndef LZ4_H_98237428734687
+#define LZ4_H_98237428734687
+
+/*-************************************************************
+ * Private Definitions
+ **************************************************************
+ * Do not use these definitions directly.
+ * They are only exposed to allow static allocation of `LZ4_stream_t` and `LZ4_streamDecode_t`.
+ * Accessing members will expose user code to API and/or ABI break in future versions of the library.
+ **************************************************************/
#define LZ4_HASHLOG (LZ4_MEMORY_USAGE-2)
#define LZ4_HASHTABLESIZE (1 << LZ4_MEMORY_USAGE)
#define LZ4_HASH_SIZE_U32 (1 << LZ4_HASHLOG) /* required as macro for static allocation */
#if defined(__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
-#include <stdint.h>
-
-typedef struct LZ4_stream_t_internal LZ4_stream_t_internal;
-struct LZ4_stream_t_internal {
- uint32_t hashTable[LZ4_HASH_SIZE_U32];
- uint32_t currentOffset;
- uint16_t initCheck;
- uint16_t tableType;
- const uint8_t* dictionary;
- const LZ4_stream_t_internal* dictCtx;
- uint32_t dictSize;
-};
-
-typedef struct {
- const uint8_t* externalDict;
- size_t extDictSize;
- const uint8_t* prefixEnd;
- size_t prefixSize;
-} LZ4_streamDecode_t_internal;
-
+# include <stdint.h>
+ typedef int8_t LZ4_i8;
+ typedef uint8_t LZ4_byte;
+ typedef uint16_t LZ4_u16;
+ typedef uint32_t LZ4_u32;
#else
+ typedef signed char LZ4_i8;
+ typedef unsigned char LZ4_byte;
+ typedef unsigned short LZ4_u16;
+ typedef unsigned int LZ4_u32;
+#endif
typedef struct LZ4_stream_t_internal LZ4_stream_t_internal;
struct LZ4_stream_t_internal {
- unsigned int hashTable[LZ4_HASH_SIZE_U32];
- unsigned int currentOffset;
- unsigned short initCheck;
- unsigned short tableType;
- const unsigned char* dictionary;
+ LZ4_u32 hashTable[LZ4_HASH_SIZE_U32];
+ LZ4_u32 currentOffset;
+ LZ4_u32 tableType;
+ const LZ4_byte* dictionary;
const LZ4_stream_t_internal* dictCtx;
- unsigned int dictSize;
+ LZ4_u32 dictSize;
};
typedef struct {
- const unsigned char* externalDict;
+ const LZ4_byte* externalDict;
size_t extDictSize;
- const unsigned char* prefixEnd;
+ const LZ4_byte* prefixEnd;
size_t prefixSize;
} LZ4_streamDecode_t_internal;
-#endif
-/*!
- * LZ4_stream_t :
- * information structure to track an LZ4 stream.
- * init this structure before first use.
- * note : only use in association with static linking !
- * this definition is not API/ABI safe,
- * it may change in a future version !
+/*! LZ4_stream_t :
+ * Do not use below internal definitions directly !
+ * Declare or allocate an LZ4_stream_t instead.
+ * LZ4_stream_t can also be created using LZ4_createStream(), which is recommended.
+ * The structure definition can be convenient for static allocation
+ * (on stack, or as part of larger structure).
+ * Init this structure with LZ4_initStream() before first use.
+ * note : only use this definition in association with static linking !
+ * this definition is not API/ABI safe, and may change in future versions.
*/
-#define LZ4_STREAMSIZE_U64 ((1 << (LZ4_MEMORY_USAGE-3)) + 4)
-#define LZ4_STREAMSIZE (LZ4_STREAMSIZE_U64 * sizeof(unsigned long long))
+#define LZ4_STREAMSIZE 16416 /* static size, for inter-version compatibility */
+#define LZ4_STREAMSIZE_VOIDP (LZ4_STREAMSIZE / sizeof(void*))
union LZ4_stream_u {
- unsigned long long table[LZ4_STREAMSIZE_U64];
+ void* table[LZ4_STREAMSIZE_VOIDP];
LZ4_stream_t_internal internal_donotuse;
-} ; /* previously typedef'd to LZ4_stream_t */
+}; /* previously typedef'd to LZ4_stream_t */
-/*!
- * LZ4_streamDecode_t :
- * information structure to track an LZ4 stream during decompression.
- * init this structure using LZ4_setStreamDecode (or memset()) before first use
- * note : only use in association with static linking !
- * this definition is not API/ABI safe,
- * and may change in a future version !
+/*! LZ4_initStream() : v1.9.0+
+ * An LZ4_stream_t structure must be initialized at least once.
+ * This is automatically done when invoking LZ4_createStream(),
+ * but it's not when the structure is simply declared on stack (for example).
+ *
+ * Use LZ4_initStream() to properly initialize a newly declared LZ4_stream_t.
+ * It can also initialize any arbitrary buffer of sufficient size,
+ * and will @return a pointer of proper type upon initialization.
+ *
+ * Note : initialization fails if size and alignment conditions are not respected.
+ * In which case, the function will @return NULL.
+ * Note2: An LZ4_stream_t structure guarantees correct alignment and size.
+ * Note3: Before v1.9.0, use LZ4_resetStream() instead
*/
-#define LZ4_STREAMDECODESIZE_U64 4
+LZ4LIB_API LZ4_stream_t* LZ4_initStream (void* buffer, size_t size);
+
+
+/*! LZ4_streamDecode_t :
+ * information structure to track an LZ4 stream during decompression.
+ * init this structure using LZ4_setStreamDecode() before first use.
+ * note : only use in association with static linking !
+ * this definition is not API/ABI safe,
+ * and may change in a future version !
+ */
+#define LZ4_STREAMDECODESIZE_U64 (4 + ((sizeof(void*)==16) ? 2 : 0) /*AS-400*/ )
#define LZ4_STREAMDECODESIZE (LZ4_STREAMDECODESIZE_U64 * sizeof(unsigned long long))
union LZ4_streamDecode_u {
unsigned long long table[LZ4_STREAMDECODESIZE_U64];
@@ -565,47 +660,53 @@ union LZ4_streamDecode_u {
} ; /* previously typedef'd to LZ4_streamDecode_t */
+
/*-************************************
* Obsolete Functions
**************************************/
/*! Deprecation warnings
- Should deprecation warnings be a problem,
- it is generally possible to disable them,
- typically with -Wno-deprecated-declarations for gcc
- or _CRT_SECURE_NO_WARNINGS in Visual.
- Otherwise, it's also possible to define LZ4_DISABLE_DEPRECATE_WARNINGS */
+ *
+ * Deprecated functions make the compiler generate a warning when invoked.
+ * This is meant to invite users to update their source code.
+ * Should deprecation warnings be a problem, it is generally possible to disable them,
+ * typically with -Wno-deprecated-declarations for gcc
+ * or _CRT_SECURE_NO_WARNINGS in Visual.
+ *
+ * Another method is to define LZ4_DISABLE_DEPRECATE_WARNINGS
+ * before including the header file.
+ */
#ifdef LZ4_DISABLE_DEPRECATE_WARNINGS
# define LZ4_DEPRECATED(message) /* disable deprecation warnings */
#else
-# define LZ4_GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
# if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */
# define LZ4_DEPRECATED(message) [[deprecated(message)]]
-# elif (LZ4_GCC_VERSION >= 405) || defined(__clang__)
-# define LZ4_DEPRECATED(message) __attribute__((deprecated(message)))
-# elif (LZ4_GCC_VERSION >= 301)
-# define LZ4_DEPRECATED(message) __attribute__((deprecated))
# elif defined(_MSC_VER)
# define LZ4_DEPRECATED(message) __declspec(deprecated(message))
+# elif defined(__clang__) || (defined(__GNUC__) && (__GNUC__ * 10 + __GNUC_MINOR__ >= 45))
+# define LZ4_DEPRECATED(message) __attribute__((deprecated(message)))
+# elif defined(__GNUC__) && (__GNUC__ * 10 + __GNUC_MINOR__ >= 31)
+# define LZ4_DEPRECATED(message) __attribute__((deprecated))
# else
-# pragma message("WARNING: You need to implement LZ4_DEPRECATED for this compiler")
-# define LZ4_DEPRECATED(message)
+# pragma message("WARNING: LZ4_DEPRECATED needs custom implementation for this compiler")
+# define LZ4_DEPRECATED(message) /* disabled */
# endif
#endif /* LZ4_DISABLE_DEPRECATE_WARNINGS */
-/* Obsolete compression functions */
-LZ4_DEPRECATED("use LZ4_compress_default() instead") LZ4LIB_API int LZ4_compress (const char* source, char* dest, int sourceSize);
-LZ4_DEPRECATED("use LZ4_compress_default() instead") LZ4LIB_API int LZ4_compress_limitedOutput (const char* source, char* dest, int sourceSize, int maxOutputSize);
+/*! Obsolete compression functions (since v1.7.3) */
+LZ4_DEPRECATED("use LZ4_compress_default() instead") LZ4LIB_API int LZ4_compress (const char* src, char* dest, int srcSize);
+LZ4_DEPRECATED("use LZ4_compress_default() instead") LZ4LIB_API int LZ4_compress_limitedOutput (const char* src, char* dest, int srcSize, int maxOutputSize);
LZ4_DEPRECATED("use LZ4_compress_fast_extState() instead") LZ4LIB_API int LZ4_compress_withState (void* state, const char* source, char* dest, int inputSize);
LZ4_DEPRECATED("use LZ4_compress_fast_extState() instead") LZ4LIB_API int LZ4_compress_limitedOutput_withState (void* state, const char* source, char* dest, int inputSize, int maxOutputSize);
LZ4_DEPRECATED("use LZ4_compress_fast_continue() instead") LZ4LIB_API int LZ4_compress_continue (LZ4_stream_t* LZ4_streamPtr, const char* source, char* dest, int inputSize);
LZ4_DEPRECATED("use LZ4_compress_fast_continue() instead") LZ4LIB_API int LZ4_compress_limitedOutput_continue (LZ4_stream_t* LZ4_streamPtr, const char* source, char* dest, int inputSize, int maxOutputSize);
-/* Obsolete decompression functions */
+/*! Obsolete decompression functions (since v1.8.0) */
LZ4_DEPRECATED("use LZ4_decompress_fast() instead") LZ4LIB_API int LZ4_uncompress (const char* source, char* dest, int outputSize);
LZ4_DEPRECATED("use LZ4_decompress_safe() instead") LZ4LIB_API int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize);
-/* Obsolete streaming functions; degraded functionality; do not use!
+/* Obsolete streaming functions (since v1.7.0)
+ * degraded functionality; do not use!
*
* In order to perform streaming compression, these functions depended on data
* that is no longer tracked in the state. They have been preserved as well as
@@ -616,14 +717,56 @@ LZ4_DEPRECATED("use LZ4_decompress_safe() instead") LZ4LIB_API int LZ4_uncompres
*/
LZ4_DEPRECATED("Use LZ4_createStream() instead") LZ4LIB_API void* LZ4_create (char* inputBuffer);
LZ4_DEPRECATED("Use LZ4_createStream() instead") LZ4LIB_API int LZ4_sizeofStreamState(void);
-LZ4_DEPRECATED("Use LZ4_resetStream() instead") LZ4LIB_API int LZ4_resetStreamState(void* state, char* inputBuffer);
-LZ4_DEPRECATED("Use LZ4_saveDict() instead") LZ4LIB_API char* LZ4_slideInputBuffer (void* state);
+LZ4_DEPRECATED("Use LZ4_resetStream() instead") LZ4LIB_API int LZ4_resetStreamState(void* state, char* inputBuffer);
+LZ4_DEPRECATED("Use LZ4_saveDict() instead") LZ4LIB_API char* LZ4_slideInputBuffer (void* state);
-/* Obsolete streaming decoding functions */
+/*! Obsolete streaming decoding functions (since v1.7.0) */
LZ4_DEPRECATED("use LZ4_decompress_safe_usingDict() instead") LZ4LIB_API int LZ4_decompress_safe_withPrefix64k (const char* src, char* dst, int compressedSize, int maxDstSize);
LZ4_DEPRECATED("use LZ4_decompress_fast_usingDict() instead") LZ4LIB_API int LZ4_decompress_fast_withPrefix64k (const char* src, char* dst, int originalSize);
-#endif /* LZ4_H_2983827168210 */
+/*! Obsolete LZ4_decompress_fast variants (since v1.9.0) :
+ * These functions used to be faster than LZ4_decompress_safe(),
+ * but this is no longer the case. They are now slower.
+ * This is because LZ4_decompress_fast() doesn't know the input size,
+ * and therefore must progress more cautiously into the input buffer to not read beyond the end of block.
+ * On top of that `LZ4_decompress_fast()` is not protected vs malformed or malicious inputs, making it a security liability.
+ * As a consequence, LZ4_decompress_fast() is strongly discouraged, and deprecated.
+ *
+ * The last remaining LZ4_decompress_fast() specificity is that
+ * it can decompress a block without knowing its compressed size.
+ * Such functionality can be achieved in a more secure manner
+ * by employing LZ4_decompress_safe_partial().
+ *
+ * Parameters:
+ * originalSize : is the uncompressed size to regenerate.
+ * `dst` must be already allocated, its size must be >= 'originalSize' bytes.
+ * @return : number of bytes read from source buffer (== compressed size).
+ * The function expects to finish at block's end exactly.
+ * If the source stream is detected malformed, the function stops decoding and returns a negative result.
+ * note : LZ4_decompress_fast*() requires originalSize. Thanks to this information, it never writes past the output buffer.
+ * However, since it doesn't know its 'src' size, it may read an unknown amount of input, past input buffer bounds.
+ * Also, since match offsets are not validated, match reads from 'src' may underflow too.
+ * These issues never happen if input (compressed) data is correct.
+ * But they may happen if input data is invalid (error or intentional tampering).
+ * As a consequence, use these functions in trusted environments with trusted data **only**.
+ */
+LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe() instead")
+LZ4LIB_API int LZ4_decompress_fast (const char* src, char* dst, int originalSize);
+LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_continue() instead")
+LZ4LIB_API int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* src, char* dst, int originalSize);
+LZ4_DEPRECATED("This function is deprecated and unsafe. Consider using LZ4_decompress_safe_usingDict() instead")
+LZ4LIB_API int LZ4_decompress_fast_usingDict (const char* src, char* dst, int originalSize, const char* dictStart, int dictSize);
+
+/*! LZ4_resetStream() :
+ * An LZ4_stream_t structure must be initialized at least once.
+ * This is done with LZ4_initStream(), or LZ4_resetStream().
+ * Consider switching to LZ4_initStream(),
+ * invoking LZ4_resetStream() will trigger deprecation warnings in the future.
+ */
+LZ4LIB_API void LZ4_resetStream (LZ4_stream_t* streamPtr);
+
+
+#endif /* LZ4_H_98237428734687 */
#if defined (__cplusplus)
diff --git a/lib/lz4frame.c b/lib/lz4frame.c
index 08bf0fa..ec02c92 100644
--- a/lib/lz4frame.c
+++ b/lib/lz4frame.c
@@ -1,41 +1,44 @@
/*
-LZ4 auto-framing library
-Copyright (C) 2011-2016, Yann Collet.
-
-BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
-
-Redistribution and use in source and binary forms, with or without
-modification, are permitted provided that the following conditions are
-met:
-
-* Redistributions of source code must retain the above copyright
-notice, this list of conditions and the following disclaimer.
-* Redistributions in binary form must reproduce the above
-copyright notice, this list of conditions and the following disclaimer
-in the documentation and/or other materials provided with the
-distribution.
-
-THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
-A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
-OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
-LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
-DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
-THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-You can contact the author at :
-- LZ4 homepage : http://www.lz4.org
-- LZ4 source repository : https://github.com/lz4/lz4
-*/
+ * LZ4 auto-framing library
+ * Copyright (C) 2011-2016, Yann Collet.
+ *
+ * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are
+ * met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following disclaimer
+ * in the documentation and/or other materials provided with the
+ * distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * You can contact the author at :
+ * - LZ4 homepage : http://www.lz4.org
+ * - LZ4 source repository : https://github.com/lz4/lz4
+ */
/* LZ4F is a stand-alone API to create LZ4-compressed Frames
-* in full conformance with specification v1.5.0
-* All related operations, including memory management, are handled by the library.
-* */
+ * in full conformance with specification v1.6.1 .
+ * This library rely upon memory management capabilities (malloc, free)
+ * provided either by <stdlib.h>,
+ * or redirected towards another library of user's choice
+ * (see Memory Routines below).
+ */
/*-************************************
@@ -62,16 +65,27 @@ You can contact the author at :
/*-************************************
* Memory routines
**************************************/
-#include <stdlib.h> /* malloc, calloc, free */
-#define ALLOC(s) malloc(s)
-#define ALLOC_AND_ZERO(s) calloc(1,s)
-#define FREEMEM free
+/*
+ * User may redirect invocations of
+ * malloc(), calloc() and free()
+ * towards another library or solution of their choice
+ * by modifying below section.
+ */
+#ifndef LZ4_SRC_INCLUDED /* avoid redefinition when sources are coalesced */
+# include <stdlib.h> /* malloc, calloc, free */
+# define ALLOC(s) malloc(s)
+# define ALLOC_AND_ZERO(s) calloc(1,(s))
+# define FREEMEM(p) free(p)
+#endif
+
#include <string.h> /* memset, memcpy, memmove */
-#define MEM_INIT memset
+#ifndef LZ4_SRC_INCLUDED /* avoid redefinition when sources are coalesced */
+# define MEM_INIT(p,v,s) memset((p),(v),(s))
+#endif
/*-************************************
-* Includes
+* Library declarations
**************************************/
#define LZ4F_STATIC_LINKING_ONLY
#include "lz4frame.h"
@@ -134,8 +148,8 @@ static U32 LZ4F_readLE32 (const void* src)
{
const BYTE* const srcPtr = (const BYTE*)src;
U32 value32 = srcPtr[0];
- value32 += (srcPtr[1]<<8);
- value32 += (srcPtr[2]<<16);
+ value32 += ((U32)srcPtr[1])<< 8;
+ value32 += ((U32)srcPtr[2])<<16;
value32 += ((U32)srcPtr[3])<<24;
return value32;
}
@@ -180,9 +194,11 @@ static void LZ4F_writeLE64 (void* dst, U64 value64)
/*-************************************
* Constants
**************************************/
-#define KB *(1<<10)
-#define MB *(1<<20)
-#define GB *(1<<30)
+#ifndef LZ4_SRC_INCLUDED /* avoid double definition */
+# define KB *(1<<10)
+# define MB *(1<<20)
+# define GB *(1<<30)
+#endif
#define _1BIT 0x01
#define _2BITS 0x03
@@ -195,9 +211,10 @@ static void LZ4F_writeLE64 (void* dst, U64 value64)
#define LZ4F_BLOCKUNCOMPRESSED_FLAG 0x80000000U
#define LZ4F_BLOCKSIZEID_DEFAULT LZ4F_max64KB
-static const size_t minFHSize = 7;
+static const size_t minFHSize = LZ4F_HEADER_SIZE_MIN; /* 7 */
static const size_t maxFHSize = LZ4F_HEADER_SIZE_MAX; /* 19 */
-static const size_t BHSize = 4;
+static const size_t BHSize = LZ4F_BLOCK_HEADER_SIZE; /* block header : size, and compress flag */
+static const size_t BFSize = LZ4F_BLOCK_CHECKSUM_SIZE; /* block footer : checksum (optional) */
/*-************************************
@@ -258,22 +275,22 @@ unsigned LZ4F_getVersion(void) { return LZ4F_VERSION; }
int LZ4F_compressionLevel_max(void) { return LZ4HC_CLEVEL_MAX; }
-
-/*-************************************
-* Private functions
-**************************************/
-#define MIN(a,b) ( (a) < (b) ? (a) : (b) )
-
-static size_t LZ4F_getBlockSize(unsigned blockSizeID)
+size_t LZ4F_getBlockSize(unsigned blockSizeID)
{
static const size_t blockSizes[4] = { 64 KB, 256 KB, 1 MB, 4 MB };
if (blockSizeID == 0) blockSizeID = LZ4F_BLOCKSIZEID_DEFAULT;
- blockSizeID -= 4;
- if (blockSizeID > 3) return err0r(LZ4F_ERROR_maxBlockSize_invalid);
+ if (blockSizeID < LZ4F_max64KB || blockSizeID > LZ4F_max4MB)
+ return err0r(LZ4F_ERROR_maxBlockSize_invalid);
+ blockSizeID -= LZ4F_max64KB;
return blockSizes[blockSizeID];
}
+/*-************************************
+* Private functions
+**************************************/
+#define MIN(a,b) ( (a) < (b) ? (a) : (b) )
+
static BYTE LZ4F_headerChecksum (const void* header, size_t length)
{
U32 const xxh = XXH32(header, length, 0);
@@ -308,9 +325,9 @@ static size_t LZ4F_compressBound_internal(size_t srcSize,
const LZ4F_preferences_t* preferencesPtr,
size_t alreadyBuffered)
{
- LZ4F_preferences_t prefsNull;
- MEM_INIT(&prefsNull, 0, sizeof(prefsNull));
+ LZ4F_preferences_t prefsNull = LZ4F_INIT_PREFERENCES;
prefsNull.frameInfo.contentChecksumFlag = LZ4F_contentChecksumEnabled; /* worst case */
+ prefsNull.frameInfo.blockChecksumFlag = LZ4F_blockChecksumEnabled; /* worst case */
{ const LZ4F_preferences_t* const prefsPtr = (preferencesPtr==NULL) ? &prefsNull : preferencesPtr;
U32 const flush = prefsPtr->autoFlush | (srcSize==0);
LZ4F_blockSizeID_t const blockID = prefsPtr->frameInfo.blockSizeID;
@@ -323,11 +340,10 @@ static size_t LZ4F_compressBound_internal(size_t srcSize,
size_t const lastBlockSize = flush ? partialBlockSize : 0;
unsigned const nbBlocks = nbFullBlocks + (lastBlockSize>0);
- size_t const blockHeaderSize = 4;
- size_t const blockCRCSize = 4 * prefsPtr->frameInfo.blockChecksumFlag;
- size_t const frameEnd = 4 + (prefsPtr->frameInfo.contentChecksumFlag*4);
+ size_t const blockCRCSize = BFSize * prefsPtr->frameInfo.blockChecksumFlag;
+ size_t const frameEnd = BHSize + (prefsPtr->frameInfo.contentChecksumFlag*BFSize);
- return ((blockHeaderSize + blockCRCSize) * nbBlocks) +
+ return ((BHSize + blockCRCSize) * nbBlocks) +
(blockSize * nbFullBlocks) + lastBlockSize + frameEnd;
}
}
@@ -388,15 +404,18 @@ size_t LZ4F_compressFrame_usingCDict(LZ4F_cctx* cctx,
if (LZ4F_isError(headerSize)) return headerSize;
dstPtr += headerSize; /* header size */ }
- { size_t const cSize = LZ4F_compressUpdate(cctx, dstPtr, dstEnd-dstPtr, srcBuffer, srcSize, &options);
+ assert(dstEnd >= dstPtr);
+ { size_t const cSize = LZ4F_compressUpdate(cctx, dstPtr, (size_t)(dstEnd-dstPtr), srcBuffer, srcSize, &options);
if (LZ4F_isError(cSize)) return cSize;
dstPtr += cSize; }
- { size_t const tailSize = LZ4F_compressEnd(cctx, dstPtr, dstEnd-dstPtr, &options); /* flush last block, and generate suffix */
+ assert(dstEnd >= dstPtr);
+ { size_t const tailSize = LZ4F_compressEnd(cctx, dstPtr, (size_t)(dstEnd-dstPtr), &options); /* flush last block, and generate suffix */
if (LZ4F_isError(tailSize)) return tailSize;
dstPtr += tailSize; }
- return (dstPtr - dstStart);
+ assert(dstEnd >= dstStart);
+ return (size_t)(dstPtr - dstStart);
}
@@ -428,7 +447,7 @@ size_t LZ4F_compressFrame(void* dstBuffer, size_t dstCapacity,
if (preferencesPtr == NULL ||
preferencesPtr->compressionLevel < LZ4HC_CLEVEL_MIN)
{
- LZ4_resetStream(&lz4ctx);
+ LZ4_initStream(&lz4ctx, sizeof(lz4ctx));
cctxPtr->lz4CtxPtr = &lz4ctx;
cctxPtr->lz4CtxAlloc = 1;
cctxPtr->lz4CtxState = 1;
@@ -514,7 +533,7 @@ void LZ4F_freeCDict(LZ4F_CDict* cdict)
* If the result LZ4F_errorCode_t is not OK_NoError, there was an error during context creation.
* Object can release its memory using LZ4F_freeCompressionContext();
*/
-LZ4F_errorCode_t LZ4F_createCompressionContext(LZ4F_compressionContext_t* LZ4F_compressionContextPtr, unsigned version)
+LZ4F_errorCode_t LZ4F_createCompressionContext(LZ4F_cctx** LZ4F_compressionContextPtr, unsigned version)
{
LZ4F_cctx_t* const cctxPtr = (LZ4F_cctx_t*)ALLOC_AND_ZERO(sizeof(LZ4F_cctx_t));
if (cctxPtr==NULL) return err0r(LZ4F_ERROR_allocation_failed);
@@ -522,20 +541,18 @@ LZ4F_errorCode_t LZ4F_createCompressionContext(LZ4F_compressionContext_t* LZ4F_c
cctxPtr->version = version;
cctxPtr->cStage = 0; /* Next stage : init stream */
- *LZ4F_compressionContextPtr = (LZ4F_compressionContext_t)cctxPtr;
+ *LZ4F_compressionContextPtr = cctxPtr;
return LZ4F_OK_NoError;
}
-LZ4F_errorCode_t LZ4F_freeCompressionContext(LZ4F_compressionContext_t LZ4F_compressionContext)
+LZ4F_errorCode_t LZ4F_freeCompressionContext(LZ4F_cctx* cctxPtr)
{
- LZ4F_cctx_t* const cctxPtr = (LZ4F_cctx_t*)LZ4F_compressionContext;
-
if (cctxPtr != NULL) { /* support free on NULL */
- FREEMEM(cctxPtr->lz4CtxPtr); /* works because LZ4_streamHC_t and LZ4_stream_t are simple POD types */
+ FREEMEM(cctxPtr->lz4CtxPtr); /* note: LZ4_streamHC_t and LZ4_stream_t are simple POD types */
FREEMEM(cctxPtr->tmpBuff);
- FREEMEM(LZ4F_compressionContext);
+ FREEMEM(cctxPtr);
}
return LZ4F_OK_NoError;
@@ -598,20 +615,22 @@ size_t LZ4F_compressBegin_usingCDict(LZ4F_cctx* cctxPtr,
if (cctxPtr->lz4CtxAlloc < ctxTypeID) {
FREEMEM(cctxPtr->lz4CtxPtr);
if (cctxPtr->prefs.compressionLevel < LZ4HC_CLEVEL_MIN) {
- cctxPtr->lz4CtxPtr = (void*)LZ4_createStream();
+ cctxPtr->lz4CtxPtr = LZ4_createStream();
} else {
- cctxPtr->lz4CtxPtr = (void*)LZ4_createStreamHC();
+ cctxPtr->lz4CtxPtr = LZ4_createStreamHC();
}
- if (cctxPtr->lz4CtxPtr == NULL) return err0r(LZ4F_ERROR_allocation_failed);
+ if (cctxPtr->lz4CtxPtr == NULL)
+ return err0r(LZ4F_ERROR_allocation_failed);
cctxPtr->lz4CtxAlloc = ctxTypeID;
cctxPtr->lz4CtxState = ctxTypeID;
} else if (cctxPtr->lz4CtxState != ctxTypeID) {
/* otherwise, a sufficient buffer is allocated, but we need to
* reset it to the correct context type */
if (cctxPtr->prefs.compressionLevel < LZ4HC_CLEVEL_MIN) {
- LZ4_resetStream((LZ4_stream_t *) cctxPtr->lz4CtxPtr);
+ LZ4_initStream((LZ4_stream_t *) cctxPtr->lz4CtxPtr, sizeof (LZ4_stream_t));
} else {
- LZ4_resetStreamHC((LZ4_streamHC_t *) cctxPtr->lz4CtxPtr, cctxPtr->prefs.compressionLevel);
+ LZ4_initStreamHC((LZ4_streamHC_t *) cctxPtr->lz4CtxPtr, sizeof(LZ4_streamHC_t));
+ LZ4_setCompressionLevel((LZ4_streamHC_t *) cctxPtr->lz4CtxPtr, cctxPtr->prefs.compressionLevel);
}
cctxPtr->lz4CtxState = ctxTypeID;
}
@@ -623,8 +642,8 @@ size_t LZ4F_compressBegin_usingCDict(LZ4F_cctx* cctxPtr,
cctxPtr->maxBlockSize = LZ4F_getBlockSize(cctxPtr->prefs.frameInfo.blockSizeID);
{ size_t const requiredBuffSize = preferencesPtr->autoFlush ?
- (cctxPtr->prefs.frameInfo.blockMode == LZ4F_blockLinked) * 64 KB : /* only needs windows size */
- cctxPtr->maxBlockSize + ((cctxPtr->prefs.frameInfo.blockMode == LZ4F_blockLinked) * 128 KB);
+ ((cctxPtr->prefs.frameInfo.blockMode == LZ4F_blockLinked) ? 64 KB : 0) : /* only needs past data up to window size */
+ cctxPtr->maxBlockSize + ((cctxPtr->prefs.frameInfo.blockMode == LZ4F_blockLinked) ? 128 KB : 0);
if (cctxPtr->maxBufferSize < requiredBuffSize) {
cctxPtr->maxBufferSize = 0;
@@ -635,7 +654,7 @@ size_t LZ4F_compressBegin_usingCDict(LZ4F_cctx* cctxPtr,
} }
cctxPtr->tmpIn = cctxPtr->tmpBuff;
cctxPtr->tmpInSize = 0;
- XXH32_reset(&(cctxPtr->xxh), 0);
+ (void)XXH32_reset(&(cctxPtr->xxh), 0);
/* context init */
cctxPtr->cdict = cdict;
@@ -644,7 +663,7 @@ size_t LZ4F_compressBegin_usingCDict(LZ4F_cctx* cctxPtr,
LZ4F_initStream(cctxPtr->lz4CtxPtr, cdict, cctxPtr->prefs.compressionLevel, LZ4F_blockLinked);
}
if (preferencesPtr->compressionLevel >= LZ4HC_CLEVEL_MIN) {
- LZ4_favorDecompressionSpeed((LZ4_streamHC_t*)cctxPtr->lz4CtxPtr, (int)preferencesPtr->favorDecSpeed);
+ LZ4_favorDecompressionSpeed((LZ4_streamHC_t*)cctxPtr->lz4CtxPtr, (int)preferencesPtr->favorDecSpeed);
}
/* Magic Number */
@@ -656,7 +675,7 @@ size_t LZ4F_compressBegin_usingCDict(LZ4F_cctx* cctxPtr,
*dstPtr++ = (BYTE)(((1 & _2BITS) << 6) /* Version('01') */
+ ((cctxPtr->prefs.frameInfo.blockMode & _1BIT ) << 5)
+ ((cctxPtr->prefs.frameInfo.blockChecksumFlag & _1BIT ) << 4)
- + ((cctxPtr->prefs.frameInfo.contentSize > 0) << 3)
+ + ((unsigned)(cctxPtr->prefs.frameInfo.contentSize > 0) << 3)
+ ((cctxPtr->prefs.frameInfo.contentChecksumFlag & _1BIT ) << 2)
+ (cctxPtr->prefs.frameInfo.dictID > 0) );
/* BD Byte */
@@ -673,11 +692,11 @@ size_t LZ4F_compressBegin_usingCDict(LZ4F_cctx* cctxPtr,
dstPtr += 4;
}
/* Header CRC Byte */
- *dstPtr = LZ4F_headerChecksum(headerStart, dstPtr - headerStart);
+ *dstPtr = LZ4F_headerChecksum(headerStart, (size_t)(dstPtr - headerStart));
dstPtr++;
cctxPtr->cStage = 1; /* header written, now request input data block */
- return (dstPtr - dstStart);
+ return (size_t)(dstPtr - dstStart);
}
@@ -686,7 +705,7 @@ size_t LZ4F_compressBegin_usingCDict(LZ4F_cctx* cctxPtr,
* dstBuffer must be >= LZ4F_HEADER_SIZE_MAX bytes.
* preferencesPtr can be NULL, in which case default parameters are selected.
* @return : number of bytes written into dstBuffer for the header
- * or an error code (can be tested using LZ4F_isError())
+ * or an error code (can be tested using LZ4F_isError())
*/
size_t LZ4F_compressBegin(LZ4F_cctx* cctxPtr,
void* dstBuffer, size_t dstCapacity,
@@ -704,6 +723,9 @@ size_t LZ4F_compressBegin(LZ4F_cctx* cctxPtr,
*/
size_t LZ4F_compressBound(size_t srcSize, const LZ4F_preferences_t* preferencesPtr)
{
+ if (preferencesPtr && preferencesPtr->autoFlush) {
+ return LZ4F_compressBound_internal(srcSize, preferencesPtr, 0);
+ }
return LZ4F_compressBound_internal(srcSize, preferencesPtr, (size_t)-1);
}
@@ -712,27 +734,32 @@ typedef int (*compressFunc_t)(void* ctx, const char* src, char* dst, int srcSize
/*! LZ4F_makeBlock():
- * compress a single block, add header and checksum
- * assumption : dst buffer capacity is >= srcSize */
-static size_t LZ4F_makeBlock(void* dst, const void* src, size_t srcSize,
+ * compress a single block, add header and optional checksum.
+ * assumption : dst buffer capacity is >= BHSize + srcSize + crcSize
+ */
+static size_t LZ4F_makeBlock(void* dst,
+ const void* src, size_t srcSize,
compressFunc_t compress, void* lz4ctx, int level,
- const LZ4F_CDict* cdict, LZ4F_blockChecksum_t crcFlag)
+ const LZ4F_CDict* cdict,
+ LZ4F_blockChecksum_t crcFlag)
{
BYTE* const cSizePtr = (BYTE*)dst;
- U32 cSize = (U32)compress(lz4ctx, (const char*)src, (char*)(cSizePtr+4),
+ U32 cSize = (U32)compress(lz4ctx, (const char*)src, (char*)(cSizePtr+BHSize),
(int)(srcSize), (int)(srcSize-1),
level, cdict);
- LZ4F_writeLE32(cSizePtr, cSize);
if (cSize == 0) { /* compression failed */
+ DEBUGLOG(5, "LZ4F_makeBlock: compression failed, creating a raw block (size %u)", (U32)srcSize);
cSize = (U32)srcSize;
LZ4F_writeLE32(cSizePtr, cSize | LZ4F_BLOCKUNCOMPRESSED_FLAG);
- memcpy(cSizePtr+4, src, srcSize);
+ memcpy(cSizePtr+BHSize, src, srcSize);
+ } else {
+ LZ4F_writeLE32(cSizePtr, cSize);
}
if (crcFlag) {
- U32 const crc32 = XXH32(cSizePtr+4, cSize, 0); /* checksum of compressed data */
- LZ4F_writeLE32(cSizePtr+4+cSize, crc32);
+ U32 const crc32 = XXH32(cSizePtr+BHSize, cSize, 0); /* checksum of compressed data */
+ LZ4F_writeLE32(cSizePtr+BHSize+cSize, crc32);
}
- return 4 + cSize + ((U32)crcFlag)*4;
+ return BHSize + cSize + ((U32)crcFlag)*BFSize;
}
@@ -832,9 +859,11 @@ size_t LZ4F_compressUpdate(LZ4F_cctx* cctxPtr,
memcpy(cctxPtr->tmpIn + cctxPtr->tmpInSize, srcBuffer, sizeToCopy);
srcPtr += sizeToCopy;
- dstPtr += LZ4F_makeBlock(dstPtr, cctxPtr->tmpIn, blockSize,
+ dstPtr += LZ4F_makeBlock(dstPtr,
+ cctxPtr->tmpIn, blockSize,
compress, cctxPtr->lz4CtxPtr, cctxPtr->prefs.compressionLevel,
- cctxPtr->cdict, cctxPtr->prefs.frameInfo.blockChecksumFlag);
+ cctxPtr->cdict,
+ cctxPtr->prefs.frameInfo.blockChecksumFlag);
if (cctxPtr->prefs.frameInfo.blockMode==LZ4F_blockLinked) cctxPtr->tmpIn += blockSize;
cctxPtr->tmpInSize = 0;
@@ -844,18 +873,22 @@ size_t LZ4F_compressUpdate(LZ4F_cctx* cctxPtr,
while ((size_t)(srcEnd - srcPtr) >= blockSize) {
/* compress full blocks */
lastBlockCompressed = fromSrcBuffer;
- dstPtr += LZ4F_makeBlock(dstPtr, srcPtr, blockSize,
+ dstPtr += LZ4F_makeBlock(dstPtr,
+ srcPtr, blockSize,
compress, cctxPtr->lz4CtxPtr, cctxPtr->prefs.compressionLevel,
- cctxPtr->cdict, cctxPtr->prefs.frameInfo.blockChecksumFlag);
+ cctxPtr->cdict,
+ cctxPtr->prefs.frameInfo.blockChecksumFlag);
srcPtr += blockSize;
}
if ((cctxPtr->prefs.autoFlush) && (srcPtr < srcEnd)) {
/* compress remaining input < blockSize */
lastBlockCompressed = fromSrcBuffer;
- dstPtr += LZ4F_makeBlock(dstPtr, srcPtr, srcEnd - srcPtr,
+ dstPtr += LZ4F_makeBlock(dstPtr,
+ srcPtr, (size_t)(srcEnd - srcPtr),
compress, cctxPtr->lz4CtxPtr, cctxPtr->prefs.compressionLevel,
- cctxPtr->cdict, cctxPtr->prefs.frameInfo.blockChecksumFlag);
+ cctxPtr->cdict,
+ cctxPtr->prefs.frameInfo.blockChecksumFlag);
srcPtr = srcEnd;
}
@@ -881,28 +914,30 @@ size_t LZ4F_compressUpdate(LZ4F_cctx* cctxPtr,
/* some input data left, necessarily < blockSize */
if (srcPtr < srcEnd) {
/* fill tmp buffer */
- size_t const sizeToCopy = srcEnd - srcPtr;
+ size_t const sizeToCopy = (size_t)(srcEnd - srcPtr);
memcpy(cctxPtr->tmpIn, srcPtr, sizeToCopy);
cctxPtr->tmpInSize = sizeToCopy;
}
if (cctxPtr->prefs.frameInfo.contentChecksumFlag == LZ4F_contentChecksumEnabled)
- XXH32_update(&(cctxPtr->xxh), srcBuffer, srcSize);
+ (void)XXH32_update(&(cctxPtr->xxh), srcBuffer, srcSize);
cctxPtr->totalInSize += srcSize;
- return dstPtr - dstStart;
+ return (size_t)(dstPtr - dstStart);
}
/*! LZ4F_flush() :
- * Should you need to create compressed data immediately, without waiting for a block to be filled,
- * you can call LZ4_flush(), which will immediately compress any remaining data stored within compressionContext.
- * The result of the function is the number of bytes written into dstBuffer
- * (it can be zero, this means there was no data left within compressionContext)
+ * When compressed data must be sent immediately, without waiting for a block to be filled,
+ * invoke LZ4_flush(), which will immediately compress any remaining data stored within LZ4F_cctx.
+ * The result of the function is the number of bytes written into dstBuffer.
+ * It can be zero, this means there was no data left within LZ4F_cctx.
* The function outputs an error code if it fails (can be tested using LZ4F_isError())
- * The LZ4F_compressOptions_t structure is optional : you can provide NULL as argument.
+ * LZ4F_compressOptions_t* is optional. NULL is a valid argument.
*/
-size_t LZ4F_flush(LZ4F_cctx* cctxPtr, void* dstBuffer, size_t dstCapacity, const LZ4F_compressOptions_t* compressOptionsPtr)
+size_t LZ4F_flush(LZ4F_cctx* cctxPtr,
+ void* dstBuffer, size_t dstCapacity,
+ const LZ4F_compressOptions_t* compressOptionsPtr)
{
BYTE* const dstStart = (BYTE*)dstBuffer;
BYTE* dstPtr = dstStart;
@@ -910,52 +945,67 @@ size_t LZ4F_flush(LZ4F_cctx* cctxPtr, void* dstBuffer, size_t dstCapacity, const
if (cctxPtr->tmpInSize == 0) return 0; /* nothing to flush */
if (cctxPtr->cStage != 1) return err0r(LZ4F_ERROR_GENERIC);
- if (dstCapacity < (cctxPtr->tmpInSize + 4)) return err0r(LZ4F_ERROR_dstMaxSize_tooSmall); /* +4 : block header(4) */
+ if (dstCapacity < (cctxPtr->tmpInSize + BHSize + BFSize))
+ return err0r(LZ4F_ERROR_dstMaxSize_tooSmall);
(void)compressOptionsPtr; /* not yet useful */
/* select compression function */
compress = LZ4F_selectCompression(cctxPtr->prefs.frameInfo.blockMode, cctxPtr->prefs.compressionLevel);
/* compress tmp buffer */
- dstPtr += LZ4F_makeBlock(dstPtr, cctxPtr->tmpIn, cctxPtr->tmpInSize,
+ dstPtr += LZ4F_makeBlock(dstPtr,
+ cctxPtr->tmpIn, cctxPtr->tmpInSize,
compress, cctxPtr->lz4CtxPtr, cctxPtr->prefs.compressionLevel,
- cctxPtr->cdict, cctxPtr->prefs.frameInfo.blockChecksumFlag);
- if (cctxPtr->prefs.frameInfo.blockMode==LZ4F_blockLinked) cctxPtr->tmpIn += cctxPtr->tmpInSize;
+ cctxPtr->cdict,
+ cctxPtr->prefs.frameInfo.blockChecksumFlag);
+ assert(((void)"flush overflows dstBuffer!", (size_t)(dstPtr - dstStart) <= dstCapacity));
+
+ if (cctxPtr->prefs.frameInfo.blockMode == LZ4F_blockLinked)
+ cctxPtr->tmpIn += cctxPtr->tmpInSize;
cctxPtr->tmpInSize = 0;
/* keep tmpIn within limits */
if ((cctxPtr->tmpIn + cctxPtr->maxBlockSize) > (cctxPtr->tmpBuff + cctxPtr->maxBufferSize)) { /* necessarily LZ4F_blockLinked */
- int realDictSize = LZ4F_localSaveDict(cctxPtr);
+ int const realDictSize = LZ4F_localSaveDict(cctxPtr);
cctxPtr->tmpIn = cctxPtr->tmpBuff + realDictSize;
}
- return dstPtr - dstStart;
+ return (size_t)(dstPtr - dstStart);
}
/*! LZ4F_compressEnd() :
- * When you want to properly finish the compressed frame, just call LZ4F_compressEnd().
- * It will flush whatever data remained within compressionContext (like LZ4_flush())
- * but also properly finalize the frame, with an endMark and a checksum.
- * The result of the function is the number of bytes written into dstBuffer (necessarily >= 4 (endMark size))
- * The function outputs an error code if it fails (can be tested using LZ4F_isError())
- * The LZ4F_compressOptions_t structure is optional : you can provide NULL as argument.
- * compressionContext can then be used again, starting with LZ4F_compressBegin(). The preferences will remain the same.
+ * When you want to properly finish the compressed frame, just call LZ4F_compressEnd().
+ * It will flush whatever data remained within compressionContext (like LZ4_flush())
+ * but also properly finalize the frame, with an endMark and an (optional) checksum.
+ * LZ4F_compressOptions_t structure is optional : you can provide NULL as argument.
+ * @return: the number of bytes written into dstBuffer (necessarily >= 4 (endMark size))
+ * or an error code if it fails (can be tested using LZ4F_isError())
+ * The context can then be used again to compress a new frame, starting with LZ4F_compressBegin().
*/
-size_t LZ4F_compressEnd(LZ4F_cctx* cctxPtr, void* dstBuffer, size_t dstMaxSize, const LZ4F_compressOptions_t* compressOptionsPtr)
+size_t LZ4F_compressEnd(LZ4F_cctx* cctxPtr,
+ void* dstBuffer, size_t dstCapacity,
+ const LZ4F_compressOptions_t* compressOptionsPtr)
{
BYTE* const dstStart = (BYTE*)dstBuffer;
BYTE* dstPtr = dstStart;
- size_t const flushSize = LZ4F_flush(cctxPtr, dstBuffer, dstMaxSize, compressOptionsPtr);
+ size_t const flushSize = LZ4F_flush(cctxPtr, dstBuffer, dstCapacity, compressOptionsPtr);
+ DEBUGLOG(5,"LZ4F_compressEnd: dstCapacity=%u", (unsigned)dstCapacity);
if (LZ4F_isError(flushSize)) return flushSize;
dstPtr += flushSize;
+ assert(flushSize <= dstCapacity);
+ dstCapacity -= flushSize;
+
+ if (dstCapacity < 4) return err0r(LZ4F_ERROR_dstMaxSize_tooSmall);
LZ4F_writeLE32(dstPtr, 0);
- dstPtr+=4; /* endMark */
+ dstPtr += 4; /* endMark */
if (cctxPtr->prefs.frameInfo.contentChecksumFlag == LZ4F_contentChecksumEnabled) {
U32 const xxh = XXH32_digest(&(cctxPtr->xxh));
+ if (dstCapacity < 8) return err0r(LZ4F_ERROR_dstMaxSize_tooSmall);
+ DEBUGLOG(5,"Writing 32-bit content checksum");
LZ4F_writeLE32(dstPtr, xxh);
dstPtr+=4; /* content Checksum */
}
@@ -968,7 +1018,7 @@ size_t LZ4F_compressEnd(LZ4F_cctx* cctxPtr, void* dstBuffer, size_t dstMaxSize,
return err0r(LZ4F_ERROR_frameSize_wrong);
}
- return dstPtr - dstStart;
+ return (size_t)(dstPtr - dstStart);
}
@@ -1019,7 +1069,10 @@ struct LZ4F_dctx_s {
LZ4F_errorCode_t LZ4F_createDecompressionContext(LZ4F_dctx** LZ4F_decompressionContextPtr, unsigned versionNumber)
{
LZ4F_dctx* const dctx = (LZ4F_dctx*)ALLOC_AND_ZERO(sizeof(LZ4F_dctx));
- if (dctx==NULL) return err0r(LZ4F_ERROR_GENERIC);
+ if (dctx == NULL) { /* failed allocation */
+ *LZ4F_decompressionContextPtr = NULL;
+ return err0r(LZ4F_ERROR_allocation_failed);
+ }
dctx->version = versionNumber;
*LZ4F_decompressionContextPtr = dctx;
@@ -1049,31 +1102,6 @@ void LZ4F_resetDecompressionContext(LZ4F_dctx* dctx)
}
-/*! LZ4F_headerSize() :
- * @return : size of frame header
- * or an error code, which can be tested using LZ4F_isError()
- */
-static size_t LZ4F_headerSize(const void* src, size_t srcSize)
-{
- /* minimal srcSize to determine header size */
- if (srcSize < 5) return err0r(LZ4F_ERROR_frameHeader_incomplete);
-
- /* special case : skippable frames */
- if ((LZ4F_readLE32(src) & 0xFFFFFFF0U) == LZ4F_MAGIC_SKIPPABLE_START) return 8;
-
- /* control magic number */
- if (LZ4F_readLE32(src) != LZ4F_MAGICNUMBER)
- return err0r(LZ4F_ERROR_frameType_unknown);
-
- /* Frame Header Size */
- { BYTE const FLG = ((const BYTE*)src)[4];
- U32 const contentSizeFlag = (FLG>>3) & _1BIT;
- U32 const dictIDFlag = FLG & _1BIT;
- return minFHSize + (contentSizeFlag*8) + (dictIDFlag*4);
- }
-}
-
-
/*! LZ4F_decodeHeader() :
* input : `src` points at the **beginning of the frame**
* output : set internal values of dctx, such as
@@ -1088,6 +1116,7 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
size_t frameHeaderSize;
const BYTE* srcPtr = (const BYTE*)src;
+ DEBUGLOG(5, "LZ4F_decodeHeader");
/* need to decode header to get frameInfo */
if (srcSize < minFHSize) return err0r(LZ4F_ERROR_frameHeader_incomplete); /* minimal frame header size */
MEM_INIT(&(dctx->frameInfo), 0, sizeof(dctx->frameInfo));
@@ -1107,8 +1136,12 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
}
/* control magic number */
- if (LZ4F_readLE32(srcPtr) != LZ4F_MAGICNUMBER)
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
+ if (LZ4F_readLE32(srcPtr) != LZ4F_MAGICNUMBER) {
+ DEBUGLOG(4, "frame header error : unknown magic number");
return err0r(LZ4F_ERROR_frameType_unknown);
+ }
+#endif
dctx->frameInfo.frameType = LZ4F_frame;
/* Flags */
@@ -1125,7 +1158,7 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
}
/* Frame Header Size */
- frameHeaderSize = minFHSize + (contentSizeFlag*8) + (dictIDFlag*4);
+ frameHeaderSize = minFHSize + (contentSizeFlag?8:0) + (dictIDFlag?4:0);
if (srcSize < frameHeaderSize) {
/* not enough input to fully decode frame header */
@@ -1146,10 +1179,13 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
}
/* check header */
+ assert(frameHeaderSize > 5);
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
{ BYTE const HC = LZ4F_headerChecksum(srcPtr+4, frameHeaderSize-5);
if (HC != srcPtr[frameHeaderSize-1])
return err0r(LZ4F_ERROR_headerChecksum_invalid);
}
+#endif
/* save */
dctx->frameInfo.blockMode = (LZ4F_blockMode_t)blockMode;
@@ -1169,6 +1205,36 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
}
+/*! LZ4F_headerSize() :
+ * @return : size of frame header
+ * or an error code, which can be tested using LZ4F_isError()
+ */
+size_t LZ4F_headerSize(const void* src, size_t srcSize)
+{
+ if (src == NULL) return err0r(LZ4F_ERROR_srcPtr_wrong);
+
+ /* minimal srcSize to determine header size */
+ if (srcSize < LZ4F_MIN_SIZE_TO_KNOW_HEADER_LENGTH)
+ return err0r(LZ4F_ERROR_frameHeader_incomplete);
+
+ /* special case : skippable frames */
+ if ((LZ4F_readLE32(src) & 0xFFFFFFF0U) == LZ4F_MAGIC_SKIPPABLE_START)
+ return 8;
+
+ /* control magic number */
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
+ if (LZ4F_readLE32(src) != LZ4F_MAGICNUMBER)
+ return err0r(LZ4F_ERROR_frameType_unknown);
+#endif
+
+ /* Frame Header Size */
+ { BYTE const FLG = ((const BYTE*)src)[4];
+ U32 const contentSizeFlag = (FLG>>3) & _1BIT;
+ U32 const dictIDFlag = FLG & _1BIT;
+ return minFHSize + (contentSizeFlag?8:0) + (dictIDFlag?4:0);
+ }
+}
+
/*! LZ4F_getFrameInfo() :
* This function extracts frame parameters (max blockSize, frame checksum, etc.).
* Usage is optional. Objective is to provide relevant information for allocation purposes.
@@ -1184,10 +1250,12 @@ static size_t LZ4F_decodeHeader(LZ4F_dctx* dctx, const void* src, size_t srcSize
* note 1 : in case of error, dctx is not modified. Decoding operations can resume from where they stopped.
* note 2 : frame parameters are *copied into* an already allocated LZ4F_frameInfo_t structure.
*/
-LZ4F_errorCode_t LZ4F_getFrameInfo(LZ4F_dctx* dctx, LZ4F_frameInfo_t* frameInfoPtr,
- const void* srcBuffer, size_t* srcSizePtr)
+LZ4F_errorCode_t LZ4F_getFrameInfo(LZ4F_dctx* dctx,
+ LZ4F_frameInfo_t* frameInfoPtr,
+ const void* srcBuffer, size_t* srcSizePtr)
{
- if (dctx->dStage > dstage_storeFrameHeader) { /* assumption : dstage_* header enum at beginning of range */
+ LZ4F_STATIC_ASSERT(dstage_getFrameHeader < dstage_storeFrameHeader);
+ if (dctx->dStage > dstage_storeFrameHeader) {
/* frameInfo already decoded */
size_t o=0, i=0;
*srcSizePtr = 0;
@@ -1200,7 +1268,6 @@ LZ4F_errorCode_t LZ4F_getFrameInfo(LZ4F_dctx* dctx, LZ4F_frameInfo_t* frameInfoP
*srcSizePtr = 0;
return err0r(LZ4F_ERROR_frameDecoding_alreadyStarted);
} else {
- size_t decodeResult;
size_t const hSize = LZ4F_headerSize(srcBuffer, *srcSizePtr);
if (LZ4F_isError(hSize)) { *srcSizePtr=0; return hSize; }
if (*srcSizePtr < hSize) {
@@ -1208,44 +1275,51 @@ LZ4F_errorCode_t LZ4F_getFrameInfo(LZ4F_dctx* dctx, LZ4F_frameInfo_t* frameInfoP
return err0r(LZ4F_ERROR_frameHeader_incomplete);
}
- decodeResult = LZ4F_decodeHeader(dctx, srcBuffer, hSize);
- if (LZ4F_isError(decodeResult)) {
- *srcSizePtr = 0;
- } else {
- *srcSizePtr = decodeResult;
- decodeResult = BHSize; /* block header size */
- }
- *frameInfoPtr = dctx->frameInfo;
- return decodeResult;
- } }
+ { size_t decodeResult = LZ4F_decodeHeader(dctx, srcBuffer, hSize);
+ if (LZ4F_isError(decodeResult)) {
+ *srcSizePtr = 0;
+ } else {
+ *srcSizePtr = decodeResult;
+ decodeResult = BHSize; /* block header size */
+ }
+ *frameInfoPtr = dctx->frameInfo;
+ return decodeResult;
+ } } }
}
/* LZ4F_updateDict() :
- * only used for LZ4F_blockLinked mode */
+ * only used for LZ4F_blockLinked mode
+ * Condition : dstPtr != NULL
+ */
static void LZ4F_updateDict(LZ4F_dctx* dctx,
const BYTE* dstPtr, size_t dstSize, const BYTE* dstBufferStart,
unsigned withinTmp)
{
- if (dctx->dictSize==0)
- dctx->dict = (const BYTE*)dstPtr; /* priority to dictionary continuity */
+ assert(dstPtr != NULL);
+ if (dctx->dictSize==0) {
+ dctx->dict = (const BYTE*)dstPtr; /* priority to prefix mode */
+ }
+ assert(dctx->dict != NULL);
- if (dctx->dict + dctx->dictSize == dstPtr) { /* dictionary continuity, directly within dstBuffer */
+ if (dctx->dict + dctx->dictSize == dstPtr) { /* prefix mode, everything within dstBuffer */
dctx->dictSize += dstSize;
return;
}
- if (dstPtr - dstBufferStart + dstSize >= 64 KB) { /* history in dstBuffer becomes large enough to become dictionary */
+ assert(dstPtr >= dstBufferStart);
+ if ((size_t)(dstPtr - dstBufferStart) + dstSize >= 64 KB) { /* history in dstBuffer becomes large enough to become dictionary */
dctx->dict = (const BYTE*)dstBufferStart;
- dctx->dictSize = dstPtr - dstBufferStart + dstSize;
+ dctx->dictSize = (size_t)(dstPtr - dstBufferStart) + dstSize;
return;
}
assert(dstSize < 64 KB); /* if dstSize >= 64 KB, dictionary would be set into dstBuffer directly */
- /* dstBuffer does not contain whole useful history (64 KB), so it must be saved within tmpOut */
+ /* dstBuffer does not contain whole useful history (64 KB), so it must be saved within tmpOutBuffer */
+ assert(dctx->tmpOutBuffer != NULL);
- if ((withinTmp) && (dctx->dict == dctx->tmpOutBuffer)) { /* continue history within tmpOutBuffer */
+ if (withinTmp && (dctx->dict == dctx->tmpOutBuffer)) { /* continue history within tmpOutBuffer */
/* withinTmp expectation : content of [dstPtr,dstSize] is same as [dict+dictSize,dstSize], so we just extend it */
assert(dctx->dict + dctx->dictSize == dctx->tmpOut + dctx->tmpOutStart);
dctx->dictSize += dstSize;
@@ -1253,7 +1327,7 @@ static void LZ4F_updateDict(LZ4F_dctx* dctx,
}
if (withinTmp) { /* copy relevant dict portion in front of tmpOut within tmpOutBuffer */
- size_t const preserveSize = dctx->tmpOut - dctx->tmpOutBuffer;
+ size_t const preserveSize = (size_t)(dctx->tmpOut - dctx->tmpOutBuffer);
size_t copySize = 64 KB - dctx->tmpOutSize;
const BYTE* const oldDictEnd = dctx->dict + dctx->dictSize - dctx->tmpOutStart;
if (dctx->tmpOutSize > 64 KB) copySize = 0;
@@ -1317,17 +1391,21 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
const BYTE* const srcEnd = srcStart + *srcSizePtr;
const BYTE* srcPtr = srcStart;
BYTE* const dstStart = (BYTE*)dstBuffer;
- BYTE* const dstEnd = dstStart + *dstSizePtr;
+ BYTE* const dstEnd = dstStart ? dstStart + *dstSizePtr : NULL;
BYTE* dstPtr = dstStart;
const BYTE* selectedIn = NULL;
unsigned doAnotherStage = 1;
size_t nextSrcSizeHint = 1;
+ DEBUGLOG(5, "LZ4F_decompress : %p,%u => %p,%u",
+ srcBuffer, (unsigned)*srcSizePtr, dstBuffer, (unsigned)*dstSizePtr);
+ if (dstBuffer == NULL) assert(*dstSizePtr == 0);
MEM_INIT(&optionsNull, 0, sizeof(optionsNull));
if (decompressOptionsPtr==NULL) decompressOptionsPtr = &optionsNull;
*srcSizePtr = 0;
*dstSizePtr = 0;
+ assert(dctx != NULL);
/* behaves as a state machine */
@@ -1337,8 +1415,9 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
{
case dstage_getFrameHeader:
+ DEBUGLOG(6, "dstage_getFrameHeader");
if ((size_t)(srcEnd-srcPtr) >= maxFHSize) { /* enough to decode - shortcut */
- size_t const hSize = LZ4F_decodeHeader(dctx, srcPtr, srcEnd-srcPtr); /* will update dStage appropriately */
+ size_t const hSize = LZ4F_decodeHeader(dctx, srcPtr, (size_t)(srcEnd-srcPtr)); /* will update dStage appropriately */
if (LZ4F_isError(hSize)) return hSize;
srcPtr += hSize;
break;
@@ -1350,6 +1429,7 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
/* fall-through */
case dstage_storeFrameHeader:
+ DEBUGLOG(6, "dstage_storeFrameHeader");
{ size_t const sizeToCopy = MIN(dctx->tmpInTarget - dctx->tmpInSize, (size_t)(srcEnd - srcPtr));
memcpy(dctx->header + dctx->tmpInSize, srcPtr, sizeToCopy);
dctx->tmpInSize += sizeToCopy;
@@ -1366,14 +1446,15 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
break;
case dstage_init:
- if (dctx->frameInfo.contentChecksumFlag) XXH32_reset(&(dctx->xxh), 0);
+ DEBUGLOG(6, "dstage_init");
+ if (dctx->frameInfo.contentChecksumFlag) (void)XXH32_reset(&(dctx->xxh), 0);
/* internal buffers allocation */
{ size_t const bufferNeeded = dctx->maxBlockSize
- + ((dctx->frameInfo.blockMode==LZ4F_blockLinked) * 128 KB);
+ + ((dctx->frameInfo.blockMode==LZ4F_blockLinked) ? 128 KB : 0);
if (bufferNeeded > dctx->maxBufferSize) { /* tmp buffers too small */
dctx->maxBufferSize = 0; /* ensure allocation will be re-attempted on next entry*/
FREEMEM(dctx->tmpIn);
- dctx->tmpIn = (BYTE*)ALLOC(dctx->maxBlockSize + 4 /* block checksum */);
+ dctx->tmpIn = (BYTE*)ALLOC(dctx->maxBlockSize + BFSize /* block checksum */);
if (dctx->tmpIn == NULL)
return err0r(LZ4F_ERROR_allocation_failed);
FREEMEM(dctx->tmpOutBuffer);
@@ -1419,19 +1500,23 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
} /* if (dctx->dStage == dstage_storeBlockHeader) */
/* decode block header */
- { size_t const nextCBlockSize = LZ4F_readLE32(selectedIn) & 0x7FFFFFFFU;
- size_t const crcSize = dctx->frameInfo.blockChecksumFlag * 4;
- if (nextCBlockSize==0) { /* frameEnd signal, no more block */
+ { U32 const blockHeader = LZ4F_readLE32(selectedIn);
+ size_t const nextCBlockSize = blockHeader & 0x7FFFFFFFU;
+ size_t const crcSize = dctx->frameInfo.blockChecksumFlag * BFSize;
+ if (blockHeader==0) { /* frameEnd signal, no more block */
+ DEBUGLOG(5, "end of frame");
dctx->dStage = dstage_getSuffix;
break;
}
- if (nextCBlockSize > dctx->maxBlockSize)
+ if (nextCBlockSize > dctx->maxBlockSize) {
return err0r(LZ4F_ERROR_maxBlockSize_invalid);
- if (LZ4F_readLE32(selectedIn) & LZ4F_BLOCKUNCOMPRESSED_FLAG) {
+ }
+ if (blockHeader & LZ4F_BLOCKUNCOMPRESSED_FLAG) {
/* next block is uncompressed */
dctx->tmpInTarget = nextCBlockSize;
+ DEBUGLOG(5, "next block is uncompressed (size %u)", (U32)nextCBlockSize);
if (dctx->frameInfo.blockChecksumFlag) {
- XXH32_reset(&dctx->blockChecksum, 0);
+ (void)XXH32_reset(&dctx->blockChecksum, 0);
}
dctx->dStage = dstage_copyDirect;
break;
@@ -1439,28 +1524,34 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
/* next block is a compressed block */
dctx->tmpInTarget = nextCBlockSize + crcSize;
dctx->dStage = dstage_getCBlock;
- if (dstPtr==dstEnd) {
- nextSrcSizeHint = nextCBlockSize + crcSize + BHSize;
+ if (dstPtr==dstEnd || srcPtr==srcEnd) {
+ nextSrcSizeHint = BHSize + nextCBlockSize + crcSize;
doAnotherStage = 0;
}
break;
}
case dstage_copyDirect: /* uncompressed block */
- { size_t const minBuffSize = MIN((size_t)(srcEnd-srcPtr), (size_t)(dstEnd-dstPtr));
- size_t const sizeToCopy = MIN(dctx->tmpInTarget, minBuffSize);
- memcpy(dstPtr, srcPtr, sizeToCopy);
- if (dctx->frameInfo.blockChecksumFlag) {
- XXH32_update(&dctx->blockChecksum, srcPtr, sizeToCopy);
- }
- if (dctx->frameInfo.contentChecksumFlag)
- XXH32_update(&dctx->xxh, srcPtr, sizeToCopy);
- if (dctx->frameInfo.contentSize)
- dctx->frameRemainingSize -= sizeToCopy;
+ DEBUGLOG(6, "dstage_copyDirect");
+ { size_t sizeToCopy;
+ if (dstPtr == NULL) {
+ sizeToCopy = 0;
+ } else {
+ size_t const minBuffSize = MIN((size_t)(srcEnd-srcPtr), (size_t)(dstEnd-dstPtr));
+ sizeToCopy = MIN(dctx->tmpInTarget, minBuffSize);
+ memcpy(dstPtr, srcPtr, sizeToCopy);
+ if (dctx->frameInfo.blockChecksumFlag) {
+ (void)XXH32_update(&dctx->blockChecksum, srcPtr, sizeToCopy);
+ }
+ if (dctx->frameInfo.contentChecksumFlag)
+ (void)XXH32_update(&dctx->xxh, srcPtr, sizeToCopy);
+ if (dctx->frameInfo.contentSize)
+ dctx->frameRemainingSize -= sizeToCopy;
- /* history management (linked blocks only)*/
- if (dctx->frameInfo.blockMode == LZ4F_blockLinked)
- LZ4F_updateDict(dctx, dstPtr, sizeToCopy, dstStart, 0);
+ /* history management (linked blocks only)*/
+ if (dctx->frameInfo.blockMode == LZ4F_blockLinked) {
+ LZ4F_updateDict(dctx, dstPtr, sizeToCopy, dstStart, 0);
+ } }
srcPtr += sizeToCopy;
dstPtr += sizeToCopy;
@@ -1473,15 +1564,16 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
break;
}
dctx->tmpInTarget -= sizeToCopy; /* need to copy more */
- nextSrcSizeHint = dctx->tmpInTarget +
- + dctx->frameInfo.contentChecksumFlag * 4 /* block checksum */
- + BHSize /* next header size */;
- doAnotherStage = 0;
- break;
}
+ nextSrcSizeHint = dctx->tmpInTarget +
+ +(dctx->frameInfo.blockChecksumFlag ? BFSize : 0)
+ + BHSize /* next header size */;
+ doAnotherStage = 0;
+ break;
/* check block checksum for recently transferred uncompressed block */
case dstage_getBlockChecksum:
+ DEBUGLOG(6, "dstage_getBlockChecksum");
{ const void* crcSrc;
if ((srcEnd-srcPtr >= 4) && (dctx->tmpInSize==0)) {
crcSrc = srcPtr;
@@ -1500,13 +1592,23 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
}
{ U32 const readCRC = LZ4F_readLE32(crcSrc);
U32 const calcCRC = XXH32_digest(&dctx->blockChecksum);
- if (readCRC != calcCRC)
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
+ DEBUGLOG(6, "compare block checksum");
+ if (readCRC != calcCRC) {
+ DEBUGLOG(4, "incorrect block checksum: %08X != %08X",
+ readCRC, calcCRC);
return err0r(LZ4F_ERROR_blockChecksum_invalid);
+ }
+#else
+ (void)readCRC;
+ (void)calcCRC;
+#endif
} }
dctx->dStage = dstage_getBlockHeader; /* new block */
break;
case dstage_getCBlock:
+ DEBUGLOG(6, "dstage_getCBlock");
if ((size_t)(srcEnd-srcPtr) < dctx->tmpInTarget) {
dctx->tmpInSize = 0;
dctx->dStage = dstage_storeCBlock;
@@ -1516,7 +1618,7 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
selectedIn = srcPtr;
srcPtr += dctx->tmpInTarget;
- if (0) /* jump over next block */
+ if (0) /* always jump over next block */
case dstage_storeCBlock:
{ size_t const wantedData = dctx->tmpInTarget - dctx->tmpInSize;
size_t const inputLeft = (size_t)(srcEnd-srcPtr);
@@ -1525,8 +1627,10 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
dctx->tmpInSize += sizeToCopy;
srcPtr += sizeToCopy;
if (dctx->tmpInSize < dctx->tmpInTarget) { /* need more input */
- nextSrcSizeHint = (dctx->tmpInTarget - dctx->tmpInSize) + BHSize;
- doAnotherStage=0;
+ nextSrcSizeHint = (dctx->tmpInTarget - dctx->tmpInSize)
+ + (dctx->frameInfo.blockChecksumFlag ? BFSize : 0)
+ + BHSize /* next header size */;
+ doAnotherStage = 0;
break;
}
selectedIn = dctx->tmpIn;
@@ -1538,14 +1642,20 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
assert(selectedIn != NULL); /* selectedIn is defined at this stage (either srcPtr, or dctx->tmpIn) */
{ U32 const readBlockCrc = LZ4F_readLE32(selectedIn + dctx->tmpInTarget);
U32 const calcBlockCrc = XXH32(selectedIn, dctx->tmpInTarget, 0);
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
if (readBlockCrc != calcBlockCrc)
return err0r(LZ4F_ERROR_blockChecksum_invalid);
+#else
+ (void)readBlockCrc;
+ (void)calcBlockCrc;
+#endif
} }
if ((size_t)(dstEnd-dstPtr) >= dctx->maxBlockSize) {
const char* dict = (const char*)dctx->dict;
size_t dictSize = dctx->dictSize;
int decodedSize;
+ assert(dstPtr != NULL);
if (dict && dictSize > 1 GB) {
/* the dictSize param is an int, avoid truncation / sign issues */
dict += dictSize - 64 KB;
@@ -1558,13 +1668,14 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
dict, (int)dictSize);
if (decodedSize < 0) return err0r(LZ4F_ERROR_GENERIC); /* decompression failed */
if (dctx->frameInfo.contentChecksumFlag)
- XXH32_update(&(dctx->xxh), dstPtr, decodedSize);
+ XXH32_update(&(dctx->xxh), dstPtr, (size_t)decodedSize);
if (dctx->frameInfo.contentSize)
- dctx->frameRemainingSize -= decodedSize;
+ dctx->frameRemainingSize -= (size_t)decodedSize;
/* dictionary management */
- if (dctx->frameInfo.blockMode==LZ4F_blockLinked)
- LZ4F_updateDict(dctx, dstPtr, decodedSize, dstStart, 0);
+ if (dctx->frameInfo.blockMode==LZ4F_blockLinked) {
+ LZ4F_updateDict(dctx, dstPtr, (size_t)decodedSize, dstStart, 0);
+ }
dstPtr += decodedSize;
dctx->dStage = dstage_getBlockHeader;
@@ -1601,17 +1712,19 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
if (decodedSize < 0) /* decompression failed */
return err0r(LZ4F_ERROR_decompressionFailed);
if (dctx->frameInfo.contentChecksumFlag)
- XXH32_update(&(dctx->xxh), dctx->tmpOut, decodedSize);
+ XXH32_update(&(dctx->xxh), dctx->tmpOut, (size_t)decodedSize);
if (dctx->frameInfo.contentSize)
- dctx->frameRemainingSize -= decodedSize;
- dctx->tmpOutSize = decodedSize;
+ dctx->frameRemainingSize -= (size_t)decodedSize;
+ dctx->tmpOutSize = (size_t)decodedSize;
dctx->tmpOutStart = 0;
dctx->dStage = dstage_flushOut;
}
/* fall-through */
case dstage_flushOut: /* flush decoded data from tmpOut to dstBuffer */
- { size_t const sizeToCopy = MIN(dctx->tmpOutSize - dctx->tmpOutStart, (size_t)(dstEnd-dstPtr));
+ DEBUGLOG(6, "dstage_flushOut");
+ if (dstPtr != NULL) {
+ size_t const sizeToCopy = MIN(dctx->tmpOutSize - dctx->tmpOutStart, (size_t)(dstEnd-dstPtr));
memcpy(dstPtr, dctx->tmpOut + dctx->tmpOutStart, sizeToCopy);
/* dictionary management */
@@ -1620,16 +1733,15 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
dctx->tmpOutStart += sizeToCopy;
dstPtr += sizeToCopy;
-
- if (dctx->tmpOutStart == dctx->tmpOutSize) { /* all flushed */
- dctx->dStage = dstage_getBlockHeader; /* get next block */
- break;
- }
- /* could not flush everything : stop there, just request a block header */
- doAnotherStage = 0;
- nextSrcSizeHint = BHSize;
+ }
+ if (dctx->tmpOutStart == dctx->tmpOutSize) { /* all flushed */
+ dctx->dStage = dstage_getBlockHeader; /* get next block */
break;
}
+ /* could not flush everything : stop there, just request a block header */
+ doAnotherStage = 0;
+ nextSrcSizeHint = BHSize;
+ break;
case dstage_getSuffix:
if (dctx->frameRemainingSize)
@@ -1667,8 +1779,13 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
/* case dstage_checkSuffix: */ /* no direct entry, avoid initialization risks */
{ U32 const readCRC = LZ4F_readLE32(selectedIn);
U32 const resultCRC = XXH32_digest(&(dctx->xxh));
+#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
if (readCRC != resultCRC)
return err0r(LZ4F_ERROR_contentChecksum_invalid);
+#else
+ (void)readCRC;
+ (void)resultCRC;
+#endif
nextSrcSizeHint = 0;
LZ4F_resetDecompressionContext(dctx);
doAnotherStage = 0;
@@ -1728,18 +1845,19 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
LZ4F_STATIC_ASSERT((unsigned)dstage_init == 2);
if ( (dctx->frameInfo.blockMode==LZ4F_blockLinked) /* next block will use up to 64KB from previous ones */
&& (dctx->dict != dctx->tmpOutBuffer) /* dictionary is not already within tmp */
+ && (dctx->dict != NULL) /* dictionary exists */
&& (!decompressOptionsPtr->stableDst) /* cannot rely on dst data to remain there for next call */
&& ((unsigned)(dctx->dStage)-2 < (unsigned)(dstage_getSuffix)-2) ) /* valid stages : [init ... getSuffix[ */
{
if (dctx->dStage == dstage_flushOut) {
- size_t const preserveSize = dctx->tmpOut - dctx->tmpOutBuffer;
+ size_t const preserveSize = (size_t)(dctx->tmpOut - dctx->tmpOutBuffer);
size_t copySize = 64 KB - dctx->tmpOutSize;
const BYTE* oldDictEnd = dctx->dict + dctx->dictSize - dctx->tmpOutStart;
if (dctx->tmpOutSize > 64 KB) copySize = 0;
if (copySize > preserveSize) copySize = preserveSize;
+ assert(dctx->tmpOutBuffer != NULL);
- if (copySize > 0)
- memcpy(dctx->tmpOutBuffer + preserveSize - copySize, oldDictEnd - copySize, copySize);
+ memcpy(dctx->tmpOutBuffer + preserveSize - copySize, oldDictEnd - copySize, copySize);
dctx->dict = dctx->tmpOutBuffer;
dctx->dictSize = preserveSize + dctx->tmpOutStart;
@@ -1747,8 +1865,7 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
const BYTE* const oldDictEnd = dctx->dict + dctx->dictSize;
size_t const newDictSize = MIN(dctx->dictSize, 64 KB);
- if (newDictSize > 0)
- memcpy(dctx->tmpOutBuffer, oldDictEnd - newDictSize, newDictSize);
+ memcpy(dctx->tmpOutBuffer, oldDictEnd - newDictSize, newDictSize);
dctx->dict = dctx->tmpOutBuffer;
dctx->dictSize = newDictSize;
@@ -1756,8 +1873,8 @@ size_t LZ4F_decompress(LZ4F_dctx* dctx,
}
}
- *srcSizePtr = (srcPtr - srcStart);
- *dstSizePtr = (dstPtr - dstStart);
+ *srcSizePtr = (size_t)(srcPtr - srcStart);
+ *dstSizePtr = (size_t)(dstPtr - dstStart);
return nextSrcSizeHint;
}
diff --git a/lib/lz4frame.h b/lib/lz4frame.h
index 75f1fd9..4573317 100644
--- a/lib/lz4frame.h
+++ b/lib/lz4frame.h
@@ -32,11 +32,14 @@
- LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c
*/
-/* LZ4F is a stand-alone API to create LZ4-compressed frames
- * conformant with specification v1.6.1.
- * It also offers streaming capabilities.
+/* LZ4F is a stand-alone API able to create and decode LZ4 frames
+ * conformant with specification v1.6.1 in doc/lz4_Frame_format.md .
+ * Generated frames are compatible with `lz4` CLI.
+ *
+ * LZ4F also offers streaming capabilities.
+ *
* lz4.h is not required when using lz4frame.h,
- * except to get constant such as LZ4_VERSION_NUMBER.
+ * except to extract common constant such as LZ4_VERSION_NUMBER.
* */
#ifndef LZ4F_H_09782039843
@@ -63,17 +66,22 @@ extern "C" {
*****************************************************************/
/* LZ4_DLL_EXPORT :
* Enable exporting of functions when building a Windows DLL
- * LZ4FLIB_API :
+ * LZ4FLIB_VISIBILITY :
* Control library symbols visibility.
*/
+#ifndef LZ4FLIB_VISIBILITY
+# if defined(__GNUC__) && (__GNUC__ >= 4)
+# define LZ4FLIB_VISIBILITY __attribute__ ((visibility ("default")))
+# else
+# define LZ4FLIB_VISIBILITY
+# endif
+#endif
#if defined(LZ4_DLL_EXPORT) && (LZ4_DLL_EXPORT==1)
-# define LZ4FLIB_API __declspec(dllexport)
+# define LZ4FLIB_API __declspec(dllexport) LZ4FLIB_VISIBILITY
#elif defined(LZ4_DLL_IMPORT) && (LZ4_DLL_IMPORT==1)
-# define LZ4FLIB_API __declspec(dllimport)
-#elif defined(__GNUC__) && (__GNUC__ >= 4)
-# define LZ4FLIB_API __attribute__ ((__visibility__ ("default")))
+# define LZ4FLIB_API __declspec(dllimport) LZ4FLIB_VISIBILITY
#else
-# define LZ4FLIB_API
+# define LZ4FLIB_API LZ4FLIB_VISIBILITY
#endif
#ifdef LZ4F_DISABLE_DEPRECATE_WARNINGS
@@ -100,7 +108,7 @@ LZ4FLIB_API const char* LZ4F_getErrorName(LZ4F_errorCode_t code); /**< return
/*-************************************
* Frame compression types
- **************************************/
+ ************************************* */
/* #define LZ4F_ENABLE_OBSOLETE_ENUMS // uncomment to enable obsolete enums */
#ifdef LZ4F_ENABLE_OBSOLETE_ENUMS
# define LZ4F_OBSOLETE_ENUM(x) , LZ4F_DEPRECATE(x) = LZ4F_##x
@@ -110,7 +118,8 @@ LZ4FLIB_API const char* LZ4F_getErrorName(LZ4F_errorCode_t code); /**< return
/* The larger the block size, the (slightly) better the compression ratio,
* though there are diminishing returns.
- * Larger blocks also increase memory usage on both compression and decompression sides. */
+ * Larger blocks also increase memory usage on both compression and decompression sides.
+ */
typedef enum {
LZ4F_default=0,
LZ4F_max64KB=4,
@@ -173,7 +182,7 @@ typedef struct {
LZ4F_blockChecksum_t blockChecksumFlag; /* 1: each block followed by a checksum of block's compressed data; 0: disabled (default) */
} LZ4F_frameInfo_t;
-#define LZ4F_INIT_FRAMEINFO { 0, 0, 0, 0, 0, 0, 0 } /* v1.8.3+ */
+#define LZ4F_INIT_FRAMEINFO { LZ4F_default, LZ4F_blockLinked, LZ4F_noContentChecksum, LZ4F_frame, 0ULL, 0U, LZ4F_noBlockChecksum } /* v1.8.3+ */
/*! LZ4F_preferences_t :
* makes it possible to supply advanced compression instructions to streaming interface.
@@ -188,14 +197,14 @@ typedef struct {
unsigned reserved[3]; /* must be zero for forward compatibility */
} LZ4F_preferences_t;
-#define LZ4F_INIT_PREFERENCES { LZ4F_INIT_FRAMEINFO, 0, 0, 0, { 0, 0, 0 } } /* v1.8.3+ */
+#define LZ4F_INIT_PREFERENCES { LZ4F_INIT_FRAMEINFO, 0, 0u, 0u, { 0u, 0u, 0u } } /* v1.8.3+ */
/*-*********************************
* Simple compression function
***********************************/
-LZ4FLIB_API int LZ4F_compressionLevel_max(void);
+LZ4FLIB_API int LZ4F_compressionLevel_max(void); /* v1.8.0+ */
/*! LZ4F_compressFrameBound() :
* Returns the maximum possible compressed size with LZ4F_compressFrame() given srcSize and preferences.
@@ -247,7 +256,18 @@ LZ4FLIB_API LZ4F_errorCode_t LZ4F_freeCompressionContext(LZ4F_cctx* cctx);
/*---- Compression ----*/
-#define LZ4F_HEADER_SIZE_MAX 19 /* LZ4 Frame header size can vary from 7 to 19 bytes */
+#define LZ4F_HEADER_SIZE_MIN 7 /* LZ4 Frame header size can vary, depending on selected paramaters */
+#define LZ4F_HEADER_SIZE_MAX 19
+
+/* Size in bytes of a block header in little-endian format. Highest bit indicates if block data is uncompressed */
+#define LZ4F_BLOCK_HEADER_SIZE 4
+
+/* Size in bytes of a block checksum footer in little-endian format. */
+#define LZ4F_BLOCK_CHECKSUM_SIZE 4
+
+/* Size in bytes of the content checksum. */
+#define LZ4F_CONTENT_CHECKSUM_SIZE 4
+
/*! LZ4F_compressBegin() :
* will write the frame header into dstBuffer.
* dstCapacity must be >= LZ4F_HEADER_SIZE_MAX bytes.
@@ -260,15 +280,19 @@ LZ4FLIB_API size_t LZ4F_compressBegin(LZ4F_cctx* cctx,
const LZ4F_preferences_t* prefsPtr);
/*! LZ4F_compressBound() :
- * Provides minimum dstCapacity required to guarantee compression success
- * given a srcSize and preferences, covering worst case scenario.
+ * Provides minimum dstCapacity required to guarantee success of
+ * LZ4F_compressUpdate(), given a srcSize and preferences, for a worst case scenario.
+ * When srcSize==0, LZ4F_compressBound() provides an upper bound for LZ4F_flush() and LZ4F_compressEnd() instead.
+ * Note that the result is only valid for a single invocation of LZ4F_compressUpdate().
+ * When invoking LZ4F_compressUpdate() multiple times,
+ * if the output buffer is gradually filled up instead of emptied and re-used from its start,
+ * one must check if there is enough remaining capacity before each invocation, using LZ4F_compressBound().
+ * @return is always the same for a srcSize and prefsPtr.
* prefsPtr is optional : when NULL is provided, preferences will be set to cover worst case scenario.
- * Estimation is valid for either LZ4F_compressUpdate(), LZ4F_flush() or LZ4F_compressEnd(),
- * Estimation includes the possibility that internal buffer might already be filled by up to (blockSize-1) bytes.
- * It also includes frame footer (ending + checksum), which would have to be generated by LZ4F_compressEnd().
- * Estimation doesn't include frame header, as it was already generated by LZ4F_compressBegin().
- * Result is always the same for a srcSize and prefsPtr, so it can be trusted to size reusable buffers.
- * When srcSize==0, LZ4F_compressBound() provides an upper bound for LZ4F_flush() and LZ4F_compressEnd() operations.
+ * tech details :
+ * @return if automatic flushing is not enabled, includes the possibility that internal buffer might already be filled by up to (blockSize-1) bytes.
+ * It also includes frame footer (ending + checksum), since it might be generated by LZ4F_compressEnd().
+ * @return doesn't include frame header, as it was already generated by LZ4F_compressBegin().
*/
LZ4FLIB_API size_t LZ4F_compressBound(size_t srcSize, const LZ4F_preferences_t* prefsPtr);
@@ -295,6 +319,7 @@ LZ4FLIB_API size_t LZ4F_compressUpdate(LZ4F_cctx* cctx,
* `cOptPtr` is optional : it's possible to provide NULL, all options will be set to default.
* @return : nb of bytes written into dstBuffer (can be zero, when there is no data stored within cctx)
* or an error code if it fails (which can be tested using LZ4F_isError())
+ * Note : LZ4F_flush() is guaranteed to be successful when dstCapacity >= LZ4F_compressBound(0, prefsPtr).
*/
LZ4FLIB_API size_t LZ4F_flush(LZ4F_cctx* cctx,
void* dstBuffer, size_t dstCapacity,
@@ -307,6 +332,7 @@ LZ4FLIB_API size_t LZ4F_flush(LZ4F_cctx* cctx,
* `cOptPtr` is optional : NULL can be provided, in which case all options will be set to default.
* @return : nb of bytes written into dstBuffer, necessarily >= 4 (endMark),
* or an error code if it fails (which can be tested using LZ4F_isError())
+ * Note : LZ4F_compressEnd() is guaranteed to be successful when dstCapacity >= LZ4F_compressBound(0, prefsPtr).
* A successful call to LZ4F_compressEnd() makes `cctx` available again for another compression task.
*/
LZ4FLIB_API size_t LZ4F_compressEnd(LZ4F_cctx* cctx,
@@ -345,23 +371,58 @@ LZ4FLIB_API LZ4F_errorCode_t LZ4F_freeDecompressionContext(LZ4F_dctx* dctx);
* Streaming decompression functions
*************************************/
+#define LZ4F_MIN_SIZE_TO_KNOW_HEADER_LENGTH 5
+
+/*! LZ4F_headerSize() : v1.9.0+
+ * Provide the header size of a frame starting at `src`.
+ * `srcSize` must be >= LZ4F_MIN_SIZE_TO_KNOW_HEADER_LENGTH,
+ * which is enough to decode the header length.
+ * @return : size of frame header
+ * or an error code, which can be tested using LZ4F_isError()
+ * note : Frame header size is variable, but is guaranteed to be
+ * >= LZ4F_HEADER_SIZE_MIN bytes, and <= LZ4F_HEADER_SIZE_MAX bytes.
+ */
+LZ4FLIB_API size_t LZ4F_headerSize(const void* src, size_t srcSize);
+
/*! LZ4F_getFrameInfo() :
* This function extracts frame parameters (max blockSize, dictID, etc.).
- * Its usage is optional.
- * Extracted information is typically useful for allocation and dictionary.
- * This function works in 2 situations :
- * - At the beginning of a new frame, in which case
- * it will decode information from `srcBuffer`, starting the decoding process.
- * Input size must be large enough to successfully decode the entire frame header.
- * Frame header size is variable, but is guaranteed to be <= LZ4F_HEADER_SIZE_MAX bytes.
- * It's allowed to provide more input data than this minimum.
- * - After decoding has been started.
- * In which case, no input is read, frame parameters are extracted from dctx.
- * - If decoding has barely started, but not yet extracted information from header,
+ * Its usage is optional: user can call LZ4F_decompress() directly.
+ *
+ * Extracted information will fill an existing LZ4F_frameInfo_t structure.
+ * This can be useful for allocation and dictionary identification purposes.
+ *
+ * LZ4F_getFrameInfo() can work in the following situations :
+ *
+ * 1) At the beginning of a new frame, before any invocation of LZ4F_decompress().
+ * It will decode header from `srcBuffer`,
+ * consuming the header and starting the decoding process.
+ *
+ * Input size must be large enough to contain the full frame header.
+ * Frame header size can be known beforehand by LZ4F_headerSize().
+ * Frame header size is variable, but is guaranteed to be >= LZ4F_HEADER_SIZE_MIN bytes,
+ * and not more than <= LZ4F_HEADER_SIZE_MAX bytes.
+ * Hence, blindly providing LZ4F_HEADER_SIZE_MAX bytes or more will always work.
+ * It's allowed to provide more input data than the header size,
+ * LZ4F_getFrameInfo() will only consume the header.
+ *
+ * If input size is not large enough,
+ * aka if it's smaller than header size,
+ * function will fail and return an error code.
+ *
+ * 2) After decoding has been started,
+ * it's possible to invoke LZ4F_getFrameInfo() anytime
+ * to extract already decoded frame parameters stored within dctx.
+ *
+ * Note that, if decoding has barely started,
+ * and not yet read enough information to decode the header,
* LZ4F_getFrameInfo() will fail.
- * The number of bytes consumed from srcBuffer will be updated within *srcSizePtr (necessarily <= original value).
- * Decompression must resume from (srcBuffer + *srcSizePtr).
- * @return : an hint about how many srcSize bytes LZ4F_decompress() expects for next call,
+ *
+ * The number of bytes consumed from srcBuffer will be updated in *srcSizePtr (necessarily <= original value).
+ * LZ4F_getFrameInfo() only consumes bytes when decoding has not yet started,
+ * and when decoding the header has been successful.
+ * Decompression must then resume from (srcBuffer + *srcSizePtr).
+ *
+ * @return : a hint about how many srcSize bytes LZ4F_decompress() expects for next call,
* or an error code which can be tested using LZ4F_isError().
* note 1 : in case of error, dctx is not modified. Decoding operation can resume from beginning safely.
* note 2 : frame parameters are *copied into* an already allocated LZ4F_frameInfo_t structure.
@@ -371,8 +432,10 @@ LZ4FLIB_API size_t LZ4F_getFrameInfo(LZ4F_dctx* dctx,
const void* srcBuffer, size_t* srcSizePtr);
/*! LZ4F_decompress() :
- * Call this function repetitively to regenerate compressed data from `srcBuffer`.
- * The function will read up to *srcSizePtr bytes from srcBuffer,
+ * Call this function repetitively to regenerate data compressed in `srcBuffer`.
+ *
+ * The function requires a valid dctx state.
+ * It will read up to *srcSizePtr bytes from srcBuffer,
* and decompress data into dstBuffer, of capacity *dstSizePtr.
*
* The nb of bytes consumed from srcBuffer will be written into *srcSizePtr (necessarily <= original value).
@@ -427,20 +490,20 @@ LZ4FLIB_API void LZ4F_resetDecompressionContext(LZ4F_dctx* dctx); /* always su
extern "C" {
#endif
-/* These declarations are not stable and may change in the future. They are
- * therefore only safe to depend on when the caller is statically linked
- * against the library. To access their declarations, define
- * LZ4F_STATIC_LINKING_ONLY.
+/* These declarations are not stable and may change in the future.
+ * They are therefore only safe to depend on
+ * when the caller is statically linked against the library.
+ * To access their declarations, define LZ4F_STATIC_LINKING_ONLY.
*
- * There is a further protection mechanism where these symbols aren't published
- * into shared/dynamic libraries. You can override this behavior and force
- * them to be published by defining LZ4F_PUBLISH_STATIC_FUNCTIONS. Use at
- * your own risk.
+ * By default, these symbols aren't published into shared/dynamic libraries.
+ * You can override this behavior and force them to be published
+ * by defining LZ4F_PUBLISH_STATIC_FUNCTIONS.
+ * Use at your own risk.
*/
#ifdef LZ4F_PUBLISH_STATIC_FUNCTIONS
-#define LZ4FLIB_STATIC_API LZ4FLIB_API
+# define LZ4FLIB_STATIC_API LZ4FLIB_API
#else
-#define LZ4FLIB_STATIC_API
+# define LZ4FLIB_STATIC_API
#endif
@@ -471,19 +534,38 @@ extern "C" {
#define LZ4F_GENERATE_ENUM(ENUM) LZ4F_##ENUM,
/* enum list is exposed, to handle specific errors */
-typedef enum { LZ4F_LIST_ERRORS(LZ4F_GENERATE_ENUM) } LZ4F_errorCodes;
+typedef enum { LZ4F_LIST_ERRORS(LZ4F_GENERATE_ENUM)
+ _LZ4F_dummy_error_enum_for_c89_never_used } LZ4F_errorCodes;
LZ4FLIB_STATIC_API LZ4F_errorCodes LZ4F_getErrorCode(size_t functionResult);
-
+LZ4FLIB_STATIC_API size_t LZ4F_getBlockSize(unsigned);
/**********************************
* Bulk processing dictionary API
*********************************/
+
+/* A Dictionary is useful for the compression of small messages (KB range).
+ * It dramatically improves compression efficiency.
+ *
+ * LZ4 can ingest any input as dictionary, though only the last 64 KB are useful.
+ * Best results are generally achieved by using Zstandard's Dictionary Builder
+ * to generate a high-quality dictionary from a set of samples.
+ *
+ * Loading a dictionary has a cost, since it involves construction of tables.
+ * The Bulk processing dictionary API makes it possible to share this cost
+ * over an arbitrary number of compression jobs, even concurrently,
+ * markedly improving compression latency for these cases.
+ *
+ * The same dictionary will have to be used on the decompression side
+ * for decoding to be successful.
+ * To help identify the correct dictionary at decoding stage,
+ * the frame header allows optional embedding of a dictID field.
+ */
typedef struct LZ4F_CDict_s LZ4F_CDict;
/*! LZ4_createCDict() :
- * When compressing multiple messages / blocks with the same dictionary, it's recommended to load it just once.
+ * When compressing multiple messages / blocks using the same dictionary, it's recommended to load it just once.
* LZ4_createCDict() will create a digested dictionary, ready to start future compression operations without startup delay.
* LZ4_CDict can be created once and shared by multiple threads concurrently, since its usage is read-only.
* `dictBuffer` can be released after LZ4_CDict creation, since its content is copied within CDict */
diff --git a/lib/lz4hc.c b/lib/lz4hc.c
index e913ee7..77c9f43 100644
--- a/lib/lz4hc.c
+++ b/lib/lz4hc.c
@@ -53,7 +53,7 @@
#include "lz4hc.h"
-/*=== Common LZ4 definitions ===*/
+/*=== Common definitions ===*/
#if defined(__GNUC__)
# pragma GCC diagnostic ignored "-Wunused-function"
#endif
@@ -62,7 +62,13 @@
#endif
#define LZ4_COMMONDEFS_ONLY
+#ifndef LZ4_SRC_INCLUDED
#include "lz4.c" /* LZ4_count, constants, mem */
+#endif
+
+
+/*=== Enums ===*/
+typedef enum { noDictCtx, usingDictCtxHc } dictCtx_directive;
/*=== Constants ===*/
@@ -76,25 +82,24 @@
#define HASH_FUNCTION(i) (((i) * 2654435761U) >> ((MINMATCH*8)-LZ4HC_HASH_LOG))
#define DELTANEXTMAXD(p) chainTable[(p) & LZ4HC_MAXD_MASK] /* flexible, LZ4HC_MAXD dependent */
#define DELTANEXTU16(table, pos) table[(U16)(pos)] /* faster */
+/* Make fields passed to, and updated by LZ4HC_encodeSequence explicit */
+#define UPDATABLE(ip, op, anchor) &ip, &op, &anchor
static U32 LZ4HC_hashPtr(const void* ptr) { return HASH_FUNCTION(LZ4_read32(ptr)); }
-/*=== Enums ===*/
-typedef enum { noDictCtx, usingDictCtx } dictCtx_directive;
-
/**************************************
* HC Compression
**************************************/
static void LZ4HC_clearTables (LZ4HC_CCtx_internal* hc4)
{
- MEM_INIT((void*)hc4->hashTable, 0, sizeof(hc4->hashTable));
+ MEM_INIT(hc4->hashTable, 0, sizeof(hc4->hashTable));
MEM_INIT(hc4->chainTable, 0xFF, sizeof(hc4->chainTable));
}
-static void LZ4HC_init (LZ4HC_CCtx_internal* hc4, const BYTE* start)
+static void LZ4HC_init_internal (LZ4HC_CCtx_internal* hc4, const BYTE* start)
{
- uptrval startingOffset = hc4->end - hc4->base;
+ uptrval startingOffset = (uptrval)(hc4->end - hc4->base);
if (startingOffset > 1 GB) {
LZ4HC_clearTables(hc4);
startingOffset = 0;
@@ -121,7 +126,7 @@ LZ4_FORCE_INLINE void LZ4HC_Insert (LZ4HC_CCtx_internal* hc4, const BYTE* ip)
while (idx < target) {
U32 const h = LZ4HC_hashPtr(base+idx);
size_t delta = idx - hashTable[h];
- if (delta>MAX_DISTANCE) delta = MAX_DISTANCE;
+ if (delta>LZ4_DISTANCE_MAX) delta = LZ4_DISTANCE_MAX;
DELTANEXTU16(chainTable, idx) = (U16)delta;
hashTable[h] = idx;
idx++;
@@ -147,13 +152,28 @@ int LZ4HC_countBack(const BYTE* const ip, const BYTE* const match,
return back;
}
+#if defined(_MSC_VER)
+# define LZ4HC_rotl32(x,r) _rotl(x,r)
+#else
+# define LZ4HC_rotl32(x,r) ((x << r) | (x >> (32 - r)))
+#endif
+
+
+static U32 LZ4HC_rotatePattern(size_t const rotate, U32 const pattern)
+{
+ size_t const bitsToRotate = (rotate & (sizeof(pattern) - 1)) << 3;
+ if (bitsToRotate == 0) return pattern;
+ return LZ4HC_rotl32(pattern, (int)bitsToRotate);
+}
+
/* LZ4HC_countPattern() :
* pattern32 must be a sample of repetitive pattern of length 1, 2 or 4 (but not 3!) */
static unsigned
LZ4HC_countPattern(const BYTE* ip, const BYTE* const iEnd, U32 const pattern32)
{
const BYTE* const iStart = ip;
- reg_t const pattern = (sizeof(pattern)==8) ? (reg_t)pattern32 + (((reg_t)pattern32) << 32) : pattern32;
+ reg_t const pattern = (sizeof(pattern)==8) ?
+ (reg_t)pattern32 + (((reg_t)pattern32) << (sizeof(pattern)*4)) : pattern32;
while (likely(ip < iEnd-(sizeof(pattern)-1))) {
reg_t const diff = LZ4_read_ARCH(ip) ^ pattern;
@@ -199,6 +219,16 @@ LZ4HC_reverseCountPattern(const BYTE* ip, const BYTE* const iLow, U32 pattern)
return (unsigned)(iStart - ip);
}
+/* LZ4HC_protectDictEnd() :
+ * Checks if the match is in the last 3 bytes of the dictionary, so reading the
+ * 4 byte MINMATCH would overflow.
+ * @returns true if the match index is okay.
+ */
+static int LZ4HC_protectDictEnd(U32 const dictLimit, U32 const matchIndex)
+{
+ return ((U32)((dictLimit - 1) - matchIndex) >= 3);
+}
+
typedef enum { rep_untested, rep_not, rep_confirmed } repeat_state_e;
typedef enum { favorCompressionRatio=0, favorDecompressionSpeed } HCfavor_e;
@@ -224,14 +254,13 @@ LZ4HC_InsertAndGetWiderMatch (
const U32 dictLimit = hc4->dictLimit;
const BYTE* const lowPrefixPtr = base + dictLimit;
const U32 ipIndex = (U32)(ip - base);
- const U32 lowestMatchIndex = (hc4->lowLimit + 64 KB > ipIndex) ? hc4->lowLimit : ipIndex - MAX_DISTANCE;
+ const U32 lowestMatchIndex = (hc4->lowLimit + (LZ4_DISTANCE_MAX + 1) > ipIndex) ? hc4->lowLimit : ipIndex - LZ4_DISTANCE_MAX;
const BYTE* const dictBase = hc4->dictBase;
int const lookBackLength = (int)(ip-iLowLimit);
int nbAttempts = maxNbAttempts;
- int matchChainPos = 0;
+ U32 matchChainPos = 0;
U32 const pattern = LZ4_read32(ip);
U32 matchIndex;
- U32 dictMatchIndex;
repeat_state_e repeat = rep_untested;
size_t srcPatternLength = 0;
@@ -242,7 +271,7 @@ LZ4HC_InsertAndGetWiderMatch (
DEBUGLOG(7, "First match at index %u / %u (lowestMatchIndex)",
matchIndex, lowestMatchIndex);
- while ((matchIndex>=lowestMatchIndex) && (nbAttempts)) {
+ while ((matchIndex>=lowestMatchIndex) && (nbAttempts>0)) {
int matchLength=0;
nbAttempts--;
assert(matchIndex < ipIndex);
@@ -256,7 +285,7 @@ LZ4HC_InsertAndGetWiderMatch (
if (LZ4_read16(iLowLimit + longest - 1) == LZ4_read16(matchPtr - lookBackLength + longest - 1)) {
if (LZ4_read32(matchPtr) == pattern) {
int const back = lookBackLength ? LZ4HC_countBack(ip, matchPtr, iLowLimit, lowPrefixPtr) : 0;
- matchLength = MINMATCH + LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, iHighLimit);
+ matchLength = MINMATCH + (int)LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, iHighLimit);
matchLength -= back;
if (matchLength > longest) {
longest = matchLength;
@@ -270,7 +299,7 @@ LZ4HC_InsertAndGetWiderMatch (
int back = 0;
const BYTE* vLimit = ip + (dictLimit - matchIndex);
if (vLimit > iHighLimit) vLimit = iHighLimit;
- matchLength = LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, vLimit) + MINMATCH;
+ matchLength = (int)LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, vLimit) + MINMATCH;
if ((ip+matchLength == vLimit) && (vLimit < iHighLimit))
matchLength += LZ4_count(ip+matchLength, lowPrefixPtr, iHighLimit);
back = lookBackLength ? LZ4HC_countBack(ip, matchPtr, iLowLimit, dictStart) : 0;
@@ -283,15 +312,22 @@ LZ4HC_InsertAndGetWiderMatch (
if (chainSwap && matchLength==longest) { /* better match => select a better chain */
assert(lookBackLength==0); /* search forward only */
- if (matchIndex + longest <= ipIndex) {
+ if (matchIndex + (U32)longest <= ipIndex) {
+ int const kTrigger = 4;
U32 distanceToNextMatch = 1;
+ int const end = longest - MINMATCH + 1;
+ int step = 1;
+ int accel = 1 << kTrigger;
int pos;
- for (pos = 0; pos <= longest - MINMATCH; pos++) {
- U32 const candidateDist = DELTANEXTU16(chainTable, matchIndex + pos);
+ for (pos = 0; pos < end; pos += step) {
+ U32 const candidateDist = DELTANEXTU16(chainTable, matchIndex + (U32)pos);
+ step = (accel++ >> kTrigger);
if (candidateDist > distanceToNextMatch) {
distanceToNextMatch = candidateDist;
- matchChainPos = pos;
- } }
+ matchChainPos = (U32)pos;
+ accel = 1 << kTrigger;
+ }
+ }
if (distanceToNextMatch > 1) {
if (distanceToNextMatch > matchIndex) break; /* avoid overflow */
matchIndex -= distanceToNextMatch;
@@ -310,49 +346,78 @@ LZ4HC_InsertAndGetWiderMatch (
} else {
repeat = rep_not;
} }
- if ( (repeat == rep_confirmed)
- && (matchCandidateIdx >= dictLimit) ) { /* same segment only */
- const BYTE* const matchPtr = base + matchCandidateIdx;
+ if ( (repeat == rep_confirmed) && (matchCandidateIdx >= lowestMatchIndex)
+ && LZ4HC_protectDictEnd(dictLimit, matchCandidateIdx) ) {
+ const int extDict = matchCandidateIdx < dictLimit;
+ const BYTE* const matchPtr = (extDict ? dictBase : base) + matchCandidateIdx;
if (LZ4_read32(matchPtr) == pattern) { /* good candidate */
- size_t const forwardPatternLength = LZ4HC_countPattern(matchPtr+sizeof(pattern), iHighLimit, pattern) + sizeof(pattern);
- const BYTE* const lowestMatchPtr = (lowPrefixPtr + MAX_DISTANCE >= ip) ? lowPrefixPtr : ip - MAX_DISTANCE;
- size_t const backLength = LZ4HC_reverseCountPattern(matchPtr, lowestMatchPtr, pattern);
- size_t const currentSegmentLength = backLength + forwardPatternLength;
-
- if ( (currentSegmentLength >= srcPatternLength) /* current pattern segment large enough to contain full srcPatternLength */
- && (forwardPatternLength <= srcPatternLength) ) { /* haven't reached this position yet */
- matchIndex = matchCandidateIdx + (U32)forwardPatternLength - (U32)srcPatternLength; /* best position, full pattern, might be followed by more match */
- } else {
- matchIndex = matchCandidateIdx - (U32)backLength; /* farthest position in current segment, will find a match of length currentSegmentLength + maybe some back */
- if (lookBackLength==0) { /* no back possible */
- size_t const maxML = MIN(currentSegmentLength, srcPatternLength);
- if ((size_t)longest < maxML) {
- assert(base + matchIndex < ip);
- if (ip - (base+matchIndex) > MAX_DISTANCE) break;
- assert(maxML < 2 GB);
- longest = (int)maxML;
- *matchpos = base + matchIndex; /* virtual pos, relative to ip, to retrieve offset */
- *startpos = ip;
+ const BYTE* const dictStart = dictBase + hc4->lowLimit;
+ const BYTE* const iLimit = extDict ? dictBase + dictLimit : iHighLimit;
+ size_t forwardPatternLength = LZ4HC_countPattern(matchPtr+sizeof(pattern), iLimit, pattern) + sizeof(pattern);
+ if (extDict && matchPtr + forwardPatternLength == iLimit) {
+ U32 const rotatedPattern = LZ4HC_rotatePattern(forwardPatternLength, pattern);
+ forwardPatternLength += LZ4HC_countPattern(lowPrefixPtr, iHighLimit, rotatedPattern);
+ }
+ { const BYTE* const lowestMatchPtr = extDict ? dictStart : lowPrefixPtr;
+ size_t backLength = LZ4HC_reverseCountPattern(matchPtr, lowestMatchPtr, pattern);
+ size_t currentSegmentLength;
+ if (!extDict && matchPtr - backLength == lowPrefixPtr && hc4->lowLimit < dictLimit) {
+ U32 const rotatedPattern = LZ4HC_rotatePattern((U32)(-(int)backLength), pattern);
+ backLength += LZ4HC_reverseCountPattern(dictBase + dictLimit, dictStart, rotatedPattern);
+ }
+ /* Limit backLength not go further than lowestMatchIndex */
+ backLength = matchCandidateIdx - MAX(matchCandidateIdx - (U32)backLength, lowestMatchIndex);
+ assert(matchCandidateIdx - backLength >= lowestMatchIndex);
+ currentSegmentLength = backLength + forwardPatternLength;
+ /* Adjust to end of pattern if the source pattern fits, otherwise the beginning of the pattern */
+ if ( (currentSegmentLength >= srcPatternLength) /* current pattern segment large enough to contain full srcPatternLength */
+ && (forwardPatternLength <= srcPatternLength) ) { /* haven't reached this position yet */
+ U32 const newMatchIndex = matchCandidateIdx + (U32)forwardPatternLength - (U32)srcPatternLength; /* best position, full pattern, might be followed by more match */
+ if (LZ4HC_protectDictEnd(dictLimit, newMatchIndex))
+ matchIndex = newMatchIndex;
+ else {
+ /* Can only happen if started in the prefix */
+ assert(newMatchIndex >= dictLimit - 3 && newMatchIndex < dictLimit && !extDict);
+ matchIndex = dictLimit;
}
- { U32 const distToNextPattern = DELTANEXTU16(chainTable, matchIndex);
- if (distToNextPattern > matchIndex) break; /* avoid overflow */
- matchIndex -= distToNextPattern;
- } } }
+ } else {
+ U32 const newMatchIndex = matchCandidateIdx - (U32)backLength; /* farthest position in current segment, will find a match of length currentSegmentLength + maybe some back */
+ if (!LZ4HC_protectDictEnd(dictLimit, newMatchIndex)) {
+ assert(newMatchIndex >= dictLimit - 3 && newMatchIndex < dictLimit && !extDict);
+ matchIndex = dictLimit;
+ } else {
+ matchIndex = newMatchIndex;
+ if (lookBackLength==0) { /* no back possible */
+ size_t const maxML = MIN(currentSegmentLength, srcPatternLength);
+ if ((size_t)longest < maxML) {
+ assert(base + matchIndex != ip);
+ if ((size_t)(ip - base) - matchIndex > LZ4_DISTANCE_MAX) break;
+ assert(maxML < 2 GB);
+ longest = (int)maxML;
+ *matchpos = base + matchIndex; /* virtual pos, relative to ip, to retrieve offset */
+ *startpos = ip;
+ }
+ { U32 const distToNextPattern = DELTANEXTU16(chainTable, matchIndex);
+ if (distToNextPattern > matchIndex) break; /* avoid overflow */
+ matchIndex -= distToNextPattern;
+ } } } } }
continue;
} }
} } /* PA optimization */
/* follow current chain */
- matchIndex -= DELTANEXTU16(chainTable, matchIndex+matchChainPos);
+ matchIndex -= DELTANEXTU16(chainTable, matchIndex + matchChainPos);
} /* while ((matchIndex>=lowestMatchIndex) && (nbAttempts)) */
- if (dict == usingDictCtx && nbAttempts && ipIndex - lowestMatchIndex < MAX_DISTANCE) {
- size_t const dictEndOffset = dictCtx->end - dictCtx->base;
+ if ( dict == usingDictCtxHc
+ && nbAttempts > 0
+ && ipIndex - lowestMatchIndex < LZ4_DISTANCE_MAX) {
+ size_t const dictEndOffset = (size_t)(dictCtx->end - dictCtx->base);
+ U32 dictMatchIndex = dictCtx->hashTable[LZ4HC_hashPtr(ip)];
assert(dictEndOffset <= 1 GB);
- dictMatchIndex = dictCtx->hashTable[LZ4HC_hashPtr(ip)];
matchIndex = dictMatchIndex + lowestMatchIndex - (U32)dictEndOffset;
- while (ipIndex - matchIndex <= MAX_DISTANCE && nbAttempts--) {
+ while (ipIndex - matchIndex <= LZ4_DISTANCE_MAX && nbAttempts--) {
const BYTE* const matchPtr = dictCtx->base + dictMatchIndex;
if (LZ4_read32(matchPtr) == pattern) {
@@ -360,22 +425,19 @@ LZ4HC_InsertAndGetWiderMatch (
int back = 0;
const BYTE* vLimit = ip + (dictEndOffset - dictMatchIndex);
if (vLimit > iHighLimit) vLimit = iHighLimit;
- mlt = LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, vLimit) + MINMATCH;
+ mlt = (int)LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, vLimit) + MINMATCH;
back = lookBackLength ? LZ4HC_countBack(ip, matchPtr, iLowLimit, dictCtx->base + dictCtx->dictLimit) : 0;
mlt -= back;
if (mlt > longest) {
longest = mlt;
*matchpos = base + matchIndex + back;
*startpos = ip + back;
- }
- }
+ } }
{ U32 const nextOffset = DELTANEXTU16(dictCtx->chainTable, dictMatchIndex);
dictMatchIndex -= nextOffset;
matchIndex -= nextOffset;
- }
- }
- }
+ } } }
return longest;
}
@@ -395,86 +457,94 @@ int LZ4HC_InsertAndFindBestMatch(LZ4HC_CCtx_internal* const hc4, /* Index tabl
return LZ4HC_InsertAndGetWiderMatch(hc4, ip, ip, iLimit, MINMATCH-1, matchpos, &uselessPtr, maxNbAttempts, patternAnalysis, 0 /*chainSwap*/, dict, favorCompressionRatio);
}
-
-
-typedef enum {
- noLimit = 0,
- limitedOutput = 1,
- limitedDestSize = 2,
-} limitedOutput_directive;
-
/* LZ4HC_encodeSequence() :
* @return : 0 if ok,
* 1 if buffer issue detected */
LZ4_FORCE_INLINE int LZ4HC_encodeSequence (
- const BYTE** ip,
- BYTE** op,
- const BYTE** anchor,
+ const BYTE** _ip,
+ BYTE** _op,
+ const BYTE** _anchor,
int matchLength,
const BYTE* const match,
limitedOutput_directive limit,
BYTE* oend)
{
+#define ip (*_ip)
+#define op (*_op)
+#define anchor (*_anchor)
+
size_t length;
- BYTE* const token = (*op)++;
+ BYTE* const token = op++;
#if defined(LZ4_DEBUG) && (LZ4_DEBUG >= 6)
static const BYTE* start = NULL;
static U32 totalCost = 0;
- U32 const pos = (start==NULL) ? 0 : (U32)(*anchor - start);
- U32 const ll = (U32)(*ip - *anchor);
+ U32 const pos = (start==NULL) ? 0 : (U32)(anchor - start);
+ U32 const ll = (U32)(ip - anchor);
U32 const llAdd = (ll>=15) ? ((ll-15) / 255) + 1 : 0;
U32 const mlAdd = (matchLength>=19) ? ((matchLength-19) / 255) + 1 : 0;
U32 const cost = 1 + llAdd + ll + 2 + mlAdd;
- if (start==NULL) start = *anchor; /* only works for single segment */
+ if (start==NULL) start = anchor; /* only works for single segment */
/* g_debuglog_enable = (pos >= 2228) & (pos <= 2262); */
- DEBUGLOG(6, "pos:%7u -- literals:%3u, match:%4i, offset:%5u, cost:%3u + %u",
+ DEBUGLOG(6, "pos:%7u -- literals:%4u, match:%4i, offset:%5u, cost:%4u + %5u",
pos,
- (U32)(*ip - *anchor), matchLength, (U32)(*ip-match),
+ (U32)(ip - anchor), matchLength, (U32)(ip-match),
cost, totalCost);
totalCost += cost;
#endif
/* Encode Literal length */
- length = (size_t)(*ip - *anchor);
- if ((limit) && ((*op + (length >> 8) + length + (2 + 1 + LASTLITERALS)) > oend)) return 1; /* Check output limit */
+ length = (size_t)(ip - anchor);
+ LZ4_STATIC_ASSERT(notLimited == 0);
+ /* Check output limit */
+ if (limit && ((op + (length / 255) + length + (2 + 1 + LASTLITERALS)) > oend)) {
+ DEBUGLOG(6, "Not enough room to write %i literals (%i bytes remaining)",
+ (int)length, (int)(oend - op));
+ return 1;
+ }
if (length >= RUN_MASK) {
size_t len = length - RUN_MASK;
*token = (RUN_MASK << ML_BITS);
- for(; len >= 255 ; len -= 255) *(*op)++ = 255;
- *(*op)++ = (BYTE)len;
+ for(; len >= 255 ; len -= 255) *op++ = 255;
+ *op++ = (BYTE)len;
} else {
*token = (BYTE)(length << ML_BITS);
}
/* Copy Literals */
- LZ4_wildCopy(*op, *anchor, (*op) + length);
- *op += length;
+ LZ4_wildCopy8(op, anchor, op + length);
+ op += length;
/* Encode Offset */
- assert( (*ip - match) <= MAX_DISTANCE ); /* note : consider providing offset as a value, rather than as a pointer difference */
- LZ4_writeLE16(*op, (U16)(*ip-match)); *op += 2;
+ assert( (ip - match) <= LZ4_DISTANCE_MAX ); /* note : consider providing offset as a value, rather than as a pointer difference */
+ LZ4_writeLE16(op, (U16)(ip - match)); op += 2;
/* Encode MatchLength */
assert(matchLength >= MINMATCH);
- length = (size_t)(matchLength - MINMATCH);
- if ((limit) && (*op + (length >> 8) + (1 + LASTLITERALS) > oend)) return 1; /* Check output limit */
+ length = (size_t)matchLength - MINMATCH;
+ if (limit && (op + (length / 255) + (1 + LASTLITERALS) > oend)) {
+ DEBUGLOG(6, "Not enough room to write match length");
+ return 1; /* Check output limit */
+ }
if (length >= ML_MASK) {
*token += ML_MASK;
length -= ML_MASK;
- for(; length >= 510 ; length -= 510) { *(*op)++ = 255; *(*op)++ = 255; }
- if (length >= 255) { length -= 255; *(*op)++ = 255; }
- *(*op)++ = (BYTE)length;
+ for(; length >= 510 ; length -= 510) { *op++ = 255; *op++ = 255; }
+ if (length >= 255) { length -= 255; *op++ = 255; }
+ *op++ = (BYTE)length;
} else {
*token += (BYTE)(length);
}
/* Prepare next loop */
- *ip += matchLength;
- *anchor = *ip;
+ ip += matchLength;
+ anchor = ip;
return 0;
}
+#undef ip
+#undef op
+#undef anchor
LZ4_FORCE_INLINE int LZ4HC_compress_hashChain (
LZ4HC_CCtx_internal* const ctx,
@@ -482,7 +552,7 @@ LZ4_FORCE_INLINE int LZ4HC_compress_hashChain (
char* const dest,
int* srcSizePtr,
int const maxOutputSize,
- unsigned maxNbAttempts,
+ int maxNbAttempts,
const limitedOutput_directive limit,
const dictCtx_directive dict
)
@@ -511,12 +581,12 @@ LZ4_FORCE_INLINE int LZ4HC_compress_hashChain (
/* init */
*srcSizePtr = 0;
- if (limit == limitedDestSize) oend -= LASTLITERALS; /* Hack for support LZ4 format restriction */
- if (inputSize < LZ4_minLength) goto _last_literals; /* Input too small, no compression (all literals) */
+ if (limit == fillOutput) oend -= LASTLITERALS; /* Hack for support LZ4 format restriction */
+ if (inputSize < LZ4_minLength) goto _last_literals; /* Input too small, no compression (all literals) */
/* Main Loop */
while (ip <= mflimit) {
- ml = LZ4HC_InsertAndFindBestMatch (ctx, ip, matchlimit, &ref, maxNbAttempts, patternAnalysis, dict);
+ ml = LZ4HC_InsertAndFindBestMatch(ctx, ip, matchlimit, &ref, maxNbAttempts, patternAnalysis, dict);
if (ml<MINMATCH) { ip++; continue; }
/* saved, in case we would skip too much */
@@ -533,7 +603,7 @@ _Search2:
if (ml2 == ml) { /* No better match => encode ML1 */
optr = op;
- if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) goto _dest_overflow;
+ if (LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ml, ref, limit, oend)) goto _dest_overflow;
continue;
}
@@ -581,10 +651,14 @@ _Search3:
if (start2 < ip+ml) ml = (int)(start2 - ip);
/* Now, encode 2 sequences */
optr = op;
- if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) goto _dest_overflow;
+ if (LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ml, ref, limit, oend)) goto _dest_overflow;
ip = start2;
optr = op;
- if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml2, ref2, limit, oend)) goto _dest_overflow;
+ if (LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ml2, ref2, limit, oend)) {
+ ml = ml2;
+ ref = ref2;
+ goto _dest_overflow;
+ }
continue;
}
@@ -603,7 +677,7 @@ _Search3:
}
optr = op;
- if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) goto _dest_overflow;
+ if (LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ml, ref, limit, oend)) goto _dest_overflow;
ip = start3;
ref = ref3;
ml = ml3;
@@ -641,7 +715,7 @@ _Search3:
}
}
optr = op;
- if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) goto _dest_overflow;
+ if (LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ml, ref, limit, oend)) goto _dest_overflow;
/* ML2 becomes ML1 */
ip = start2; ref = ref2; ml = ml2;
@@ -656,17 +730,18 @@ _Search3:
_last_literals:
/* Encode Last Literals */
{ size_t lastRunSize = (size_t)(iend - anchor); /* literals */
- size_t litLength = (lastRunSize + 255 - RUN_MASK) / 255;
- size_t const totalSize = 1 + litLength + lastRunSize;
- if (limit == limitedDestSize) oend += LASTLITERALS; /* restore correct value */
+ size_t llAdd = (lastRunSize + 255 - RUN_MASK) / 255;
+ size_t const totalSize = 1 + llAdd + lastRunSize;
+ if (limit == fillOutput) oend += LASTLITERALS; /* restore correct value */
if (limit && (op + totalSize > oend)) {
- if (limit == limitedOutput) return 0; /* Check output limit */
+ if (limit == limitedOutput) return 0;
/* adapt lastRunSize to fill 'dest' */
- lastRunSize = (size_t)(oend - op) - 1;
- litLength = (lastRunSize + 255 - RUN_MASK) / 255;
- lastRunSize -= litLength;
+ lastRunSize = (size_t)(oend - op) - 1 /*token*/;
+ llAdd = (lastRunSize + 256 - RUN_MASK) / 256;
+ lastRunSize -= llAdd;
}
- ip = anchor + lastRunSize;
+ DEBUGLOG(6, "Final literal run : %i literals", (int)lastRunSize);
+ ip = anchor + lastRunSize; /* can be != iend if limit==fillOutput */
if (lastRunSize >= RUN_MASK) {
size_t accumulator = lastRunSize - RUN_MASK;
@@ -685,10 +760,26 @@ _last_literals:
return (int) (((char*)op)-dest);
_dest_overflow:
- if (limit == limitedDestSize) {
+ if (limit == fillOutput) {
+ /* Assumption : ip, anchor, ml and ref must be set correctly */
+ size_t const ll = (size_t)(ip - anchor);
+ size_t const ll_addbytes = (ll + 240) / 255;
+ size_t const ll_totalCost = 1 + ll_addbytes + ll;
+ BYTE* const maxLitPos = oend - 3; /* 2 for offset, 1 for token */
+ DEBUGLOG(6, "Last sequence overflowing");
op = optr; /* restore correct out pointer */
+ if (op + ll_totalCost <= maxLitPos) {
+ /* ll validated; now adjust match length */
+ size_t const bytesLeftForMl = (size_t)(maxLitPos - (op+ll_totalCost));
+ size_t const maxMlSize = MINMATCH + (ML_MASK-1) + (bytesLeftForMl * 255);
+ assert(maxMlSize < INT_MAX); assert(ml >= 0);
+ if ((size_t)ml > maxMlSize) ml = (int)maxMlSize;
+ if ((oend + LASTLITERALS) - (op + ll_totalCost + 2) - 1 + ml >= MFLIMIT) {
+ LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ml, ref, notLimited, oend);
+ } }
goto _last_literals;
}
+ /* compression failed */
return 0;
}
@@ -699,7 +790,7 @@ static int LZ4HC_compress_optimal( LZ4HC_CCtx_internal* ctx,
int const nbSearches, size_t sufficient_len,
const limitedOutput_directive limit, int const fullUpdate,
const dictCtx_directive dict,
- HCfavor_e favorDecSpeed);
+ const HCfavor_e favorDecSpeed);
LZ4_FORCE_INLINE int LZ4HC_compress_generic_internal (
@@ -716,7 +807,7 @@ LZ4_FORCE_INLINE int LZ4HC_compress_generic_internal (
typedef enum { lz4hc, lz4opt } lz4hc_strat_e;
typedef struct {
lz4hc_strat_e strat;
- U32 nbSearches;
+ int nbSearches;
U32 targetLength;
} cParams_t;
static const cParams_t clTable[LZ4HC_CLEVEL_MAX+1] = {
@@ -735,56 +826,65 @@ LZ4_FORCE_INLINE int LZ4HC_compress_generic_internal (
{ lz4opt,16384,LZ4_OPT_NUM }, /* 12==LZ4HC_CLEVEL_MAX */
};
- DEBUGLOG(4, "LZ4HC_compress_generic(%p, %p, %d)", ctx, src, *srcSizePtr);
+ DEBUGLOG(4, "LZ4HC_compress_generic(ctx=%p, src=%p, srcSize=%d, limit=%d)",
+ ctx, src, *srcSizePtr, limit);
- if (limit == limitedDestSize && dstCapacity < 1) return 0; /* Impossible to store anything */
- if ((U32)*srcSizePtr > (U32)LZ4_MAX_INPUT_SIZE) return 0; /* Unsupported input size (too large or negative) */
+ if (limit == fillOutput && dstCapacity < 1) return 0; /* Impossible to store anything */
+ if ((U32)*srcSizePtr > (U32)LZ4_MAX_INPUT_SIZE) return 0; /* Unsupported input size (too large or negative) */
ctx->end += *srcSizePtr;
if (cLevel < 1) cLevel = LZ4HC_CLEVEL_DEFAULT; /* note : convention is different from lz4frame, maybe something to review */
cLevel = MIN(LZ4HC_CLEVEL_MAX, cLevel);
{ cParams_t const cParam = clTable[cLevel];
HCfavor_e const favor = ctx->favorDecSpeed ? favorDecompressionSpeed : favorCompressionRatio;
- if (cParam.strat == lz4hc)
- return LZ4HC_compress_hashChain(ctx,
+ int result;
+
+ if (cParam.strat == lz4hc) {
+ result = LZ4HC_compress_hashChain(ctx,
src, dst, srcSizePtr, dstCapacity,
cParam.nbSearches, limit, dict);
- assert(cParam.strat == lz4opt);
- return LZ4HC_compress_optimal(ctx,
- src, dst, srcSizePtr, dstCapacity,
- cParam.nbSearches, cParam.targetLength, limit,
- cLevel == LZ4HC_CLEVEL_MAX, /* ultra mode */
- dict, favor);
+ } else {
+ assert(cParam.strat == lz4opt);
+ result = LZ4HC_compress_optimal(ctx,
+ src, dst, srcSizePtr, dstCapacity,
+ cParam.nbSearches, cParam.targetLength, limit,
+ cLevel == LZ4HC_CLEVEL_MAX, /* ultra mode */
+ dict, favor);
+ }
+ if (result <= 0) ctx->dirty = 1;
+ return result;
}
}
static void LZ4HC_setExternalDict(LZ4HC_CCtx_internal* ctxPtr, const BYTE* newBlock);
-static int LZ4HC_compress_generic_noDictCtx (
- LZ4HC_CCtx_internal* const ctx,
- const char* const src,
- char* const dst,
- int* const srcSizePtr,
- int const dstCapacity,
- int cLevel,
- limitedOutput_directive limit
- )
+static int
+LZ4HC_compress_generic_noDictCtx (
+ LZ4HC_CCtx_internal* const ctx,
+ const char* const src,
+ char* const dst,
+ int* const srcSizePtr,
+ int const dstCapacity,
+ int cLevel,
+ limitedOutput_directive limit
+ )
{
assert(ctx->dictCtx == NULL);
return LZ4HC_compress_generic_internal(ctx, src, dst, srcSizePtr, dstCapacity, cLevel, limit, noDictCtx);
}
-static int LZ4HC_compress_generic_dictCtx (
- LZ4HC_CCtx_internal* const ctx,
- const char* const src,
- char* const dst,
- int* const srcSizePtr,
- int const dstCapacity,
- int cLevel,
- limitedOutput_directive limit
- )
+static int
+LZ4HC_compress_generic_dictCtx (
+ LZ4HC_CCtx_internal* const ctx,
+ const char* const src,
+ char* const dst,
+ int* const srcSizePtr,
+ int const dstCapacity,
+ int cLevel,
+ limitedOutput_directive limit
+ )
{
- const size_t position = ctx->end - ctx->base - ctx->lowLimit;
+ const size_t position = (size_t)(ctx->end - ctx->base) - ctx->lowLimit;
assert(ctx->dictCtx != NULL);
if (position >= 64 KB) {
ctx->dictCtx = NULL;
@@ -795,19 +895,20 @@ static int LZ4HC_compress_generic_dictCtx (
ctx->compressionLevel = (short)cLevel;
return LZ4HC_compress_generic_noDictCtx(ctx, src, dst, srcSizePtr, dstCapacity, cLevel, limit);
} else {
- return LZ4HC_compress_generic_internal(ctx, src, dst, srcSizePtr, dstCapacity, cLevel, limit, usingDictCtx);
+ return LZ4HC_compress_generic_internal(ctx, src, dst, srcSizePtr, dstCapacity, cLevel, limit, usingDictCtxHc);
}
}
-static int LZ4HC_compress_generic (
- LZ4HC_CCtx_internal* const ctx,
- const char* const src,
- char* const dst,
- int* const srcSizePtr,
- int const dstCapacity,
- int cLevel,
- limitedOutput_directive limit
- )
+static int
+LZ4HC_compress_generic (
+ LZ4HC_CCtx_internal* const ctx,
+ const char* const src,
+ char* const dst,
+ int* const srcSizePtr,
+ int const dstCapacity,
+ int cLevel,
+ limitedOutput_directive limit
+ )
{
if (ctx->dictCtx == NULL) {
return LZ4HC_compress_generic_noDictCtx(ctx, src, dst, srcSizePtr, dstCapacity, cLevel, limit);
@@ -817,24 +918,36 @@ static int LZ4HC_compress_generic (
}
-int LZ4_sizeofStateHC(void) { return sizeof(LZ4_streamHC_t); }
+int LZ4_sizeofStateHC(void) { return (int)sizeof(LZ4_streamHC_t); }
+
+static size_t LZ4_streamHC_t_alignment(void)
+{
+#if LZ4_ALIGN_TEST
+ typedef struct { char c; LZ4_streamHC_t t; } t_a;
+ return sizeof(t_a) - sizeof(LZ4_streamHC_t);
+#else
+ return 1; /* effectively disabled */
+#endif
+}
+/* state is presumed correctly initialized,
+ * in which case its size and alignment have already been validate */
int LZ4_compress_HC_extStateHC_fastReset (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int compressionLevel)
{
LZ4HC_CCtx_internal* const ctx = &((LZ4_streamHC_t*)state)->internal_donotuse;
- if (((size_t)(state)&(sizeof(void*)-1)) != 0) return 0; /* Error : state is not aligned for pointers (32 or 64 bits) */
+ if (!LZ4_isAligned(state, LZ4_streamHC_t_alignment())) return 0;
LZ4_resetStreamHC_fast((LZ4_streamHC_t*)state, compressionLevel);
- LZ4HC_init (ctx, (const BYTE*)src);
+ LZ4HC_init_internal (ctx, (const BYTE*)src);
if (dstCapacity < LZ4_compressBound(srcSize))
return LZ4HC_compress_generic (ctx, src, dst, &srcSize, dstCapacity, compressionLevel, limitedOutput);
else
- return LZ4HC_compress_generic (ctx, src, dst, &srcSize, dstCapacity, compressionLevel, noLimit);
+ return LZ4HC_compress_generic (ctx, src, dst, &srcSize, dstCapacity, compressionLevel, notLimited);
}
int LZ4_compress_HC_extStateHC (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int compressionLevel)
{
- if (((size_t)(state)&(sizeof(void*)-1)) != 0) return 0; /* Error : state is not aligned for pointers (32 or 64 bits) */
- LZ4_resetStreamHC ((LZ4_streamHC_t*)state, compressionLevel);
+ LZ4_streamHC_t* const ctx = LZ4_initStreamHC(state, sizeof(*ctx));
+ if (ctx==NULL) return 0; /* init failure */
return LZ4_compress_HC_extStateHC_fastReset(state, src, dst, srcSize, dstCapacity, compressionLevel);
}
@@ -848,19 +961,19 @@ int LZ4_compress_HC(const char* src, char* dst, int srcSize, int dstCapacity, in
#endif
int const cSize = LZ4_compress_HC_extStateHC(statePtr, src, dst, srcSize, dstCapacity, compressionLevel);
#if defined(LZ4HC_HEAPMODE) && LZ4HC_HEAPMODE==1
- free(statePtr);
+ FREEMEM(statePtr);
#endif
return cSize;
}
-/* LZ4_compress_HC_destSize() :
- * only compatible with regular HC parser */
-int LZ4_compress_HC_destSize(void* LZ4HC_Data, const char* source, char* dest, int* sourceSizePtr, int targetDestSize, int cLevel)
+/* state is presumed sized correctly (>= sizeof(LZ4_streamHC_t)) */
+int LZ4_compress_HC_destSize(void* state, const char* source, char* dest, int* sourceSizePtr, int targetDestSize, int cLevel)
{
- LZ4HC_CCtx_internal* const ctx = &((LZ4_streamHC_t*)LZ4HC_Data)->internal_donotuse;
- LZ4_resetStreamHC((LZ4_streamHC_t*)LZ4HC_Data, cLevel);
- LZ4HC_init(ctx, (const BYTE*) source);
- return LZ4HC_compress_generic(ctx, source, dest, sourceSizePtr, targetDestSize, cLevel, limitedDestSize);
+ LZ4_streamHC_t* const ctx = LZ4_initStreamHC(state, sizeof(*ctx));
+ if (ctx==NULL) return 0; /* init failure */
+ LZ4HC_init_internal(&ctx->internal_donotuse, (const BYTE*) source);
+ LZ4_setCompressionLevel(ctx, cLevel);
+ return LZ4HC_compress_generic(&ctx->internal_donotuse, source, dest, sourceSizePtr, targetDestSize, cLevel, fillOutput);
}
@@ -869,44 +982,65 @@ int LZ4_compress_HC_destSize(void* LZ4HC_Data, const char* source, char* dest, i
* Streaming Functions
**************************************/
/* allocation */
-LZ4_streamHC_t* LZ4_createStreamHC(void) {
- LZ4_streamHC_t* const LZ4_streamHCPtr = (LZ4_streamHC_t*)ALLOC(sizeof(LZ4_streamHC_t));
- if (LZ4_streamHCPtr==NULL) return NULL;
- LZ4_resetStreamHC(LZ4_streamHCPtr, LZ4HC_CLEVEL_DEFAULT);
- return LZ4_streamHCPtr;
+LZ4_streamHC_t* LZ4_createStreamHC(void)
+{
+ LZ4_streamHC_t* const state =
+ (LZ4_streamHC_t*)ALLOC_AND_ZERO(sizeof(LZ4_streamHC_t));
+ if (state == NULL) return NULL;
+ LZ4_setCompressionLevel(state, LZ4HC_CLEVEL_DEFAULT);
+ return state;
}
-int LZ4_freeStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr) {
+int LZ4_freeStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr)
+{
DEBUGLOG(4, "LZ4_freeStreamHC(%p)", LZ4_streamHCPtr);
if (!LZ4_streamHCPtr) return 0; /* support free on NULL */
- free(LZ4_streamHCPtr);
+ FREEMEM(LZ4_streamHCPtr);
return 0;
}
-/* initialization */
+LZ4_streamHC_t* LZ4_initStreamHC (void* buffer, size_t size)
+{
+ LZ4_streamHC_t* const LZ4_streamHCPtr = (LZ4_streamHC_t*)buffer;
+ /* if compilation fails here, LZ4_STREAMHCSIZE must be increased */
+ LZ4_STATIC_ASSERT(sizeof(LZ4HC_CCtx_internal) <= LZ4_STREAMHCSIZE);
+ DEBUGLOG(4, "LZ4_initStreamHC(%p, %u)", buffer, (unsigned)size);
+ /* check conditions */
+ if (buffer == NULL) return NULL;
+ if (size < sizeof(LZ4_streamHC_t)) return NULL;
+ if (!LZ4_isAligned(buffer, LZ4_streamHC_t_alignment())) return NULL;
+ /* init */
+ { LZ4HC_CCtx_internal* const hcstate = &(LZ4_streamHCPtr->internal_donotuse);
+ MEM_INIT(hcstate, 0, sizeof(*hcstate)); }
+ LZ4_setCompressionLevel(LZ4_streamHCPtr, LZ4HC_CLEVEL_DEFAULT);
+ return LZ4_streamHCPtr;
+}
+
+/* just a stub */
void LZ4_resetStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel)
{
- LZ4_STATIC_ASSERT(sizeof(LZ4HC_CCtx_internal) <= sizeof(size_t) * LZ4_STREAMHCSIZE_SIZET); /* if compilation fails here, LZ4_STREAMHCSIZE must be increased */
- DEBUGLOG(4, "LZ4_resetStreamHC(%p, %d)", LZ4_streamHCPtr, compressionLevel);
- LZ4_streamHCPtr->internal_donotuse.end = (const BYTE *)(ptrdiff_t)-1;
- LZ4_streamHCPtr->internal_donotuse.base = NULL;
- LZ4_streamHCPtr->internal_donotuse.dictCtx = NULL;
- LZ4_streamHCPtr->internal_donotuse.favorDecSpeed = 0;
+ LZ4_initStreamHC(LZ4_streamHCPtr, sizeof(*LZ4_streamHCPtr));
LZ4_setCompressionLevel(LZ4_streamHCPtr, compressionLevel);
}
void LZ4_resetStreamHC_fast (LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel)
{
DEBUGLOG(4, "LZ4_resetStreamHC_fast(%p, %d)", LZ4_streamHCPtr, compressionLevel);
- LZ4_streamHCPtr->internal_donotuse.end -= (uptrval)LZ4_streamHCPtr->internal_donotuse.base;
- LZ4_streamHCPtr->internal_donotuse.base = NULL;
- LZ4_streamHCPtr->internal_donotuse.dictCtx = NULL;
+ if (LZ4_streamHCPtr->internal_donotuse.dirty) {
+ LZ4_initStreamHC(LZ4_streamHCPtr, sizeof(*LZ4_streamHCPtr));
+ } else {
+ /* preserve end - base : can trigger clearTable's threshold */
+ LZ4_streamHCPtr->internal_donotuse.end -= (uptrval)LZ4_streamHCPtr->internal_donotuse.base;
+ LZ4_streamHCPtr->internal_donotuse.base = NULL;
+ LZ4_streamHCPtr->internal_donotuse.dictCtx = NULL;
+ }
LZ4_setCompressionLevel(LZ4_streamHCPtr, compressionLevel);
}
void LZ4_setCompressionLevel(LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel)
{
+ DEBUGLOG(5, "LZ4_setCompressionLevel(%p, %d)", LZ4_streamHCPtr, compressionLevel);
if (compressionLevel < 1) compressionLevel = LZ4HC_CLEVEL_DEFAULT;
if (compressionLevel > LZ4HC_CLEVEL_MAX) compressionLevel = LZ4HC_CLEVEL_MAX;
LZ4_streamHCPtr->internal_donotuse.compressionLevel = (short)compressionLevel;
@@ -917,16 +1051,24 @@ void LZ4_favorDecompressionSpeed(LZ4_streamHC_t* LZ4_streamHCPtr, int favor)
LZ4_streamHCPtr->internal_donotuse.favorDecSpeed = (favor!=0);
}
-int LZ4_loadDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, const char* dictionary, int dictSize)
+/* LZ4_loadDictHC() :
+ * LZ4_streamHCPtr is presumed properly initialized */
+int LZ4_loadDictHC (LZ4_streamHC_t* LZ4_streamHCPtr,
+ const char* dictionary, int dictSize)
{
LZ4HC_CCtx_internal* const ctxPtr = &LZ4_streamHCPtr->internal_donotuse;
- DEBUGLOG(4, "LZ4_loadDictHC(%p, %p, %d)", LZ4_streamHCPtr, dictionary, dictSize);
+ DEBUGLOG(4, "LZ4_loadDictHC(ctx:%p, dict:%p, dictSize:%d)", LZ4_streamHCPtr, dictionary, dictSize);
+ assert(LZ4_streamHCPtr != NULL);
if (dictSize > 64 KB) {
- dictionary += dictSize - 64 KB;
+ dictionary += (size_t)dictSize - 64 KB;
dictSize = 64 KB;
}
- LZ4_resetStreamHC(LZ4_streamHCPtr, ctxPtr->compressionLevel);
- LZ4HC_init (ctxPtr, (const BYTE*)dictionary);
+ /* need a full initialization, there are bad side-effects when using resetFast() */
+ { int const cLevel = ctxPtr->compressionLevel;
+ LZ4_initStreamHC(LZ4_streamHCPtr, sizeof(*LZ4_streamHCPtr));
+ LZ4_setCompressionLevel(LZ4_streamHCPtr, cLevel);
+ }
+ LZ4HC_init_internal (ctxPtr, (const BYTE*)dictionary);
ctxPtr->end = (const BYTE*)dictionary + dictSize;
if (dictSize >= 4) LZ4HC_Insert (ctxPtr, ctxPtr->end-3);
return dictSize;
@@ -951,17 +1093,23 @@ static void LZ4HC_setExternalDict(LZ4HC_CCtx_internal* ctxPtr, const BYTE* newBl
ctxPtr->base = newBlock - ctxPtr->dictLimit;
ctxPtr->end = newBlock;
ctxPtr->nextToUpdate = ctxPtr->dictLimit; /* match referencing will resume from there */
+
+ /* cannot reference an extDict and a dictCtx at the same time */
+ ctxPtr->dictCtx = NULL;
}
-static int LZ4_compressHC_continue_generic (LZ4_streamHC_t* LZ4_streamHCPtr,
- const char* src, char* dst,
- int* srcSizePtr, int dstCapacity,
- limitedOutput_directive limit)
+static int
+LZ4_compressHC_continue_generic (LZ4_streamHC_t* LZ4_streamHCPtr,
+ const char* src, char* dst,
+ int* srcSizePtr, int dstCapacity,
+ limitedOutput_directive limit)
{
LZ4HC_CCtx_internal* const ctxPtr = &LZ4_streamHCPtr->internal_donotuse;
- DEBUGLOG(4, "LZ4_compressHC_continue_generic(%p, %p, %d)", LZ4_streamHCPtr, src, *srcSizePtr);
+ DEBUGLOG(5, "LZ4_compressHC_continue_generic(ctx=%p, src=%p, srcSize=%d, limit=%d)",
+ LZ4_streamHCPtr, src, *srcSizePtr, limit);
+ assert(ctxPtr != NULL);
/* auto-init if forgotten */
- if (ctxPtr->base == NULL) LZ4HC_init (ctxPtr, (const BYTE*) src);
+ if (ctxPtr->base == NULL) LZ4HC_init_internal (ctxPtr, (const BYTE*) src);
/* Check overflow */
if ((size_t)(ctxPtr->end - ctxPtr->base) > 2 GB) {
@@ -971,7 +1119,8 @@ static int LZ4_compressHC_continue_generic (LZ4_streamHC_t* LZ4_streamHCPtr,
}
/* Check if blocks follow each other */
- if ((const BYTE*)src != ctxPtr->end) LZ4HC_setExternalDict(ctxPtr, (const BYTE*)src);
+ if ((const BYTE*)src != ctxPtr->end)
+ LZ4HC_setExternalDict(ctxPtr, (const BYTE*)src);
/* Check overlapping input/dictionary space */
{ const BYTE* sourceEnd = (const BYTE*) src + *srcSizePtr;
@@ -981,8 +1130,7 @@ static int LZ4_compressHC_continue_generic (LZ4_streamHC_t* LZ4_streamHCPtr,
if (sourceEnd > dictEnd) sourceEnd = dictEnd;
ctxPtr->lowLimit = (U32)(sourceEnd - ctxPtr->dictBase);
if (ctxPtr->dictLimit - ctxPtr->lowLimit < 4) ctxPtr->lowLimit = ctxPtr->dictLimit;
- }
- }
+ } }
return LZ4HC_compress_generic (ctxPtr, src, dst, srcSizePtr, dstCapacity, ctxPtr->compressionLevel, limit);
}
@@ -992,43 +1140,52 @@ int LZ4_compress_HC_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* src,
if (dstCapacity < LZ4_compressBound(srcSize))
return LZ4_compressHC_continue_generic (LZ4_streamHCPtr, src, dst, &srcSize, dstCapacity, limitedOutput);
else
- return LZ4_compressHC_continue_generic (LZ4_streamHCPtr, src, dst, &srcSize, dstCapacity, noLimit);
+ return LZ4_compressHC_continue_generic (LZ4_streamHCPtr, src, dst, &srcSize, dstCapacity, notLimited);
}
int LZ4_compress_HC_continue_destSize (LZ4_streamHC_t* LZ4_streamHCPtr, const char* src, char* dst, int* srcSizePtr, int targetDestSize)
{
- return LZ4_compressHC_continue_generic(LZ4_streamHCPtr, src, dst, srcSizePtr, targetDestSize, limitedDestSize);
+ return LZ4_compressHC_continue_generic(LZ4_streamHCPtr, src, dst, srcSizePtr, targetDestSize, fillOutput);
}
-/* dictionary saving */
-
+/* LZ4_saveDictHC :
+ * save history content
+ * into a user-provided buffer
+ * which is then used to continue compression
+ */
int LZ4_saveDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, char* safeBuffer, int dictSize)
{
LZ4HC_CCtx_internal* const streamPtr = &LZ4_streamHCPtr->internal_donotuse;
int const prefixSize = (int)(streamPtr->end - (streamPtr->base + streamPtr->dictLimit));
- DEBUGLOG(4, "LZ4_saveDictHC(%p, %p, %d)", LZ4_streamHCPtr, safeBuffer, dictSize);
+ DEBUGLOG(5, "LZ4_saveDictHC(%p, %p, %d)", LZ4_streamHCPtr, safeBuffer, dictSize);
+ assert(prefixSize >= 0);
if (dictSize > 64 KB) dictSize = 64 KB;
if (dictSize < 4) dictSize = 0;
if (dictSize > prefixSize) dictSize = prefixSize;
- memmove(safeBuffer, streamPtr->end - dictSize, dictSize);
+ if (safeBuffer == NULL) assert(dictSize == 0);
+ if (dictSize > 0)
+ memmove(safeBuffer, streamPtr->end - dictSize, dictSize);
{ U32 const endIndex = (U32)(streamPtr->end - streamPtr->base);
streamPtr->end = (const BYTE*)safeBuffer + dictSize;
streamPtr->base = streamPtr->end - endIndex;
- streamPtr->dictLimit = endIndex - dictSize;
- streamPtr->lowLimit = endIndex - dictSize;
- if (streamPtr->nextToUpdate < streamPtr->dictLimit) streamPtr->nextToUpdate = streamPtr->dictLimit;
+ streamPtr->dictLimit = endIndex - (U32)dictSize;
+ streamPtr->lowLimit = endIndex - (U32)dictSize;
+ if (streamPtr->nextToUpdate < streamPtr->dictLimit)
+ streamPtr->nextToUpdate = streamPtr->dictLimit;
}
return dictSize;
}
-/***********************************
+/***************************************************
* Deprecated Functions
-***********************************/
+***************************************************/
+
/* These functions currently generate deprecation warnings */
-/* Deprecated compression functions */
+
+/* Wrappers for deprecated compression functions */
int LZ4_compressHC(const char* src, char* dst, int srcSize) { return LZ4_compress_HC (src, dst, srcSize, LZ4_compressBound(srcSize), 0); }
int LZ4_compressHC_limitedOutput(const char* src, char* dst, int srcSize, int maxDstSize) { return LZ4_compress_HC(src, dst, srcSize, maxDstSize, 0); }
int LZ4_compressHC2(const char* src, char* dst, int srcSize, int cLevel) { return LZ4_compress_HC (src, dst, srcSize, LZ4_compressBound(srcSize), cLevel); }
@@ -1044,25 +1201,26 @@ int LZ4_compressHC_limitedOutput_continue (LZ4_streamHC_t* ctx, const char* src,
/* Deprecated streaming functions */
int LZ4_sizeofStreamStateHC(void) { return LZ4_STREAMHCSIZE; }
+/* state is presumed correctly sized, aka >= sizeof(LZ4_streamHC_t)
+ * @return : 0 on success, !=0 if error */
int LZ4_resetStreamStateHC(void* state, char* inputBuffer)
{
- LZ4HC_CCtx_internal *ctx = &((LZ4_streamHC_t*)state)->internal_donotuse;
- if ((((size_t)state) & (sizeof(void*)-1)) != 0) return 1; /* Error : pointer is not aligned for pointer (32 or 64 bits) */
- LZ4_resetStreamHC((LZ4_streamHC_t*)state, ((LZ4_streamHC_t*)state)->internal_donotuse.compressionLevel);
- LZ4HC_init(ctx, (const BYTE*)inputBuffer);
+ LZ4_streamHC_t* const hc4 = LZ4_initStreamHC(state, sizeof(*hc4));
+ if (hc4 == NULL) return 1; /* init failed */
+ LZ4HC_init_internal (&hc4->internal_donotuse, (const BYTE*)inputBuffer);
return 0;
}
void* LZ4_createHC (const char* inputBuffer)
{
- LZ4_streamHC_t* hc4 = (LZ4_streamHC_t*)ALLOC(sizeof(LZ4_streamHC_t));
+ LZ4_streamHC_t* const hc4 = LZ4_createStreamHC();
if (hc4 == NULL) return NULL; /* not enough memory */
- LZ4_resetStreamHC(hc4, 0 /* compressionLevel */);
- LZ4HC_init (&hc4->internal_donotuse, (const BYTE*)inputBuffer);
+ LZ4HC_init_internal (&hc4->internal_donotuse, (const BYTE*)inputBuffer);
return hc4;
}
-int LZ4_freeHC (void* LZ4HC_Data) {
+int LZ4_freeHC (void* LZ4HC_Data)
+{
if (!LZ4HC_Data) return 0; /* support free on NULL */
FREEMEM(LZ4HC_Data);
return 0;
@@ -1070,7 +1228,7 @@ int LZ4_freeHC (void* LZ4HC_Data) {
int LZ4_compressHC2_continue (void* LZ4HC_Data, const char* src, char* dst, int srcSize, int cLevel)
{
- return LZ4HC_compress_generic (&((LZ4_streamHC_t*)LZ4HC_Data)->internal_donotuse, src, dst, &srcSize, 0, cLevel, noLimit);
+ return LZ4HC_compress_generic (&((LZ4_streamHC_t*)LZ4HC_Data)->internal_donotuse, src, dst, &srcSize, 0, cLevel, notLimited);
}
int LZ4_compressHC2_limitedOutput_continue (void* LZ4HC_Data, const char* src, char* dst, int srcSize, int dstCapacity, int cLevel)
@@ -1089,7 +1247,7 @@ char* LZ4_slideInputBufferHC(void* LZ4HC_Data)
/* ================================================
- * LZ4 Optimal parser (levels 10-12)
+ * LZ4 Optimal parser (levels [LZ4HC_CLEVEL_OPT_MIN - LZ4HC_CLEVEL_MAX])
* ===============================================*/
typedef struct {
int price;
@@ -1102,8 +1260,9 @@ typedef struct {
LZ4_FORCE_INLINE int LZ4HC_literalsPrice(int const litlen)
{
int price = litlen;
+ assert(litlen >= 0);
if (litlen >= (int)RUN_MASK)
- price += 1 + (litlen-RUN_MASK)/255;
+ price += 1 + ((litlen-(int)RUN_MASK) / 255);
return price;
}
@@ -1112,11 +1271,13 @@ LZ4_FORCE_INLINE int LZ4HC_literalsPrice(int const litlen)
LZ4_FORCE_INLINE int LZ4HC_sequencePrice(int litlen, int mlen)
{
int price = 1 + 2 ; /* token + 16-bit offset */
+ assert(litlen >= 0);
+ assert(mlen >= MINMATCH);
price += LZ4HC_literalsPrice(litlen);
if (mlen >= (int)(ML_MASK+MINMATCH))
- price += 1 + (mlen-(ML_MASK+MINMATCH))/255;
+ price += 1 + ((mlen-(int)(ML_MASK+MINMATCH)) / 255);
return price;
}
@@ -1162,8 +1323,13 @@ static int LZ4HC_compress_optimal ( LZ4HC_CCtx_internal* ctx,
const dictCtx_directive dict,
const HCfavor_e favorDecSpeed)
{
+ int retval = 0;
#define TRAILING_LITERALS 3
+#ifdef LZ4HC_HEAPMODE
+ LZ4HC_optimal_t* const opt = (LZ4HC_optimal_t*)ALLOC(sizeof(LZ4HC_optimal_t) * (LZ4_OPT_NUM + TRAILING_LITERALS));
+#else
LZ4HC_optimal_t opt[LZ4_OPT_NUM + TRAILING_LITERALS]; /* ~64 KB, which is a bit large for stack... */
+#endif
const BYTE* ip = (const BYTE*) source;
const BYTE* anchor = ip;
@@ -1173,15 +1339,19 @@ static int LZ4HC_compress_optimal ( LZ4HC_CCtx_internal* ctx,
BYTE* op = (BYTE*) dst;
BYTE* opSaved = (BYTE*) dst;
BYTE* oend = op + dstCapacity;
+ int ovml = MINMATCH; /* overflow - last sequence */
+ const BYTE* ovref = NULL;
/* init */
- DEBUGLOG(5, "LZ4HC_compress_optimal");
+#ifdef LZ4HC_HEAPMODE
+ if (opt == NULL) goto _return_label;
+#endif
+ DEBUGLOG(5, "LZ4HC_compress_optimal(dst=%p, dstCapa=%u)", dst, (unsigned)dstCapacity);
*srcSizePtr = 0;
- if (limit == limitedDestSize) oend -= LASTLITERALS; /* Hack for support LZ4 format restriction */
+ if (limit == fillOutput) oend -= LASTLITERALS; /* Hack for support LZ4 format restriction */
if (sufficient_len >= LZ4_OPT_NUM) sufficient_len = LZ4_OPT_NUM-1;
/* Main Loop */
- assert(ip - anchor < LZ4_MAX_INPUT_SIZE);
while (ip <= mflimit) {
int const llen = (int)(ip - anchor);
int best_mlen, best_off;
@@ -1195,8 +1365,11 @@ static int LZ4HC_compress_optimal ( LZ4HC_CCtx_internal* ctx,
int const firstML = firstMatch.len;
const BYTE* const matchPos = ip - firstMatch.off;
opSaved = op;
- if ( LZ4HC_encodeSequence(&ip, &op, &anchor, firstML, matchPos, limit, oend) ) /* updates ip, op and anchor */
+ if ( LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), firstML, matchPos, limit, oend) ) { /* updates ip, op and anchor */
+ ovml = firstML;
+ ovref = matchPos;
goto _dest_overflow;
+ }
continue;
}
@@ -1333,11 +1506,12 @@ static int LZ4HC_compress_optimal ( LZ4HC_CCtx_internal* ctx,
} }
} /* for (cur = 1; cur <= last_match_pos; cur++) */
+ assert(last_match_pos < LZ4_OPT_NUM + TRAILING_LITERALS);
best_mlen = opt[last_match_pos].mlen;
best_off = opt[last_match_pos].off;
cur = last_match_pos - best_mlen;
- encode: /* cur, last_match_pos, best_mlen, best_off must be set */
+encode: /* cur, last_match_pos, best_mlen, best_off must be set */
assert(cur < LZ4_OPT_NUM);
assert(last_match_pos >= 1); /* == 1 when only one candidate */
DEBUGLOG(6, "reverse traversal, looking for shortest path (last_match_pos=%i)", last_match_pos);
@@ -1365,27 +1539,33 @@ static int LZ4HC_compress_optimal ( LZ4HC_CCtx_internal* ctx,
if (ml == 1) { ip++; rPos++; continue; } /* literal; note: can end up with several literals, in which case, skip them */
rPos += ml;
assert(ml >= MINMATCH);
- assert((offset >= 1) && (offset <= MAX_DISTANCE));
+ assert((offset >= 1) && (offset <= LZ4_DISTANCE_MAX));
opSaved = op;
- if ( LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ip - offset, limit, oend) ) /* updates ip, op and anchor */
+ if ( LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ml, ip - offset, limit, oend) ) { /* updates ip, op and anchor */
+ ovml = ml;
+ ovref = ip - offset;
goto _dest_overflow;
- } }
+ } } }
} /* while (ip <= mflimit) */
- _last_literals:
+_last_literals:
/* Encode Last Literals */
{ size_t lastRunSize = (size_t)(iend - anchor); /* literals */
- size_t litLength = (lastRunSize + 255 - RUN_MASK) / 255;
- size_t const totalSize = 1 + litLength + lastRunSize;
- if (limit == limitedDestSize) oend += LASTLITERALS; /* restore correct value */
+ size_t llAdd = (lastRunSize + 255 - RUN_MASK) / 255;
+ size_t const totalSize = 1 + llAdd + lastRunSize;
+ if (limit == fillOutput) oend += LASTLITERALS; /* restore correct value */
if (limit && (op + totalSize > oend)) {
- if (limit == limitedOutput) return 0; /* Check output limit */
+ if (limit == limitedOutput) { /* Check output limit */
+ retval = 0;
+ goto _return_label;
+ }
/* adapt lastRunSize to fill 'dst' */
- lastRunSize = (size_t)(oend - op) - 1;
- litLength = (lastRunSize + 255 - RUN_MASK) / 255;
- lastRunSize -= litLength;
+ lastRunSize = (size_t)(oend - op) - 1 /*token*/;
+ llAdd = (lastRunSize + 256 - RUN_MASK) / 256;
+ lastRunSize -= llAdd;
}
- ip = anchor + lastRunSize;
+ DEBUGLOG(6, "Final literal run : %i literals", (int)lastRunSize);
+ ip = anchor + lastRunSize; /* can be != iend if limit==fillOutput */
if (lastRunSize >= RUN_MASK) {
size_t accumulator = lastRunSize - RUN_MASK;
@@ -1401,12 +1581,35 @@ static int LZ4HC_compress_optimal ( LZ4HC_CCtx_internal* ctx,
/* End */
*srcSizePtr = (int) (((const char*)ip) - source);
- return (int) ((char*)op-dst);
+ retval = (int) ((char*)op-dst);
+ goto _return_label;
- _dest_overflow:
- if (limit == limitedDestSize) {
- op = opSaved; /* restore correct out pointer */
- goto _last_literals;
- }
- return 0;
- }
+_dest_overflow:
+if (limit == fillOutput) {
+ /* Assumption : ip, anchor, ovml and ovref must be set correctly */
+ size_t const ll = (size_t)(ip - anchor);
+ size_t const ll_addbytes = (ll + 240) / 255;
+ size_t const ll_totalCost = 1 + ll_addbytes + ll;
+ BYTE* const maxLitPos = oend - 3; /* 2 for offset, 1 for token */
+ DEBUGLOG(6, "Last sequence overflowing (only %i bytes remaining)", (int)(oend-1-opSaved));
+ op = opSaved; /* restore correct out pointer */
+ if (op + ll_totalCost <= maxLitPos) {
+ /* ll validated; now adjust match length */
+ size_t const bytesLeftForMl = (size_t)(maxLitPos - (op+ll_totalCost));
+ size_t const maxMlSize = MINMATCH + (ML_MASK-1) + (bytesLeftForMl * 255);
+ assert(maxMlSize < INT_MAX); assert(ovml >= 0);
+ if ((size_t)ovml > maxMlSize) ovml = (int)maxMlSize;
+ if ((oend + LASTLITERALS) - (op + ll_totalCost + 2) - 1 + ovml >= MFLIMIT) {
+ DEBUGLOG(6, "Space to end : %i + ml (%i)", (int)((oend + LASTLITERALS) - (op + ll_totalCost + 2) - 1), ovml);
+ DEBUGLOG(6, "Before : ip = %p, anchor = %p", ip, anchor);
+ LZ4HC_encodeSequence(UPDATABLE(ip, op, anchor), ovml, ovref, notLimited, oend);
+ DEBUGLOG(6, "After : ip = %p, anchor = %p", ip, anchor);
+ } }
+ goto _last_literals;
+}
+_return_label:
+#ifdef LZ4HC_HEAPMODE
+ FREEMEM(opt);
+#endif
+ return retval;
+}
diff --git a/lib/lz4hc.h b/lib/lz4hc.h
index 970fa39..3d441fb 100644
--- a/lib/lz4hc.h
+++ b/lib/lz4hc.h
@@ -54,7 +54,7 @@ extern "C" {
* Block Compression
**************************************/
/*! LZ4_compress_HC() :
- * Compress data from `src` into `dst`, using the more powerful but slower "HC" algorithm.
+ * Compress data from `src` into `dst`, using the powerful but slower "HC" algorithm.
* `dst` must be already allocated.
* Compression is guaranteed to succeed if `dstCapacity >= LZ4_compressBound(srcSize)` (see "lz4.h")
* Max supported `srcSize` value is LZ4_MAX_INPUT_SIZE (see "lz4.h")
@@ -77,7 +77,21 @@ LZ4LIB_API int LZ4_compress_HC (const char* src, char* dst, int srcSize, int dst
* Memory segment must be aligned on 8-bytes boundaries (which a normal malloc() should do properly).
*/
LZ4LIB_API int LZ4_sizeofStateHC(void);
-LZ4LIB_API int LZ4_compress_HC_extStateHC(void* state, const char* src, char* dst, int srcSize, int maxDstSize, int compressionLevel);
+LZ4LIB_API int LZ4_compress_HC_extStateHC(void* stateHC, const char* src, char* dst, int srcSize, int maxDstSize, int compressionLevel);
+
+
+/*! LZ4_compress_HC_destSize() : v1.9.0+
+ * Will compress as much data as possible from `src`
+ * to fit into `targetDstSize` budget.
+ * Result is provided in 2 parts :
+ * @return : the number of bytes written into 'dst' (necessarily <= targetDstSize)
+ * or 0 if compression fails.
+ * `srcSizePtr` : on success, *srcSizePtr is updated to indicate how much bytes were read from `src`
+ */
+LZ4LIB_API int LZ4_compress_HC_destSize(void* stateHC,
+ const char* src, char* dst,
+ int* srcSizePtr, int targetDstSize,
+ int compressionLevel);
/*-************************************
@@ -89,46 +103,92 @@ LZ4LIB_API int LZ4_compress_HC_extStateHC(void* state, const char* src, char* ds
/*! LZ4_createStreamHC() and LZ4_freeStreamHC() :
* These functions create and release memory for LZ4 HC streaming state.
* Newly created states are automatically initialized.
- * Existing states can be re-used several times, using LZ4_resetStreamHC().
- * These methods are API and ABI stable, they can be used in combination with a DLL.
+ * A same state can be used multiple times consecutively,
+ * starting with LZ4_resetStreamHC_fast() to start a new stream of blocks.
*/
LZ4LIB_API LZ4_streamHC_t* LZ4_createStreamHC(void);
LZ4LIB_API int LZ4_freeStreamHC (LZ4_streamHC_t* streamHCPtr);
-LZ4LIB_API void LZ4_resetStreamHC (LZ4_streamHC_t* streamHCPtr, int compressionLevel);
-LZ4LIB_API int LZ4_loadDictHC (LZ4_streamHC_t* streamHCPtr, const char* dictionary, int dictSize);
-
-LZ4LIB_API int LZ4_compress_HC_continue (LZ4_streamHC_t* streamHCPtr, const char* src, char* dst, int srcSize, int maxDstSize);
-
-LZ4LIB_API int LZ4_saveDictHC (LZ4_streamHC_t* streamHCPtr, char* safeBuffer, int maxDictSize);
-
/*
- These functions compress data in successive blocks of any size, using previous blocks as dictionary.
+ These functions compress data in successive blocks of any size,
+ using previous blocks as dictionary, to improve compression ratio.
One key assumption is that previous blocks (up to 64 KB) remain read-accessible while compressing next blocks.
There is an exception for ring buffers, which can be smaller than 64 KB.
- Ring buffers scenario is automatically detected and handled by LZ4_compress_HC_continue().
+ Ring-buffer scenario is automatically detected and handled within LZ4_compress_HC_continue().
+
+ Before starting compression, state must be allocated and properly initialized.
+ LZ4_createStreamHC() does both, though compression level is set to LZ4HC_CLEVEL_DEFAULT.
+
+ Selecting the compression level can be done with LZ4_resetStreamHC_fast() (starts a new stream)
+ or LZ4_setCompressionLevel() (anytime, between blocks in the same stream) (experimental).
+ LZ4_resetStreamHC_fast() only works on states which have been properly initialized at least once,
+ which is automatically the case when state is created using LZ4_createStreamHC().
+
+ After reset, a first "fictional block" can be designated as initial dictionary,
+ using LZ4_loadDictHC() (Optional).
+
+ Invoke LZ4_compress_HC_continue() to compress each successive block.
+ The number of blocks is unlimited.
+ Previous input blocks, including initial dictionary when present,
+ must remain accessible and unmodified during compression.
+
+ It's allowed to update compression level anytime between blocks,
+ using LZ4_setCompressionLevel() (experimental).
+
+ 'dst' buffer should be sized to handle worst case scenarios
+ (see LZ4_compressBound(), it ensures compression success).
+ In case of failure, the API does not guarantee recovery,
+ so the state _must_ be reset.
+ To ensure compression success
+ whenever `dst` buffer size cannot be made >= LZ4_compressBound(),
+ consider using LZ4_compress_HC_continue_destSize().
+
+ Whenever previous input blocks can't be preserved unmodified in-place during compression of next blocks,
+ it's possible to copy the last blocks into a more stable memory space, using LZ4_saveDictHC().
+ Return value of LZ4_saveDictHC() is the size of dictionary effectively saved into 'safeBuffer' (<= 64 KB)
+
+ After completing a streaming compression,
+ it's possible to start a new stream of blocks, using the same LZ4_streamHC_t state,
+ just by resetting it, using LZ4_resetStreamHC_fast().
+*/
- Before starting compression, state must be properly initialized, using LZ4_resetStreamHC().
- A first "fictional block" can then be designated as initial dictionary, using LZ4_loadDictHC() (Optional).
+LZ4LIB_API void LZ4_resetStreamHC_fast(LZ4_streamHC_t* streamHCPtr, int compressionLevel); /* v1.9.0+ */
+LZ4LIB_API int LZ4_loadDictHC (LZ4_streamHC_t* streamHCPtr, const char* dictionary, int dictSize);
+
+LZ4LIB_API int LZ4_compress_HC_continue (LZ4_streamHC_t* streamHCPtr,
+ const char* src, char* dst,
+ int srcSize, int maxDstSize);
+
+/*! LZ4_compress_HC_continue_destSize() : v1.9.0+
+ * Similar to LZ4_compress_HC_continue(),
+ * but will read as much data as possible from `src`
+ * to fit into `targetDstSize` budget.
+ * Result is provided into 2 parts :
+ * @return : the number of bytes written into 'dst' (necessarily <= targetDstSize)
+ * or 0 if compression fails.
+ * `srcSizePtr` : on success, *srcSizePtr will be updated to indicate how much bytes were read from `src`.
+ * Note that this function may not consume the entire input.
+ */
+LZ4LIB_API int LZ4_compress_HC_continue_destSize(LZ4_streamHC_t* LZ4_streamHCPtr,
+ const char* src, char* dst,
+ int* srcSizePtr, int targetDstSize);
+
+LZ4LIB_API int LZ4_saveDictHC (LZ4_streamHC_t* streamHCPtr, char* safeBuffer, int maxDictSize);
- Then, use LZ4_compress_HC_continue() to compress each successive block.
- Previous memory blocks (including initial dictionary when present) must remain accessible and unmodified during compression.
- 'dst' buffer should be sized to handle worst case scenarios (see LZ4_compressBound()), to ensure operation success.
- Because in case of failure, the API does not guarantee context recovery, and context will have to be reset.
- If `dst` buffer budget cannot be >= LZ4_compressBound(), consider using LZ4_compress_HC_continue_destSize() instead.
- If, for any reason, previous data block can't be preserved unmodified in memory for next compression block,
- you can save it to a more stable memory space, using LZ4_saveDictHC().
- Return value of LZ4_saveDictHC() is the size of dictionary effectively saved into 'safeBuffer'.
-*/
+/*^**********************************************
+ * !!!!!! STATIC LINKING ONLY !!!!!!
+ ***********************************************/
-/*-**************************************************************
+/*-******************************************************************
* PRIVATE DEFINITIONS :
- * Do not use these definitions.
- * They are exposed to allow static allocation of `LZ4_streamHC_t`.
- * Using these definitions makes the code vulnerable to potential API break when upgrading LZ4
- ****************************************************************/
+ * Do not use these definitions directly.
+ * They are merely exposed to allow static allocation of `LZ4_streamHC_t`.
+ * Declare an `LZ4_streamHC_t` directly, rather than any type below.
+ * Even then, only do so in the context of static linking, as definitions may change between versions.
+ ********************************************************************/
+
#define LZ4HC_DICTIONARY_LOGSIZE 16
#define LZ4HC_MAXD (1<<LZ4HC_DICTIONARY_LOGSIZE)
#define LZ4HC_MAXD_MASK (LZ4HC_MAXD - 1)
@@ -138,59 +198,53 @@ LZ4LIB_API int LZ4_saveDictHC (LZ4_streamHC_t* streamHCPtr, char* safeBuffer, in
#define LZ4HC_HASH_MASK (LZ4HC_HASHTABLESIZE - 1)
-#if defined(__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
-#include <stdint.h>
-
-typedef struct LZ4HC_CCtx_internal LZ4HC_CCtx_internal;
-struct LZ4HC_CCtx_internal
-{
- uint32_t hashTable[LZ4HC_HASHTABLESIZE];
- uint16_t chainTable[LZ4HC_MAXD];
- const uint8_t* end; /* next block here to continue on current prefix */
- const uint8_t* base; /* All index relative to this position */
- const uint8_t* dictBase; /* alternate base for extDict */
- uint32_t dictLimit; /* below that point, need extDict */
- uint32_t lowLimit; /* below that point, no more dict */
- uint32_t nextToUpdate; /* index from which to continue dictionary update */
- short compressionLevel;
- short favorDecSpeed;
- const LZ4HC_CCtx_internal* dictCtx;
-};
-
-#else
-
typedef struct LZ4HC_CCtx_internal LZ4HC_CCtx_internal;
struct LZ4HC_CCtx_internal
{
- unsigned int hashTable[LZ4HC_HASHTABLESIZE];
- unsigned short chainTable[LZ4HC_MAXD];
- const unsigned char* end; /* next block here to continue on current prefix */
- const unsigned char* base; /* All index relative to this position */
- const unsigned char* dictBase; /* alternate base for extDict */
- unsigned int dictLimit; /* below that point, need extDict */
- unsigned int lowLimit; /* below that point, no more dict */
- unsigned int nextToUpdate; /* index from which to continue dictionary update */
- short compressionLevel;
- short favorDecSpeed;
+ LZ4_u32 hashTable[LZ4HC_HASHTABLESIZE];
+ LZ4_u16 chainTable[LZ4HC_MAXD];
+ const LZ4_byte* end; /* next block here to continue on current prefix */
+ const LZ4_byte* base; /* All index relative to this position */
+ const LZ4_byte* dictBase; /* alternate base for extDict */
+ LZ4_u32 dictLimit; /* below that point, need extDict */
+ LZ4_u32 lowLimit; /* below that point, no more dict */
+ LZ4_u32 nextToUpdate; /* index from which to continue dictionary update */
+ short compressionLevel;
+ LZ4_i8 favorDecSpeed; /* favor decompression speed if this flag set,
+ otherwise, favor compression ratio */
+ LZ4_i8 dirty; /* stream has to be fully reset if this flag is set */
const LZ4HC_CCtx_internal* dictCtx;
};
-#endif
-#define LZ4_STREAMHCSIZE (4*LZ4HC_HASHTABLESIZE + 2*LZ4HC_MAXD + 56) /* 262200 */
-#define LZ4_STREAMHCSIZE_SIZET (LZ4_STREAMHCSIZE / sizeof(size_t))
+/* Do not use these definitions directly !
+ * Declare or allocate an LZ4_streamHC_t instead.
+ */
+#define LZ4_STREAMHCSIZE 262200 /* static size, for inter-version compatibility */
+#define LZ4_STREAMHCSIZE_VOIDP (LZ4_STREAMHCSIZE / sizeof(void*))
union LZ4_streamHC_u {
- size_t table[LZ4_STREAMHCSIZE_SIZET];
+ void* table[LZ4_STREAMHCSIZE_VOIDP];
LZ4HC_CCtx_internal internal_donotuse;
-}; /* previously typedef'd to LZ4_streamHC_t */
-/*
- LZ4_streamHC_t :
- This structure allows static allocation of LZ4 HC streaming state.
- State must be initialized using LZ4_resetStreamHC() before first use.
+}; /* previously typedef'd to LZ4_streamHC_t */
- Static allocation shall only be used in combination with static linking.
- When invoking LZ4 from a DLL, use create/free functions instead, which are API and ABI stable.
-*/
+/* LZ4_streamHC_t :
+ * This structure allows static allocation of LZ4 HC streaming state.
+ * This can be used to allocate statically, on state, or as part of a larger structure.
+ *
+ * Such state **must** be initialized using LZ4_initStreamHC() before first use.
+ *
+ * Note that invoking LZ4_initStreamHC() is not required when
+ * the state was created using LZ4_createStreamHC() (which is recommended).
+ * Using the normal builder, a newly created state is automatically initialized.
+ *
+ * Static allocation shall only be used in combination with static linking.
+ */
+
+/* LZ4_initStreamHC() : v1.9.0+
+ * Required before first use of a statically allocated LZ4_streamHC_t.
+ * Before v1.9.0 : use LZ4_resetStreamHC() instead
+ */
+LZ4LIB_API LZ4_streamHC_t* LZ4_initStreamHC (void* buffer, size_t size);
/*-************************************
@@ -201,11 +255,11 @@ union LZ4_streamHC_u {
/* deprecated compression functions */
LZ4_DEPRECATED("use LZ4_compress_HC() instead") LZ4LIB_API int LZ4_compressHC (const char* source, char* dest, int inputSize);
LZ4_DEPRECATED("use LZ4_compress_HC() instead") LZ4LIB_API int LZ4_compressHC_limitedOutput (const char* source, char* dest, int inputSize, int maxOutputSize);
-LZ4_DEPRECATED("use LZ4_compress_HC() instead") LZ4LIB_API int LZ4_compressHC2 (const char* source, char* dest, int inputSize, int compressionLevel);
-LZ4_DEPRECATED("use LZ4_compress_HC() instead") LZ4LIB_API int LZ4_compressHC2_limitedOutput (const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel);
+LZ4_DEPRECATED("use LZ4_compress_HC() instead") LZ4LIB_API int LZ4_compressHC2 (const char* source, char* dest, int inputSize, int compressionLevel);
+LZ4_DEPRECATED("use LZ4_compress_HC() instead") LZ4LIB_API int LZ4_compressHC2_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel);
LZ4_DEPRECATED("use LZ4_compress_HC_extStateHC() instead") LZ4LIB_API int LZ4_compressHC_withStateHC (void* state, const char* source, char* dest, int inputSize);
LZ4_DEPRECATED("use LZ4_compress_HC_extStateHC() instead") LZ4LIB_API int LZ4_compressHC_limitedOutput_withStateHC (void* state, const char* source, char* dest, int inputSize, int maxOutputSize);
-LZ4_DEPRECATED("use LZ4_compress_HC_extStateHC() instead") LZ4LIB_API int LZ4_compressHC2_withStateHC (void* state, const char* source, char* dest, int inputSize, int compressionLevel);
+LZ4_DEPRECATED("use LZ4_compress_HC_extStateHC() instead") LZ4LIB_API int LZ4_compressHC2_withStateHC (void* state, const char* source, char* dest, int inputSize, int compressionLevel);
LZ4_DEPRECATED("use LZ4_compress_HC_extStateHC() instead") LZ4LIB_API int LZ4_compressHC2_limitedOutput_withStateHC(void* state, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel);
LZ4_DEPRECATED("use LZ4_compress_HC_continue() instead") LZ4LIB_API int LZ4_compressHC_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize);
LZ4_DEPRECATED("use LZ4_compress_HC_continue() instead") LZ4LIB_API int LZ4_compressHC_limitedOutput_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize, int maxOutputSize);
@@ -221,10 +275,21 @@ LZ4_DEPRECATED("use LZ4_compress_HC_continue() instead") LZ4LIB_API int LZ4_comp
LZ4_DEPRECATED("use LZ4_createStreamHC() instead") LZ4LIB_API void* LZ4_createHC (const char* inputBuffer);
LZ4_DEPRECATED("use LZ4_saveDictHC() instead") LZ4LIB_API char* LZ4_slideInputBufferHC (void* LZ4HC_Data);
LZ4_DEPRECATED("use LZ4_freeStreamHC() instead") LZ4LIB_API int LZ4_freeHC (void* LZ4HC_Data);
-LZ4_DEPRECATED("use LZ4_compress_HC_continue() instead") LZ4LIB_API int LZ4_compressHC2_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int compressionLevel);
+LZ4_DEPRECATED("use LZ4_compress_HC_continue() instead") LZ4LIB_API int LZ4_compressHC2_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int compressionLevel);
LZ4_DEPRECATED("use LZ4_compress_HC_continue() instead") LZ4LIB_API int LZ4_compressHC2_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel);
LZ4_DEPRECATED("use LZ4_createStreamHC() instead") LZ4LIB_API int LZ4_sizeofStreamStateHC(void);
-LZ4_DEPRECATED("use LZ4_resetStreamHC() instead") LZ4LIB_API int LZ4_resetStreamStateHC(void* state, char* inputBuffer);
+LZ4_DEPRECATED("use LZ4_initStreamHC() instead") LZ4LIB_API int LZ4_resetStreamStateHC(void* state, char* inputBuffer);
+
+
+/* LZ4_resetStreamHC() is now replaced by LZ4_initStreamHC().
+ * The intention is to emphasize the difference with LZ4_resetStreamHC_fast(),
+ * which is now the recommended function to start a new stream of blocks,
+ * but cannot be used to initialize a memory segment containing arbitrary garbage data.
+ *
+ * It is recommended to switch to LZ4_initStreamHC().
+ * LZ4_resetStreamHC() will generate deprecation warnings in a future version.
+ */
+LZ4LIB_API void LZ4_resetStreamHC (LZ4_streamHC_t* streamHCPtr, int compressionLevel);
#if defined (__cplusplus)
@@ -246,48 +311,29 @@ LZ4_DEPRECATED("use LZ4_resetStreamHC() instead") LZ4LIB_API int LZ4_resetStr
#ifndef LZ4_HC_SLO_098092834
#define LZ4_HC_SLO_098092834
+#define LZ4_STATIC_LINKING_ONLY /* LZ4LIB_STATIC_API */
+#include "lz4.h"
+
#if defined (__cplusplus)
extern "C" {
#endif
-/*! LZ4_compress_HC_destSize() : v1.8.0 (experimental)
- * Will try to compress as much data from `src` as possible
- * that can fit into `targetDstSize` budget.
- * Result is provided in 2 parts :
- * @return : the number of bytes written into 'dst'
- * or 0 if compression fails.
- * `srcSizePtr` : value will be updated to indicate how much bytes were read from `src`
- */
-int LZ4_compress_HC_destSize(void* LZ4HC_Data,
- const char* src, char* dst,
- int* srcSizePtr, int targetDstSize,
- int compressionLevel);
-
-/*! LZ4_compress_HC_continue_destSize() : v1.8.0 (experimental)
- * Similar as LZ4_compress_HC_continue(),
- * but will read a variable nb of bytes from `src`
- * to fit into `targetDstSize` budget.
- * Result is provided in 2 parts :
- * @return : the number of bytes written into 'dst'
- * or 0 if compression fails.
- * `srcSizePtr` : value will be updated to indicate how much bytes were read from `src`.
- */
-int LZ4_compress_HC_continue_destSize(LZ4_streamHC_t* LZ4_streamHCPtr,
- const char* src, char* dst,
- int* srcSizePtr, int targetDstSize);
-
-/*! LZ4_setCompressionLevel() : v1.8.0 (experimental)
- * It's possible to change compression level between 2 invocations of LZ4_compress_HC_continue*()
+/*! LZ4_setCompressionLevel() : v1.8.0+ (experimental)
+ * It's possible to change compression level
+ * between successive invocations of LZ4_compress_HC_continue*()
+ * for dynamic adaptation.
*/
-void LZ4_setCompressionLevel(LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel);
+LZ4LIB_STATIC_API void LZ4_setCompressionLevel(
+ LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel);
-/*! LZ4_favorDecompressionSpeed() : v1.8.2 (experimental)
- * Parser will select decisions favoring decompression over compression ratio.
- * Only work at highest compression settings (level >= LZ4HC_CLEVEL_OPT_MIN)
+/*! LZ4_favorDecompressionSpeed() : v1.8.2+ (experimental)
+ * Opt. Parser will favor decompression speed over compression ratio.
+ * Only applicable to levels >= LZ4HC_CLEVEL_OPT_MIN.
*/
-void LZ4_favorDecompressionSpeed(LZ4_streamHC_t* LZ4_streamHCPtr, int favor);
+LZ4LIB_STATIC_API void LZ4_favorDecompressionSpeed(
+ LZ4_streamHC_t* LZ4_streamHCPtr, int favor);
-/*! LZ4_resetStreamHC_fast() :
+/*! LZ4_resetStreamHC_fast() : v1.9.0+
* When an LZ4_streamHC_t is known to be in a internally coherent state,
* it can often be prepared for a new compression with almost no work, only
* sometimes falling back to the full, expensive reset that is always required
@@ -304,8 +350,14 @@ void LZ4_favorDecompressionSpeed(LZ4_streamHC_t* LZ4_streamHCPtr, int favor);
* - the stream was in an indeterminate state and was used in a compression
* call that fully reset the state (LZ4_compress_HC_extStateHC()) and that
* returned success
+ *
+ * Note:
+ * A stream that was last used in a compression call that returned an error
+ * may be passed to this function. However, it will be fully reset, which will
+ * clear any existing history and settings from the context.
*/
-void LZ4_resetStreamHC_fast(LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel);
+LZ4LIB_STATIC_API void LZ4_resetStreamHC_fast(
+ LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel);
/*! LZ4_compress_HC_extStateHC_fastReset() :
* A variant of LZ4_compress_HC_extStateHC().
@@ -318,7 +370,11 @@ void LZ4_resetStreamHC_fast(LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLeve
* LZ4_resetStreamHC_fast() while LZ4_compress_HC_extStateHC() starts with a
* call to LZ4_resetStreamHC().
*/
-int LZ4_compress_HC_extStateHC_fastReset (void* state, const char* src, char* dst, int srcSize, int dstCapacity, int compressionLevel);
+LZ4LIB_STATIC_API int LZ4_compress_HC_extStateHC_fastReset (
+ void* state,
+ const char* src, char* dst,
+ int srcSize, int dstCapacity,
+ int compressionLevel);
/*! LZ4_attach_HC_dictionary() :
* This is an experimental API that allows for the efficient use of a
@@ -345,7 +401,9 @@ int LZ4_compress_HC_extStateHC_fastReset (void* state, const char* src, char* ds
* stream (and source buffer) must remain in-place / accessible / unchanged
* through the lifetime of the stream session.
*/
-LZ4LIB_API void LZ4_attach_HC_dictionary(LZ4_streamHC_t *working_stream, const LZ4_streamHC_t *dictionary_stream);
+LZ4LIB_STATIC_API void LZ4_attach_HC_dictionary(
+ LZ4_streamHC_t *working_stream,
+ const LZ4_streamHC_t *dictionary_stream);
#if defined (__cplusplus)
}
diff --git a/lib/xxhash.c b/lib/xxhash.c
index 3fc97fd..ff28749 100644
--- a/lib/xxhash.c
+++ b/lib/xxhash.c
@@ -50,20 +50,26 @@
* Prefer these methods in priority order (0 > 1 > 2)
*/
#ifndef XXH_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */
-# if defined(__GNUC__) && ( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
+# if defined(__GNUC__) && ( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) \
+ || defined(__ARM_ARCH_6K__) || defined(__ARM_ARCH_6Z__) \
+ || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
# define XXH_FORCE_MEMORY_ACCESS 2
# elif (defined(__INTEL_COMPILER) && !defined(_WIN32)) || \
- (defined(__GNUC__) && ( defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7S__) ))
+ (defined(__GNUC__) && ( defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__) \
+ || defined(__ARM_ARCH_7R__) || defined(__ARM_ARCH_7M__) \
+ || defined(__ARM_ARCH_7S__) ))
# define XXH_FORCE_MEMORY_ACCESS 1
# endif
#endif
/*!XXH_ACCEPT_NULL_INPUT_POINTER :
- * If the input pointer is a null pointer, xxHash default behavior is to trigger a memory access error, since it is a bad pointer.
- * When this option is enabled, xxHash output for null input pointers will be the same as a null-length input.
- * By default, this option is disabled. To enable it, uncomment below define :
+ * If input pointer is NULL, xxHash default behavior is to dereference it, triggering a segfault.
+ * When this macro is enabled, xxHash actively checks input for null pointer.
+ * It it is, result for null input pointers is the same as a null-length input.
*/
-/* #define XXH_ACCEPT_NULL_INPUT_POINTER 1 */
+#ifndef XXH_ACCEPT_NULL_INPUT_POINTER /* can be defined externally */
+# define XXH_ACCEPT_NULL_INPUT_POINTER 0
+#endif
/*!XXH_FORCE_NATIVE_FORMAT :
* By default, xxHash library provides endian-independent Hash values, based on little-endian convention.
@@ -80,8 +86,9 @@
/*!XXH_FORCE_ALIGN_CHECK :
* This is a minor performance trick, only useful with lots of very small keys.
* It means : check for aligned/unaligned input.
- * The check costs one initial branch per hash; set to 0 when the input data
- * is guaranteed to be aligned.
+ * The check costs one initial branch per hash;
+ * set it to 0 when the input is guaranteed to be aligned,
+ * or when alignment doesn't matter for performance.
*/
#ifndef XXH_FORCE_ALIGN_CHECK /* can be defined externally */
# if defined(__i386) || defined(_M_IX86) || defined(__x86_64__) || defined(_M_X64)
@@ -104,6 +111,8 @@ static void XXH_free (void* p) { free(p); }
#include <string.h>
static void* XXH_memcpy(void* dest, const void* src, size_t size) { return memcpy(dest,src,size); }
+#include <assert.h> /* assert */
+
#define XXH_STATIC_LINKING_ONLY
#include "xxhash.h"
@@ -113,40 +122,35 @@ static void* XXH_memcpy(void* dest, const void* src, size_t size) { return memcp
***************************************/
#ifdef _MSC_VER /* Visual Studio */
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
-#endif
-
-#ifndef XXH_FORCE_INLINE
-# ifdef _MSC_VER /* Visual Studio */
-# define XXH_FORCE_INLINE static __forceinline
-# else
-# if defined (__cplusplus) || defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */
-# ifdef __GNUC__
-# define XXH_FORCE_INLINE static inline __attribute__((always_inline))
-# else
-# define XXH_FORCE_INLINE static inline
-# endif
+# define FORCE_INLINE static __forceinline
+#else
+# if defined (__cplusplus) || defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */
+# ifdef __GNUC__
+# define FORCE_INLINE static inline __attribute__((always_inline))
# else
-# define XXH_FORCE_INLINE static
-# endif /* __STDC_VERSION__ */
-# endif /* _MSC_VER */
-#endif /* XXH_FORCE_INLINE */
+# define FORCE_INLINE static inline
+# endif
+# else
+# define FORCE_INLINE static
+# endif /* __STDC_VERSION__ */
+#endif
/* *************************************
* Basic Types
***************************************/
#ifndef MEM_MODULE
-# if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
+# if !defined (__VMS) \
+ && (defined (__cplusplus) \
+ || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
# include <stdint.h>
typedef uint8_t BYTE;
typedef uint16_t U16;
typedef uint32_t U32;
- typedef int32_t S32;
# else
typedef unsigned char BYTE;
typedef unsigned short U16;
typedef unsigned int U32;
- typedef signed int S32;
# endif
#endif
@@ -213,8 +217,12 @@ typedef enum { XXH_bigEndian=0, XXH_littleEndian=1 } XXH_endianess;
/* XXH_CPU_LITTLE_ENDIAN can be defined externally, for example on the compiler command line */
#ifndef XXH_CPU_LITTLE_ENDIAN
- static const int g_one = 1;
-# define XXH_CPU_LITTLE_ENDIAN (*(const char*)(&g_one))
+static int XXH_isLittleEndian(void)
+{
+ const union { U32 u; BYTE c[4]; } one = { 1 }; /* don't use static : performance detrimental */
+ return one.c[0];
+}
+# define XXH_CPU_LITTLE_ENDIAN XXH_isLittleEndian()
#endif
@@ -223,7 +231,7 @@ typedef enum { XXH_bigEndian=0, XXH_littleEndian=1 } XXH_endianess;
*****************************/
typedef enum { XXH_aligned, XXH_unaligned } XXH_alignment;
-XXH_FORCE_INLINE U32 XXH_readLE32_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
+FORCE_INLINE U32 XXH_readLE32_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
{
if (align==XXH_unaligned)
return endian==XXH_littleEndian ? XXH_read32(ptr) : XXH_swap32(XXH_read32(ptr));
@@ -231,7 +239,7 @@ XXH_FORCE_INLINE U32 XXH_readLE32_align(const void* ptr, XXH_endianess endian, X
return endian==XXH_littleEndian ? *(const U32*)ptr : XXH_swap32(*(const U32*)ptr);
}
-XXH_FORCE_INLINE U32 XXH_readLE32(const void* ptr, XXH_endianess endian)
+FORCE_INLINE U32 XXH_readLE32(const void* ptr, XXH_endianess endian)
{
return XXH_readLE32_align(ptr, endian, XXH_unaligned);
}
@@ -245,12 +253,12 @@ static U32 XXH_readBE32(const void* ptr)
/* *************************************
* Macros
***************************************/
-#define XXH_STATIC_ASSERT(c) { enum { XXH_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
+#define XXH_STATIC_ASSERT(c) { enum { XXH_sa = 1/(int)(!!(c)) }; } /* use after variable declarations */
XXH_PUBLIC_API unsigned XXH_versionNumber (void) { return XXH_VERSION_NUMBER; }
/* *******************************************************************
-* 32-bits hash functions
+* 32-bit hash functions
*********************************************************************/
static const U32 PRIME32_1 = 2654435761U;
static const U32 PRIME32_2 = 2246822519U;
@@ -266,14 +274,89 @@ static U32 XXH32_round(U32 seed, U32 input)
return seed;
}
-XXH_FORCE_INLINE U32 XXH32_endian_align(const void* input, size_t len, U32 seed, XXH_endianess endian, XXH_alignment align)
+/* mix all bits */
+static U32 XXH32_avalanche(U32 h32)
+{
+ h32 ^= h32 >> 15;
+ h32 *= PRIME32_2;
+ h32 ^= h32 >> 13;
+ h32 *= PRIME32_3;
+ h32 ^= h32 >> 16;
+ return(h32);
+}
+
+#define XXH_get32bits(p) XXH_readLE32_align(p, endian, align)
+
+static U32
+XXH32_finalize(U32 h32, const void* ptr, size_t len,
+ XXH_endianess endian, XXH_alignment align)
+
+{
+ const BYTE* p = (const BYTE*)ptr;
+
+#define PROCESS1 \
+ h32 += (*p++) * PRIME32_5; \
+ h32 = XXH_rotl32(h32, 11) * PRIME32_1 ;
+
+#define PROCESS4 \
+ h32 += XXH_get32bits(p) * PRIME32_3; \
+ p+=4; \
+ h32 = XXH_rotl32(h32, 17) * PRIME32_4 ;
+
+ switch(len&15) /* or switch(bEnd - p) */
+ {
+ case 12: PROCESS4;
+ /* fallthrough */
+ case 8: PROCESS4;
+ /* fallthrough */
+ case 4: PROCESS4;
+ return XXH32_avalanche(h32);
+
+ case 13: PROCESS4;
+ /* fallthrough */
+ case 9: PROCESS4;
+ /* fallthrough */
+ case 5: PROCESS4;
+ PROCESS1;
+ return XXH32_avalanche(h32);
+
+ case 14: PROCESS4;
+ /* fallthrough */
+ case 10: PROCESS4;
+ /* fallthrough */
+ case 6: PROCESS4;
+ PROCESS1;
+ PROCESS1;
+ return XXH32_avalanche(h32);
+
+ case 15: PROCESS4;
+ /* fallthrough */
+ case 11: PROCESS4;
+ /* fallthrough */
+ case 7: PROCESS4;
+ /* fallthrough */
+ case 3: PROCESS1;
+ /* fallthrough */
+ case 2: PROCESS1;
+ /* fallthrough */
+ case 1: PROCESS1;
+ /* fallthrough */
+ case 0: return XXH32_avalanche(h32);
+ }
+ assert(0);
+ return h32; /* reaching this point is deemed impossible */
+}
+
+
+FORCE_INLINE U32
+XXH32_endian_align(const void* input, size_t len, U32 seed,
+ XXH_endianess endian, XXH_alignment align)
{
const BYTE* p = (const BYTE*)input;
const BYTE* bEnd = p + len;
U32 h32;
-#define XXH_get32bits(p) XXH_readLE32_align(p, endian, align)
-#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
+#if defined(XXH_ACCEPT_NULL_INPUT_POINTER) && (XXH_ACCEPT_NULL_INPUT_POINTER>=1)
if (p==NULL) {
len=0;
bEnd=p=(const BYTE*)(size_t)16;
@@ -281,7 +364,7 @@ XXH_FORCE_INLINE U32 XXH32_endian_align(const void* input, size_t len, U32 seed,
#endif
if (len>=16) {
- const BYTE* const limit = bEnd - 16;
+ const BYTE* const limit = bEnd - 15;
U32 v1 = seed + PRIME32_1 + PRIME32_2;
U32 v2 = seed + PRIME32_2;
U32 v3 = seed + 0;
@@ -292,34 +375,17 @@ XXH_FORCE_INLINE U32 XXH32_endian_align(const void* input, size_t len, U32 seed,
v2 = XXH32_round(v2, XXH_get32bits(p)); p+=4;
v3 = XXH32_round(v3, XXH_get32bits(p)); p+=4;
v4 = XXH32_round(v4, XXH_get32bits(p)); p+=4;
- } while (p<=limit);
+ } while (p < limit);
- h32 = XXH_rotl32(v1, 1) + XXH_rotl32(v2, 7) + XXH_rotl32(v3, 12) + XXH_rotl32(v4, 18);
+ h32 = XXH_rotl32(v1, 1) + XXH_rotl32(v2, 7)
+ + XXH_rotl32(v3, 12) + XXH_rotl32(v4, 18);
} else {
h32 = seed + PRIME32_5;
}
- h32 += (U32) len;
-
- while (p+4<=bEnd) {
- h32 += XXH_get32bits(p) * PRIME32_3;
- h32 = XXH_rotl32(h32, 17) * PRIME32_4 ;
- p+=4;
- }
-
- while (p<bEnd) {
- h32 += (*p) * PRIME32_5;
- h32 = XXH_rotl32(h32, 11) * PRIME32_1 ;
- p++;
- }
-
- h32 ^= h32 >> 15;
- h32 *= PRIME32_2;
- h32 ^= h32 >> 13;
- h32 *= PRIME32_3;
- h32 ^= h32 >> 16;
+ h32 += (U32)len;
- return h32;
+ return XXH32_finalize(h32, p, len&15, endian, align);
}
@@ -371,74 +437,81 @@ XXH_PUBLIC_API void XXH32_copyState(XXH32_state_t* dstState, const XXH32_state_t
XXH_PUBLIC_API XXH_errorcode XXH32_reset(XXH32_state_t* statePtr, unsigned int seed)
{
XXH32_state_t state; /* using a local state to memcpy() in order to avoid strict-aliasing warnings */
- memset(&state, 0, sizeof(state)-4); /* do not write into reserved, for future removal */
+ memset(&state, 0, sizeof(state));
state.v1 = seed + PRIME32_1 + PRIME32_2;
state.v2 = seed + PRIME32_2;
state.v3 = seed + 0;
state.v4 = seed - PRIME32_1;
- memcpy(statePtr, &state, sizeof(state));
+ /* do not write into reserved, planned to be removed in a future version */
+ memcpy(statePtr, &state, sizeof(state) - sizeof(state.reserved));
return XXH_OK;
}
-XXH_FORCE_INLINE XXH_errorcode XXH32_update_endian (XXH32_state_t* state, const void* input, size_t len, XXH_endianess endian)
+FORCE_INLINE XXH_errorcode
+XXH32_update_endian(XXH32_state_t* state, const void* input, size_t len, XXH_endianess endian)
{
- const BYTE* p = (const BYTE*)input;
- const BYTE* const bEnd = p + len;
-
-#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
- if (input==NULL) return XXH_ERROR;
+ if (input==NULL)
+#if defined(XXH_ACCEPT_NULL_INPUT_POINTER) && (XXH_ACCEPT_NULL_INPUT_POINTER>=1)
+ return XXH_OK;
+#else
+ return XXH_ERROR;
#endif
- state->total_len_32 += (unsigned)len;
- state->large_len |= (len>=16) | (state->total_len_32>=16);
+ { const BYTE* p = (const BYTE*)input;
+ const BYTE* const bEnd = p + len;
- if (state->memsize + len < 16) { /* fill in tmp buffer */
- XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, len);
- state->memsize += (unsigned)len;
- return XXH_OK;
- }
+ state->total_len_32 += (unsigned)len;
+ state->large_len |= (len>=16) | (state->total_len_32>=16);
- if (state->memsize) { /* some data left from previous update */
- XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, 16-state->memsize);
- { const U32* p32 = state->mem32;
- state->v1 = XXH32_round(state->v1, XXH_readLE32(p32, endian)); p32++;
- state->v2 = XXH32_round(state->v2, XXH_readLE32(p32, endian)); p32++;
- state->v3 = XXH32_round(state->v3, XXH_readLE32(p32, endian)); p32++;
- state->v4 = XXH32_round(state->v4, XXH_readLE32(p32, endian)); p32++;
+ if (state->memsize + len < 16) { /* fill in tmp buffer */
+ XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, len);
+ state->memsize += (unsigned)len;
+ return XXH_OK;
}
- p += 16-state->memsize;
- state->memsize = 0;
- }
-
- if (p <= bEnd-16) {
- const BYTE* const limit = bEnd - 16;
- U32 v1 = state->v1;
- U32 v2 = state->v2;
- U32 v3 = state->v3;
- U32 v4 = state->v4;
- do {
- v1 = XXH32_round(v1, XXH_readLE32(p, endian)); p+=4;
- v2 = XXH32_round(v2, XXH_readLE32(p, endian)); p+=4;
- v3 = XXH32_round(v3, XXH_readLE32(p, endian)); p+=4;
- v4 = XXH32_round(v4, XXH_readLE32(p, endian)); p+=4;
- } while (p<=limit);
+ if (state->memsize) { /* some data left from previous update */
+ XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, 16-state->memsize);
+ { const U32* p32 = state->mem32;
+ state->v1 = XXH32_round(state->v1, XXH_readLE32(p32, endian)); p32++;
+ state->v2 = XXH32_round(state->v2, XXH_readLE32(p32, endian)); p32++;
+ state->v3 = XXH32_round(state->v3, XXH_readLE32(p32, endian)); p32++;
+ state->v4 = XXH32_round(state->v4, XXH_readLE32(p32, endian));
+ }
+ p += 16-state->memsize;
+ state->memsize = 0;
+ }
- state->v1 = v1;
- state->v2 = v2;
- state->v3 = v3;
- state->v4 = v4;
- }
+ if (p <= bEnd-16) {
+ const BYTE* const limit = bEnd - 16;
+ U32 v1 = state->v1;
+ U32 v2 = state->v2;
+ U32 v3 = state->v3;
+ U32 v4 = state->v4;
+
+ do {
+ v1 = XXH32_round(v1, XXH_readLE32(p, endian)); p+=4;
+ v2 = XXH32_round(v2, XXH_readLE32(p, endian)); p+=4;
+ v3 = XXH32_round(v3, XXH_readLE32(p, endian)); p+=4;
+ v4 = XXH32_round(v4, XXH_readLE32(p, endian)); p+=4;
+ } while (p<=limit);
+
+ state->v1 = v1;
+ state->v2 = v2;
+ state->v3 = v3;
+ state->v4 = v4;
+ }
- if (p < bEnd) {
- XXH_memcpy(state->mem32, p, (size_t)(bEnd-p));
- state->memsize = (unsigned)(bEnd-p);
+ if (p < bEnd) {
+ XXH_memcpy(state->mem32, p, (size_t)(bEnd-p));
+ state->memsize = (unsigned)(bEnd-p);
+ }
}
return XXH_OK;
}
+
XXH_PUBLIC_API XXH_errorcode XXH32_update (XXH32_state_t* state_in, const void* input, size_t len)
{
XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
@@ -450,40 +523,23 @@ XXH_PUBLIC_API XXH_errorcode XXH32_update (XXH32_state_t* state_in, const void*
}
-
-XXH_FORCE_INLINE U32 XXH32_digest_endian (const XXH32_state_t* state, XXH_endianess endian)
+FORCE_INLINE U32
+XXH32_digest_endian (const XXH32_state_t* state, XXH_endianess endian)
{
- const BYTE * p = (const BYTE*)state->mem32;
- const BYTE* const bEnd = (const BYTE*)(state->mem32) + state->memsize;
U32 h32;
if (state->large_len) {
- h32 = XXH_rotl32(state->v1, 1) + XXH_rotl32(state->v2, 7) + XXH_rotl32(state->v3, 12) + XXH_rotl32(state->v4, 18);
+ h32 = XXH_rotl32(state->v1, 1)
+ + XXH_rotl32(state->v2, 7)
+ + XXH_rotl32(state->v3, 12)
+ + XXH_rotl32(state->v4, 18);
} else {
h32 = state->v3 /* == seed */ + PRIME32_5;
}
h32 += state->total_len_32;
- while (p+4<=bEnd) {
- h32 += XXH_readLE32(p, endian) * PRIME32_3;
- h32 = XXH_rotl32(h32, 17) * PRIME32_4;
- p+=4;
- }
-
- while (p<bEnd) {
- h32 += (*p) * PRIME32_5;
- h32 = XXH_rotl32(h32, 11) * PRIME32_1;
- p++;
- }
-
- h32 ^= h32 >> 15;
- h32 *= PRIME32_2;
- h32 ^= h32 >> 13;
- h32 *= PRIME32_3;
- h32 ^= h32 >> 16;
-
- return h32;
+ return XXH32_finalize(h32, state->mem32, state->memsize, endian, XXH_aligned);
}
@@ -503,7 +559,7 @@ XXH_PUBLIC_API unsigned int XXH32_digest (const XXH32_state_t* state_in)
/*! Default XXH result types are basic unsigned 32 and 64 bits.
* The canonical representation follows human-readable write convention, aka big-endian (large digits first).
* These functions allow transformation of hash result into and from its canonical format.
-* This way, hash values can be written into a file or buffer, and remain comparable across different systems and programs.
+* This way, hash values can be written into a file or buffer, remaining comparable across different systems.
*/
XXH_PUBLIC_API void XXH32_canonicalFromHash(XXH32_canonical_t* dst, XXH32_hash_t hash)
@@ -522,18 +578,21 @@ XXH_PUBLIC_API XXH32_hash_t XXH32_hashFromCanonical(const XXH32_canonical_t* src
#ifndef XXH_NO_LONG_LONG
/* *******************************************************************
-* 64-bits hash functions
+* 64-bit hash functions
*********************************************************************/
/*====== Memory access ======*/
#ifndef MEM_MODULE
# define MEM_MODULE
-# if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
+# if !defined (__VMS) \
+ && (defined (__cplusplus) \
+ || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
# include <stdint.h>
typedef uint64_t U64;
# else
- typedef unsigned long long U64; /* if your compiler doesn't support unsigned long long, replace by another 64-bit type here. Note that xxhash.h will also need to be updated. */
+ /* if compiler doesn't support unsigned long long, replace by another 64-bit type */
+ typedef unsigned long long U64;
# endif
#endif
@@ -583,7 +642,7 @@ static U64 XXH_swap64 (U64 x)
}
#endif
-XXH_FORCE_INLINE U64 XXH_readLE64_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
+FORCE_INLINE U64 XXH_readLE64_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
{
if (align==XXH_unaligned)
return endian==XXH_littleEndian ? XXH_read64(ptr) : XXH_swap64(XXH_read64(ptr));
@@ -591,7 +650,7 @@ XXH_FORCE_INLINE U64 XXH_readLE64_align(const void* ptr, XXH_endianess endian, X
return endian==XXH_littleEndian ? *(const U64*)ptr : XXH_swap64(*(const U64*)ptr);
}
-XXH_FORCE_INLINE U64 XXH_readLE64(const void* ptr, XXH_endianess endian)
+FORCE_INLINE U64 XXH_readLE64(const void* ptr, XXH_endianess endian)
{
return XXH_readLE64_align(ptr, endian, XXH_unaligned);
}
@@ -626,14 +685,137 @@ static U64 XXH64_mergeRound(U64 acc, U64 val)
return acc;
}
-XXH_FORCE_INLINE U64 XXH64_endian_align(const void* input, size_t len, U64 seed, XXH_endianess endian, XXH_alignment align)
+static U64 XXH64_avalanche(U64 h64)
+{
+ h64 ^= h64 >> 33;
+ h64 *= PRIME64_2;
+ h64 ^= h64 >> 29;
+ h64 *= PRIME64_3;
+ h64 ^= h64 >> 32;
+ return h64;
+}
+
+
+#define XXH_get64bits(p) XXH_readLE64_align(p, endian, align)
+
+static U64
+XXH64_finalize(U64 h64, const void* ptr, size_t len,
+ XXH_endianess endian, XXH_alignment align)
+{
+ const BYTE* p = (const BYTE*)ptr;
+
+#define PROCESS1_64 \
+ h64 ^= (*p++) * PRIME64_5; \
+ h64 = XXH_rotl64(h64, 11) * PRIME64_1;
+
+#define PROCESS4_64 \
+ h64 ^= (U64)(XXH_get32bits(p)) * PRIME64_1; \
+ p+=4; \
+ h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
+
+#define PROCESS8_64 { \
+ U64 const k1 = XXH64_round(0, XXH_get64bits(p)); \
+ p+=8; \
+ h64 ^= k1; \
+ h64 = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4; \
+}
+
+ switch(len&31) {
+ case 24: PROCESS8_64;
+ /* fallthrough */
+ case 16: PROCESS8_64;
+ /* fallthrough */
+ case 8: PROCESS8_64;
+ return XXH64_avalanche(h64);
+
+ case 28: PROCESS8_64;
+ /* fallthrough */
+ case 20: PROCESS8_64;
+ /* fallthrough */
+ case 12: PROCESS8_64;
+ /* fallthrough */
+ case 4: PROCESS4_64;
+ return XXH64_avalanche(h64);
+
+ case 25: PROCESS8_64;
+ /* fallthrough */
+ case 17: PROCESS8_64;
+ /* fallthrough */
+ case 9: PROCESS8_64;
+ PROCESS1_64;
+ return XXH64_avalanche(h64);
+
+ case 29: PROCESS8_64;
+ /* fallthrough */
+ case 21: PROCESS8_64;
+ /* fallthrough */
+ case 13: PROCESS8_64;
+ /* fallthrough */
+ case 5: PROCESS4_64;
+ PROCESS1_64;
+ return XXH64_avalanche(h64);
+
+ case 26: PROCESS8_64;
+ /* fallthrough */
+ case 18: PROCESS8_64;
+ /* fallthrough */
+ case 10: PROCESS8_64;
+ PROCESS1_64;
+ PROCESS1_64;
+ return XXH64_avalanche(h64);
+
+ case 30: PROCESS8_64;
+ /* fallthrough */
+ case 22: PROCESS8_64;
+ /* fallthrough */
+ case 14: PROCESS8_64;
+ /* fallthrough */
+ case 6: PROCESS4_64;
+ PROCESS1_64;
+ PROCESS1_64;
+ return XXH64_avalanche(h64);
+
+ case 27: PROCESS8_64;
+ /* fallthrough */
+ case 19: PROCESS8_64;
+ /* fallthrough */
+ case 11: PROCESS8_64;
+ PROCESS1_64;
+ PROCESS1_64;
+ PROCESS1_64;
+ return XXH64_avalanche(h64);
+
+ case 31: PROCESS8_64;
+ /* fallthrough */
+ case 23: PROCESS8_64;
+ /* fallthrough */
+ case 15: PROCESS8_64;
+ /* fallthrough */
+ case 7: PROCESS4_64;
+ /* fallthrough */
+ case 3: PROCESS1_64;
+ /* fallthrough */
+ case 2: PROCESS1_64;
+ /* fallthrough */
+ case 1: PROCESS1_64;
+ /* fallthrough */
+ case 0: return XXH64_avalanche(h64);
+ }
+
+ /* impossible to reach */
+ assert(0);
+ return 0; /* unreachable, but some compilers complain without it */
+}
+
+FORCE_INLINE U64
+XXH64_endian_align(const void* input, size_t len, U64 seed,
+ XXH_endianess endian, XXH_alignment align)
{
const BYTE* p = (const BYTE*)input;
- const BYTE* const bEnd = p + len;
+ const BYTE* bEnd = p + len;
U64 h64;
-#define XXH_get64bits(p) XXH_readLE64_align(p, endian, align)
-#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
+#if defined(XXH_ACCEPT_NULL_INPUT_POINTER) && (XXH_ACCEPT_NULL_INPUT_POINTER>=1)
if (p==NULL) {
len=0;
bEnd=p=(const BYTE*)(size_t)32;
@@ -666,32 +848,7 @@ XXH_FORCE_INLINE U64 XXH64_endian_align(const void* input, size_t len, U64 seed,
h64 += (U64) len;
- while (p+8<=bEnd) {
- U64 const k1 = XXH64_round(0, XXH_get64bits(p));
- h64 ^= k1;
- h64 = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4;
- p+=8;
- }
-
- if (p+4<=bEnd) {
- h64 ^= (U64)(XXH_get32bits(p)) * PRIME64_1;
- h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
- p+=4;
- }
-
- while (p<bEnd) {
- h64 ^= (*p) * PRIME64_5;
- h64 = XXH_rotl64(h64, 11) * PRIME64_1;
- p++;
- }
-
- h64 ^= h64 >> 33;
- h64 *= PRIME64_2;
- h64 ^= h64 >> 29;
- h64 *= PRIME64_3;
- h64 ^= h64 >> 32;
-
- return h64;
+ return XXH64_finalize(h64, p, len, endian, align);
}
@@ -741,65 +898,71 @@ XXH_PUBLIC_API void XXH64_copyState(XXH64_state_t* dstState, const XXH64_state_t
XXH_PUBLIC_API XXH_errorcode XXH64_reset(XXH64_state_t* statePtr, unsigned long long seed)
{
XXH64_state_t state; /* using a local state to memcpy() in order to avoid strict-aliasing warnings */
- memset(&state, 0, sizeof(state)-8); /* do not write into reserved, for future removal */
+ memset(&state, 0, sizeof(state));
state.v1 = seed + PRIME64_1 + PRIME64_2;
state.v2 = seed + PRIME64_2;
state.v3 = seed + 0;
state.v4 = seed - PRIME64_1;
- memcpy(statePtr, &state, sizeof(state));
+ /* do not write into reserved, planned to be removed in a future version */
+ memcpy(statePtr, &state, sizeof(state) - sizeof(state.reserved));
return XXH_OK;
}
-XXH_FORCE_INLINE XXH_errorcode XXH64_update_endian (XXH64_state_t* state, const void* input, size_t len, XXH_endianess endian)
+FORCE_INLINE XXH_errorcode
+XXH64_update_endian (XXH64_state_t* state, const void* input, size_t len, XXH_endianess endian)
{
- const BYTE* p = (const BYTE*)input;
- const BYTE* const bEnd = p + len;
-
-#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
- if (input==NULL) return XXH_ERROR;
+ if (input==NULL)
+#if defined(XXH_ACCEPT_NULL_INPUT_POINTER) && (XXH_ACCEPT_NULL_INPUT_POINTER>=1)
+ return XXH_OK;
+#else
+ return XXH_ERROR;
#endif
- state->total_len += len;
-
- if (state->memsize + len < 32) { /* fill in tmp buffer */
- XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, len);
- state->memsize += (U32)len;
- return XXH_OK;
- }
+ { const BYTE* p = (const BYTE*)input;
+ const BYTE* const bEnd = p + len;
- if (state->memsize) { /* tmp buffer is full */
- XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, 32-state->memsize);
- state->v1 = XXH64_round(state->v1, XXH_readLE64(state->mem64+0, endian));
- state->v2 = XXH64_round(state->v2, XXH_readLE64(state->mem64+1, endian));
- state->v3 = XXH64_round(state->v3, XXH_readLE64(state->mem64+2, endian));
- state->v4 = XXH64_round(state->v4, XXH_readLE64(state->mem64+3, endian));
- p += 32-state->memsize;
- state->memsize = 0;
- }
+ state->total_len += len;
- if (p+32 <= bEnd) {
- const BYTE* const limit = bEnd - 32;
- U64 v1 = state->v1;
- U64 v2 = state->v2;
- U64 v3 = state->v3;
- U64 v4 = state->v4;
+ if (state->memsize + len < 32) { /* fill in tmp buffer */
+ XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, len);
+ state->memsize += (U32)len;
+ return XXH_OK;
+ }
- do {
- v1 = XXH64_round(v1, XXH_readLE64(p, endian)); p+=8;
- v2 = XXH64_round(v2, XXH_readLE64(p, endian)); p+=8;
- v3 = XXH64_round(v3, XXH_readLE64(p, endian)); p+=8;
- v4 = XXH64_round(v4, XXH_readLE64(p, endian)); p+=8;
- } while (p<=limit);
+ if (state->memsize) { /* tmp buffer is full */
+ XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, 32-state->memsize);
+ state->v1 = XXH64_round(state->v1, XXH_readLE64(state->mem64+0, endian));
+ state->v2 = XXH64_round(state->v2, XXH_readLE64(state->mem64+1, endian));
+ state->v3 = XXH64_round(state->v3, XXH_readLE64(state->mem64+2, endian));
+ state->v4 = XXH64_round(state->v4, XXH_readLE64(state->mem64+3, endian));
+ p += 32-state->memsize;
+ state->memsize = 0;
+ }
- state->v1 = v1;
- state->v2 = v2;
- state->v3 = v3;
- state->v4 = v4;
- }
+ if (p+32 <= bEnd) {
+ const BYTE* const limit = bEnd - 32;
+ U64 v1 = state->v1;
+ U64 v2 = state->v2;
+ U64 v3 = state->v3;
+ U64 v4 = state->v4;
+
+ do {
+ v1 = XXH64_round(v1, XXH_readLE64(p, endian)); p+=8;
+ v2 = XXH64_round(v2, XXH_readLE64(p, endian)); p+=8;
+ v3 = XXH64_round(v3, XXH_readLE64(p, endian)); p+=8;
+ v4 = XXH64_round(v4, XXH_readLE64(p, endian)); p+=8;
+ } while (p<=limit);
+
+ state->v1 = v1;
+ state->v2 = v2;
+ state->v3 = v3;
+ state->v4 = v4;
+ }
- if (p < bEnd) {
- XXH_memcpy(state->mem64, p, (size_t)(bEnd-p));
- state->memsize = (unsigned)(bEnd-p);
+ if (p < bEnd) {
+ XXH_memcpy(state->mem64, p, (size_t)(bEnd-p));
+ state->memsize = (unsigned)(bEnd-p);
+ }
}
return XXH_OK;
@@ -815,10 +978,8 @@ XXH_PUBLIC_API XXH_errorcode XXH64_update (XXH64_state_t* state_in, const void*
return XXH64_update_endian(state_in, input, len, XXH_bigEndian);
}
-XXH_FORCE_INLINE U64 XXH64_digest_endian (const XXH64_state_t* state, XXH_endianess endian)
+FORCE_INLINE U64 XXH64_digest_endian (const XXH64_state_t* state, XXH_endianess endian)
{
- const BYTE * p = (const BYTE*)state->mem64;
- const BYTE* const bEnd = (const BYTE*)state->mem64 + state->memsize;
U64 h64;
if (state->total_len >= 32) {
@@ -833,37 +994,12 @@ XXH_FORCE_INLINE U64 XXH64_digest_endian (const XXH64_state_t* state, XXH_endian
h64 = XXH64_mergeRound(h64, v3);
h64 = XXH64_mergeRound(h64, v4);
} else {
- h64 = state->v3 + PRIME64_5;
+ h64 = state->v3 /*seed*/ + PRIME64_5;
}
h64 += (U64) state->total_len;
- while (p+8<=bEnd) {
- U64 const k1 = XXH64_round(0, XXH_readLE64(p, endian));
- h64 ^= k1;
- h64 = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4;
- p+=8;
- }
-
- if (p+4<=bEnd) {
- h64 ^= (U64)(XXH_readLE32(p, endian)) * PRIME64_1;
- h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
- p+=4;
- }
-
- while (p<bEnd) {
- h64 ^= (*p) * PRIME64_5;
- h64 = XXH_rotl64(h64, 11) * PRIME64_1;
- p++;
- }
-
- h64 ^= h64 >> 33;
- h64 *= PRIME64_2;
- h64 ^= h64 >> 29;
- h64 *= PRIME64_3;
- h64 ^= h64 >> 32;
-
- return h64;
+ return XXH64_finalize(h64, state->mem64, (size_t)state->total_len, endian, XXH_aligned);
}
XXH_PUBLIC_API unsigned long long XXH64_digest (const XXH64_state_t* state_in)
diff --git a/lib/xxhash.h b/lib/xxhash.h
index 870a6d9..d6bad94 100644
--- a/lib/xxhash.h
+++ b/lib/xxhash.h
@@ -57,8 +57,8 @@ Q.Score is a measure of quality of the hash function.
It depends on successfully passing SMHasher test set.
10 is a perfect score.
-A 64-bits version, named XXH64, is available since r35.
-It offers much better speed, but for 64-bits applications only.
+A 64-bit version, named XXH64, is available since r35.
+It offers much better speed, but for 64-bit applications only.
Name Speed on 64 bits Speed on 32 bits
XXH64 13.8 GB/s 1.9 GB/s
XXH32 6.8 GB/s 6.0 GB/s
@@ -80,18 +80,19 @@ typedef enum { XXH_OK=0, XXH_ERROR } XXH_errorcode;
/* ****************************
-* API modifier
-******************************/
-/** XXH_PRIVATE_API
-* This is useful to include xxhash functions in `static` mode
-* in order to inline them, and remove their symbol from the public list.
-* Methodology :
-* #define XXH_PRIVATE_API
-* #include "xxhash.h"
-* `xxhash.c` is automatically included.
-* It's not useful to compile and link it as a separate module.
-*/
-#ifdef XXH_PRIVATE_API
+ * API modifier
+ ******************************/
+/** XXH_INLINE_ALL (and XXH_PRIVATE_API)
+ * This is useful to include xxhash functions in `static` mode
+ * in order to inline them, and remove their symbol from the public list.
+ * Inlining can offer dramatic performance improvement on small keys.
+ * Methodology :
+ * #define XXH_INLINE_ALL
+ * #include "xxhash.h"
+ * `xxhash.c` is automatically included.
+ * It's not useful to compile and link it as a separate module.
+ */
+#if defined(XXH_INLINE_ALL) || defined(XXH_PRIVATE_API)
# ifndef XXH_STATIC_LINKING_ONLY
# define XXH_STATIC_LINKING_ONLY
# endif
@@ -102,23 +103,24 @@ typedef enum { XXH_OK=0, XXH_ERROR } XXH_errorcode;
# elif defined(_MSC_VER)
# define XXH_PUBLIC_API static __inline
# else
-# define XXH_PUBLIC_API static /* this version may generate warnings for unused static functions; disable the relevant warning */
+ /* this version may generate warnings for unused static functions */
+# define XXH_PUBLIC_API static
# endif
#else
# define XXH_PUBLIC_API /* do nothing */
-#endif /* XXH_PRIVATE_API */
-
-/*!XXH_NAMESPACE, aka Namespace Emulation :
-
-If you want to include _and expose_ xxHash functions from within your own library,
-but also want to avoid symbol collisions with other libraries which may also include xxHash,
-
-you can use XXH_NAMESPACE, to automatically prefix any public symbol from xxhash library
-with the value of XXH_NAMESPACE (therefore, avoid NULL and numeric values).
-
-Note that no change is required within the calling program as long as it includes `xxhash.h` :
-regular symbol name will be automatically translated by this header.
-*/
+#endif /* XXH_INLINE_ALL || XXH_PRIVATE_API */
+
+/*! XXH_NAMESPACE, aka Namespace Emulation :
+ *
+ * If you want to include _and expose_ xxHash functions from within your own library,
+ * but also want to avoid symbol collisions with other libraries which may also include xxHash,
+ *
+ * you can use XXH_NAMESPACE, to automatically prefix any public symbol from xxhash library
+ * with the value of XXH_NAMESPACE (therefore, avoid NULL and numeric values).
+ *
+ * Note that no change is required within the calling program as long as it includes `xxhash.h` :
+ * regular symbol name will be automatically translated by this header.
+ */
#ifdef XXH_NAMESPACE
# define XXH_CAT(A,B) A##B
# define XXH_NAME2(A,B) XXH_CAT(A,B)
@@ -149,18 +151,18 @@ regular symbol name will be automatically translated by this header.
***************************************/
#define XXH_VERSION_MAJOR 0
#define XXH_VERSION_MINOR 6
-#define XXH_VERSION_RELEASE 2
+#define XXH_VERSION_RELEASE 5
#define XXH_VERSION_NUMBER (XXH_VERSION_MAJOR *100*100 + XXH_VERSION_MINOR *100 + XXH_VERSION_RELEASE)
XXH_PUBLIC_API unsigned XXH_versionNumber (void);
/*-**********************************************************************
-* 32-bits hash
+* 32-bit hash
************************************************************************/
-typedef unsigned int XXH32_hash_t;
+typedef unsigned int XXH32_hash_t;
/*! XXH32() :
- Calculate the 32-bits hash of sequence "length" bytes stored at memory address "input".
+ Calculate the 32-bit hash of sequence "length" bytes stored at memory address "input".
The memory between input & input+length must be valid (allocated and read-accessible).
"seed" can be used to alter the result predictably.
Speed on Core 2 Duo @ 3 GHz (single thread, SMHasher benchmark) : 5.4 GB/s */
@@ -177,26 +179,25 @@ XXH_PUBLIC_API XXH_errorcode XXH32_update (XXH32_state_t* statePtr, const void*
XXH_PUBLIC_API XXH32_hash_t XXH32_digest (const XXH32_state_t* statePtr);
/*
-These functions generate the xxHash of an input provided in multiple segments.
-Note that, for small input, they are slower than single-call functions, due to state management.
-For small input, prefer `XXH32()` and `XXH64()` .
-
-XXH state must first be allocated, using XXH*_createState() .
-
-Start a new hash by initializing state with a seed, using XXH*_reset().
-
-Then, feed the hash state by calling XXH*_update() as many times as necessary.
-Obviously, input must be allocated and read accessible.
-The function returns an error code, with 0 meaning OK, and any other value meaning there is an error.
-
-Finally, a hash value can be produced anytime, by using XXH*_digest().
-This function returns the nn-bits hash as an int or long long.
-
-It's still possible to continue inserting input into the hash state after a digest,
-and generate some new hashes later on, by calling again XXH*_digest().
-
-When done, free XXH state space if it was allocated dynamically.
-*/
+ * Streaming functions generate the xxHash of an input provided in multiple segments.
+ * Note that, for small input, they are slower than single-call functions, due to state management.
+ * For small inputs, prefer `XXH32()` and `XXH64()`, which are better optimized.
+ *
+ * XXH state must first be allocated, using XXH*_createState() .
+ *
+ * Start a new hash by initializing state with a seed, using XXH*_reset().
+ *
+ * Then, feed the hash state by calling XXH*_update() as many times as necessary.
+ * The function returns an error code, with 0 meaning OK, and any other value meaning there is an error.
+ *
+ * Finally, a hash value can be produced anytime, by using XXH*_digest().
+ * This function returns the nn-bits hash as an int or long long.
+ *
+ * It's still possible to continue inserting input into the hash state after a digest,
+ * and generate some new hashes later on, by calling again XXH*_digest().
+ *
+ * When done, free XXH state space if it was allocated dynamically.
+ */
/*====== Canonical representation ======*/
@@ -205,22 +206,22 @@ XXH_PUBLIC_API void XXH32_canonicalFromHash(XXH32_canonical_t* dst, XXH32_hash_t
XXH_PUBLIC_API XXH32_hash_t XXH32_hashFromCanonical(const XXH32_canonical_t* src);
/* Default result type for XXH functions are primitive unsigned 32 and 64 bits.
-* The canonical representation uses human-readable write convention, aka big-endian (large digits first).
-* These functions allow transformation of hash result into and from its canonical format.
-* This way, hash values can be written into a file / memory, and remain comparable on different systems and programs.
-*/
+ * The canonical representation uses human-readable write convention, aka big-endian (large digits first).
+ * These functions allow transformation of hash result into and from its canonical format.
+ * This way, hash values can be written into a file / memory, and remain comparable on different systems and programs.
+ */
#ifndef XXH_NO_LONG_LONG
/*-**********************************************************************
-* 64-bits hash
+* 64-bit hash
************************************************************************/
typedef unsigned long long XXH64_hash_t;
/*! XXH64() :
- Calculate the 64-bits hash of sequence of length "len" stored at memory address "input".
+ Calculate the 64-bit hash of sequence of length "len" stored at memory address "input".
"seed" can be used to alter the result predictably.
- This function runs faster on 64-bits systems, but slower on 32-bits systems (see benchmark).
+ This function runs faster on 64-bit systems, but slower on 32-bit systems (see benchmark).
*/
XXH_PUBLIC_API XXH64_hash_t XXH64 (const void* input, size_t length, unsigned long long seed);
@@ -241,48 +242,82 @@ XXH_PUBLIC_API XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src
#endif /* XXH_NO_LONG_LONG */
+
#ifdef XXH_STATIC_LINKING_ONLY
/* ================================================================================================
- This section contains definitions which are not guaranteed to remain stable.
+ This section contains declarations which are not guaranteed to remain stable.
They may change in future versions, becoming incompatible with a different version of the library.
- They shall only be used with static linking.
- Never use these definitions in association with dynamic linking !
+ These declarations should only be used with static linking.
+ Never use them in association with dynamic linking !
=================================================================================================== */
-/* These definitions are only meant to allow allocation of XXH state
- statically, on stack, or in a struct for example.
- Do not use members directly. */
-
- struct XXH32_state_s {
- unsigned total_len_32;
- unsigned large_len;
- unsigned v1;
- unsigned v2;
- unsigned v3;
- unsigned v4;
- unsigned mem32[4]; /* buffer defined as U32 for alignment */
- unsigned memsize;
- unsigned reserved; /* never read nor write, will be removed in a future version */
- }; /* typedef'd to XXH32_state_t */
-
-#ifndef XXH_NO_LONG_LONG
- struct XXH64_state_s {
- unsigned long long total_len;
- unsigned long long v1;
- unsigned long long v2;
- unsigned long long v3;
- unsigned long long v4;
- unsigned long long mem64[4]; /* buffer defined as U64 for alignment */
- unsigned memsize;
- unsigned reserved[2]; /* never read nor write, will be removed in a future version */
- }; /* typedef'd to XXH64_state_t */
+/* These definitions are only present to allow
+ * static allocation of XXH state, on stack or in a struct for example.
+ * Never **ever** use members directly. */
+
+#if !defined (__VMS) \
+ && (defined (__cplusplus) \
+ || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
+# include <stdint.h>
+
+struct XXH32_state_s {
+ uint32_t total_len_32;
+ uint32_t large_len;
+ uint32_t v1;
+ uint32_t v2;
+ uint32_t v3;
+ uint32_t v4;
+ uint32_t mem32[4];
+ uint32_t memsize;
+ uint32_t reserved; /* never read nor write, might be removed in a future version */
+}; /* typedef'd to XXH32_state_t */
+
+struct XXH64_state_s {
+ uint64_t total_len;
+ uint64_t v1;
+ uint64_t v2;
+ uint64_t v3;
+ uint64_t v4;
+ uint64_t mem64[4];
+ uint32_t memsize;
+ uint32_t reserved[2]; /* never read nor write, might be removed in a future version */
+}; /* typedef'd to XXH64_state_t */
+
+# else
+
+struct XXH32_state_s {
+ unsigned total_len_32;
+ unsigned large_len;
+ unsigned v1;
+ unsigned v2;
+ unsigned v3;
+ unsigned v4;
+ unsigned mem32[4];
+ unsigned memsize;
+ unsigned reserved; /* never read nor write, might be removed in a future version */
+}; /* typedef'd to XXH32_state_t */
+
+# ifndef XXH_NO_LONG_LONG /* remove 64-bit support */
+struct XXH64_state_s {
+ unsigned long long total_len;
+ unsigned long long v1;
+ unsigned long long v2;
+ unsigned long long v3;
+ unsigned long long v4;
+ unsigned long long mem64[4];
+ unsigned memsize;
+ unsigned reserved[2]; /* never read nor write, might be removed in a future version */
+}; /* typedef'd to XXH64_state_t */
+# endif
+
+# endif
+
+
+#if defined(XXH_INLINE_ALL) || defined(XXH_PRIVATE_API)
+# include "xxhash.c" /* include xxhash function bodies as `static`, for inlining */
#endif
-# ifdef XXH_PRIVATE_API
-# include "xxhash.c" /* include xxhash function bodies as `static`, for inlining */
-# endif
-
#endif /* XXH_STATIC_LINKING_ONLY */
generated by cgit v1.2.3 (git 2.25.1) at 2025-10-03 10:51:56 +0000