Importing mkelfimage from

ftp://ftp.lnxi.com/pub/mkelfImage/mkelfImage-2.7.tar.gz

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@3103 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1

22 files changed, 2399 insertions, 0 deletions

diff --git a/util/mkelfImage/kunzip_src/arch/alpha/include/stddef.h b/util/mkelfImage/kunzip_src/arch/alpha/include/stddef.h
new file mode 100644
index 0000000000..37ee13858a
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/alpha/include/stddef.h
@@ -0,0 +1,15 @@
+#ifndef ALPHA_STDDEF_H
+#define ALPHA_STDDEF_H
+
+typedef long ptrdiff_t;
+typedef unsigned long size_t;
+typedef long ssize_t;
+
+typedef int wchar_t;
+typedef unsigned int wint_t;
+
+#define NULL 0
+
+#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
+
+#endif /* ALPHA_STDDEF_H */
diff --git a/util/mkelfImage/kunzip_src/arch/alpha/include/stdint.h b/util/mkelfImage/kunzip_src/arch/alpha/include/stdint.h
new file mode 100644
index 0000000000..9b93a0c84f
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/alpha/include/stdint.h
@@ -0,0 +1,53 @@
+#ifndef ALPHA_STDINT_H
+#define ALPHA_STDINT_H
+
+/* Exact integral types */
+typedef unsigned char      uint8_t;
+typedef signed char        int8_t; 
+
+typedef unsigned short     uint16_t;
+typedef signed short       int16_t;
+
+typedef unsigned int       uint32_t;
+typedef signed int         int32_t;
+
+typedef unsigned long      uint64_t;
+typedef signed long        int64_t;
+
+
+/* Small types */
+typedef unsigned char      uint_least8_t;
+typedef signed char        int_least8_t; 
+
+typedef unsigned short     uint_least16_t;
+typedef signed short       int_least16_t;
+
+typedef unsigned int       uint_least32_t;
+typedef signed int         int_least32_t;
+
+typedef unsigned long      uint_least64_t;
+typedef signed long        int_least64_t;
+
+/* Fast Types */
+typedef unsigned char      uint_fast8_t;
+typedef signed char        int_fast8_t; 
+
+typedef unsigned long      uint_fast16_t;
+typedef signed long        int_fast16_t;
+
+typedef unsigned long      uint_fast32_t;
+typedef signed long        int_fast32_t;
+
+typedef unsigned long      uint_fast64_t;
+typedef signed long        int_fast64_t;
+
+/* Types for `void *' pointers.  */
+typedef long               intptr_t;
+typedef unsigned long      uintptr_t;
+
+/* Largest integral types */
+typedef long               intmax_t;
+typedef unsigned long      uintmax_t;
+
+
+#endif /* ALPHA_STDINT_H */
diff --git a/util/mkelfImage/kunzip_src/arch/alpha/include/va-alpha.h b/util/mkelfImage/kunzip_src/arch/alpha/include/va-alpha.h
new file mode 100644
index 0000000000..2528a712ad
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/alpha/include/va-alpha.h
@@ -0,0 +1,128 @@
+/* GNU C varargs and stdargs support for the DEC Alpha.  */
+
+/* Note:  We must use the name __builtin_savregs.  GCC attaches special
+   significance to that name.  In particular, regardless of where in a
+   function __builtin_saveregs is called, GCC moves the call up to the
+   very start of the function.  */
+
+/* Define __gnuc_va_list.  */
+
+#ifndef __GNUC_VA_LIST
+#define __GNUC_VA_LIST
+
+/* In VMS, __gnuc_va_list is simply char *; on OSF, it's a structure.  */
+
+#ifdef __VMS__
+typedef char *__gnuc_va_list;
+#else
+
+typedef struct {
+  char *__base;			/* Pointer to first integer register. */
+  int __offset;			/* Byte offset of args so far. */
+} __gnuc_va_list;
+#endif
+
+#endif /* __GNUC_VA_LIST */
+
+/* If this is for internal libc use, don't define anything but
+   __gnuc_va_list.  */
+
+#if !defined(__GNUC_VA_LIST_1) && (defined (_STDARG_H) || defined (_VARARGS_H))
+#define __GNUC_VA_LIST_1
+
+#define _VA_LIST
+#define _VA_LIST_
+
+typedef __gnuc_va_list va_list;
+
+#if !defined(_STDARG_H)
+
+/* varargs support */
+#define va_alist __builtin_va_alist
+#define va_dcl	 int __builtin_va_alist;...
+#ifdef __VMS__
+#define va_start(pvar) ((pvar) = __builtin_saveregs ())
+#else
+#define va_start(pvar) ((pvar) = * (__gnuc_va_list *) __builtin_saveregs ())
+#endif
+
+#else /* STDARG.H */
+
+/* ANSI alternative.  */
+
+/* Call __builtin_next_arg even though we aren't using its value, so that
+   we can verify that firstarg is correct.  */
+
+#ifdef __VMS__
+#define va_start(pvar, firstarg)				\
+  (__builtin_next_arg (firstarg),				\
+   (pvar) = __builtin_saveregs ())
+#else
+#define va_start(pvar, firstarg)				\
+  (__builtin_next_arg (firstarg),				\
+   (pvar) = *(__gnuc_va_list *) __builtin_saveregs ())
+#endif
+
+#endif /* _STDARG_H */
+
+#define va_end(__va)	((void) 0)
+
+/* Values returned by __builtin_classify_type.  */
+
+enum {
+  __no_type_class = -1,
+  __void_type_class,
+  __integer_type_class,
+  __char_type_class,
+  __enumeral_type_class,
+  __boolean_type_class,
+  __pointer_type_class,
+  __reference_type_class,
+  __offset_type_class,
+  __real_type_class,
+  __complex_type_class,
+  __function_type_class,
+  __method_type_class,
+  __record_type_class,
+  __union_type_class,
+  __array_type_class,
+  __string_type_class,
+  __set_type_class,
+  __file_type_class,
+  __lang_type_class
+};
+
+/* Note that parameters are always aligned at least to a word boundary
+   (when passed) regardless of what GCC's __alignof__ operator says.  */
+
+/* Avoid errors if compiling GCC v2 with GCC v1.  */
+#if __GNUC__ == 1
+#define __extension__
+#endif
+
+/* Get the size of a type in bytes, rounded up to an integral number
+   of words.  */
+
+#define __va_tsize(__type)  \
+  (((sizeof (__type) + __extension__ sizeof (long long) - 1)   \
+    / __extension__ sizeof (long long)) * __extension__ sizeof (long long))
+
+#ifdef __VMS__
+#define va_arg(__va, __type)						\
+(*(((__va) += __va_tsize (__type)),					\
+   (__type *)(void *)((__va) - __va_tsize (__type))))
+
+#else
+
+#define va_arg(__va, __type)						\
+(*(((__va).__offset += __va_tsize (__type)),				\
+   (__type *)(void *)((__va).__base + (__va).__offset			\
+	      - (((__builtin_classify_type (* (__type *) 0)		\
+		   == __real_type_class) && (__va).__offset <= (6 * 8))	\
+		 ? (6 * 8) + 8 : __va_tsize (__type)))))
+#endif
+
+/* Copy __gnuc_va_list into another variable of this type.  */
+#define __va_copy(dest, src) (dest) = (src)
+
+#endif /* __GNUC_VA_LIST_1 */
diff --git a/util/mkelfImage/kunzip_src/arch/alpha/lib/Makefile b/util/mkelfImage/kunzip_src/arch/alpha/lib/Makefile
new file mode 100644
index 0000000000..2fd15b06a0
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/alpha/lib/Makefile
@@ -0,0 +1,17 @@
+ARCH_OPTIONS=
+
+OBJECTS += __divqu.o __remqu.o __divlu.o __remlu.o
+
+$(OBJDIR)/__divqu.o: $(SRC)/arch/alpha/lib/divide.S
+	$(CC) $(CFLAGS) -DDIV -c -o $@ $^
+
+$(OBJDIR)/__remqu.o: $(SRC)/arch/alpha/lib/divide.S
+	$(CC) $(CFLAGS) -DREM -c -o $@ $^
+
+$(OBJDIR)/__divlu.o: $(SRC)/arch/alpha/lib/divide.S
+	$(CC) $(CFLAGS) -DDIV -DINTSIZE -c -o $@ $^
+
+$(OBJDIR)/__remlu.o: $(SRC)/arch/alpha/lib/divide.S
+	$(CC) $(CFLAGS) -DREM -DINTSIZE -c -o $@ $^
+
+
diff --git a/util/mkelfImage/kunzip_src/arch/alpha/lib/divide.S b/util/mkelfImage/kunzip_src/arch/alpha/lib/divide.S
new file mode 100644
index 0000000000..a4bc1f751e
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/alpha/lib/divide.S
@@ -0,0 +1,195 @@
+/*
+ * cpu/ev6/divide.S
+ *
+ * (C) 1995 Linus Torvalds
+ *
+ * Alpha division..
+ */
+
+/*
+ * The alpha chip doesn't provide hardware division, so we have to do it
+ * by hand.  The compiler expects the functions
+ *
+ *	__divqu: 64-bit unsigned long divide
+ *	__remqu: 64-bit unsigned long remainder
+ *	__divqs/__remqs: signed 64-bit
+ *	__divlu/__remlu: unsigned 32-bit
+ *	__divls/__remls: signed 32-bit
+ *
+ * These are not normal C functions: instead of the normal
+ * calling sequence, these expect their arguments in registers
+ * $24 and $25, and return the result in $27. Register $28 may
+ * be clobbered (assembly temporary), anything else must be saved. 
+ *
+ * In short: painful.
+ *
+ * This is a rather simple bit-at-a-time algorithm: it's very good
+ * at dividing random 64-bit numbers, but the more usual case where
+ * the divisor is small is handled better by the DEC algorithm
+ * using lookup tables. This uses much less memory, though, and is
+ * nicer on the cache.. Besides, I don't know the copyright status
+ * of the DEC code.
+ */
+
+/*
+ * My temporaries:
+ *	$0 - current bit
+ *	$1 - shifted divisor
+ *	$2 - modulus/quotient
+ *
+ *	$23 - return address
+ *	$24 - dividend
+ *	$25 - divisor
+ *
+ *	$27 - quotient/modulus
+ *	$28 - compare status
+ */
+
+#define halt .long 0
+
+/*
+ * Select function type and registers
+ */
+#define mask	$0
+#define divisor	$1
+#define compare $28
+#define tmp1	$3
+#define tmp2	$4
+
+#ifdef DIV
+#define DIV_ONLY(x,y...) x,##y
+#define MOD_ONLY(x,y...)
+#define func(x) __div##x
+#define modulus $2
+#define quotient $27
+#define GETSIGN(x) xor $24,$25,x
+#define STACK 48
+#else
+#define DIV_ONLY(x,y...)
+#define MOD_ONLY(x,y...) x,##y
+#define func(x) __rem##x
+#define modulus $27
+#define quotient $2
+#define GETSIGN(x) bis $24,$24,x
+#define STACK 32
+#endif
+
+/*
+ * For 32-bit operations, we need to extend to 64-bit
+ */
+#ifdef INTSIZE
+#define ufunction func(lu)
+#define sfunction func(l)
+#define LONGIFY(x) zapnot x,15,x
+#define SLONGIFY(x) addl x,0,x
+#else
+#define ufunction func(qu)
+#define sfunction func(q)
+#define LONGIFY(x)
+#define SLONGIFY(x)
+#endif
+
+.set noat
+.align	3
+.globl	ufunction
+.ent	ufunction
+ufunction:
+	subq	$30,STACK,$30
+	.frame	$30,STACK,$23
+	.prologue 0
+
+7:	stq	$1, 0($30)
+	bis	$25,$25,divisor
+	stq	$2, 8($30)
+	bis	$24,$24,modulus
+	stq	$0,16($30)
+	bis	$31,$31,quotient
+	LONGIFY(divisor)
+	stq	tmp1,24($30)
+	LONGIFY(modulus)
+	bis	$31,1,mask
+	DIV_ONLY(stq tmp2,32($30))
+	beq	divisor, 9f			/* div by zero */
+
+#ifdef INTSIZE
+	/*
+	 * shift divisor left, using 3-bit shifts for
+	 * 32-bit divides as we can't overflow. Three-bit
+	 * shifts will result in looping three times less
+	 * here, but can result in two loops more later.
+	 * Thus using a large shift isn't worth it (and
+	 * s8add pairs better than a sll..)
+	 */
+1:	cmpult	divisor,modulus,compare
+	s8addq	divisor,$31,divisor
+	s8addq	mask,$31,mask
+	bne	compare,1b
+#else
+1:	cmpult	divisor,modulus,compare
+	blt     divisor, 2f
+	addq	divisor,divisor,divisor
+	addq	mask,mask,mask
+	bne	compare,1b
+	unop
+#endif
+
+	/* ok, start to go right again.. */
+2:	DIV_ONLY(addq quotient,mask,tmp2)
+	srl	mask,1,mask
+	cmpule	divisor,modulus,compare
+	subq	modulus,divisor,tmp1
+	DIV_ONLY(cmovne compare,tmp2,quotient)
+	srl	divisor,1,divisor
+	cmovne	compare,tmp1,modulus
+	bne	mask,2b
+
+9:	ldq	$1, 0($30)
+	ldq	$2, 8($30)
+	ldq	$0,16($30)
+	ldq	tmp1,24($30)
+	DIV_ONLY(ldq tmp2,32($30))
+	addq	$30,STACK,$30
+	ret	$31,($23),1
+	.end	ufunction
+
+/*
+ * Uhh.. Ugly signed division. I'd rather not have it at all, but
+ * it's needed in some circumstances. There are different ways to
+ * handle this, really. This does:
+ * 	-a / b = a / -b = -(a / b)
+ *	-a % b = -(a % b)
+ *	a % -b = a % b
+ * which is probably not the best solution, but at least should
+ * have the property that (x/y)*y + (x%y) = x.
+ */
+.align 3
+.globl	sfunction
+.ent	sfunction
+sfunction:
+	subq	$30,STACK,$30
+	.frame	$30,STACK,$23
+	.prologue 0
+	bis	$24,$25,$28
+	SLONGIFY($28)
+	bge	$28,7b
+	stq	$24,0($30)
+	subq	$31,$24,$28
+	stq	$25,8($30)
+	cmovlt	$24,$28,$24	/* abs($24) */
+	stq	$23,16($30)
+	subq	$31,$25,$28
+	stq	tmp1,24($30)
+	cmovlt	$25,$28,$25	/* abs($25) */
+	unop
+	bsr	$23,ufunction
+	ldq	$24,0($30)
+	ldq	$25,8($30)
+	GETSIGN($28)
+	subq	$31,$27,tmp1
+	SLONGIFY($28)
+	ldq	$23,16($30)
+	cmovlt	$28,tmp1,$27
+	ldq	tmp1,24($30)
+	addq	$30,STACK,$30
+	ret	$31,($23),1
+	.end	sfunction
diff --git a/util/mkelfImage/kunzip_src/arch/alpha/lib/kunzip.lds b/util/mkelfImage/kunzip_src/arch/alpha/lib/kunzip.lds
new file mode 100644
index 0000000000..ed632686a8
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/alpha/lib/kunzip.lds
@@ -0,0 +1,50 @@
+PAGE_SIZE = 65536;
+BASIC_ALIGN = 8;
+OUTPUT_FORMAT("elf64-alpha")
+ENTRY(__start)
+SECTIONS
+{
+	. = PAGE_SIZE;
+	_start = .;
+	/*
+	 * First we place the code and read only data (typically const declared).
+	 * This get placed in rom.
+	 */
+	.text : {
+		_text = .;
+		*(.text)
+		_etext = .;
+		_rodata = .;
+		*(.rodata);
+		_erodata = .;
+	}
+	/* Global data */
+	.data : {
+		_data = .;
+		*(.data) 
+		CONSTRUCTORS
+		*(.got)
+		*(.sdata)
+		_edata = .;
+	}
+	
+	/* Important align _bss so bss may be zeroed with quadword access */
+	. = ALIGN(BASIC_ALIGN);
+	.bss : {
+		_bss = .;
+		*(.sbss) 
+		*(.scommon)
+		*(.bss)
+		*(COMMON)
+		*(.heap)
+		*(.stack)
+		/* Important align _ebss so bss may be zeroed with quadword access */
+		. = ALIGN(BASIC_ALIGN);
+		_ebss = .;
+	}
+	_end = .;
+
+	/DISCARD/ : {
+		*(*)
+	}
+}
diff --git a/util/mkelfImage/kunzip_src/arch/alpha/lib/start.S b/util/mkelfImage/kunzip_src/arch/alpha/lib/start.S
new file mode 100644
index 0000000000..a89f4ca6de
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/alpha/lib/start.S
@@ -0,0 +1,89 @@
+.set noat
+.set noreorder
+.text
+
+__original_registers:
+	.quad 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
+
+__entry:
+	.quad entry
+
+.globl __start
+__start:
+	br $27, __save_registers
+__save_registers:
+	lda $27, (__original_registers - __save_registers)($27)
+	stq $0,    0($27)
+	stq $1,    8($27)
+	stq $2,   16($27)
+	stq $3,   24($27)
+	stq $4,   32($27)
+	stq $5,   40($27)
+	stq $6,   48($27)
+	stq $7,   56($27)
+	stq $8,   64($27)
+	stq $9,   72($27)
+	stq $10,  80($27)
+	stq $11,  88($27)
+	stq $12,  96($27)
+	stq $13, 104($27)
+	stq $14, 112($27)
+	stq $15, 120($27)
+	stq $16, 128($27)
+	stq $17, 136($27)
+	stq $18, 144($27)
+	stq $19, 152($27)
+	stq $20, 160($27)
+	stq $21, 168($27)
+	stq $22, 176($27)
+	stq $23, 184($27)
+	stq $24, 192($27)
+	stq $25, 200($27)
+	stq $26, 208($27)
+	stq $28, 224($27)
+	stq $29, 232($27)
+	stq $30, 240($27)
+
+__normal_start:
+	ldgp $29, (__normal_start - __original_registers)($27)
+	lda  $30, _estack
+	jsr  $26, kunzip
+
+.globl jmp_to_program_entry
+jmp_to_program_entry:
+	br   $27, __restore_registers
+__restore_registers:
+	lda  $27,(__original_registers - __restore_registers)($27)
+	stq  $16, (__entry - __original_registers)($27)
+	ldq $0,    0($27)
+	ldq $1,    8($27)
+	ldq $2,   16($27)
+	ldq $3,   24($27)
+	ldq $4,   32($27)
+	ldq $5,   40($27)
+	ldq $6,   48($27)
+	ldq $7,   56($27)
+	ldq $8,   64($27)
+	ldq $9,   72($27)
+	ldq $10,  80($27)
+	ldq $11,  88($27)
+	ldq $12,  96($27)
+	ldq $13, 104($27)
+	ldq $14, 112($27)
+	ldq $15, 120($27)
+	ldq $16, 128($27)
+	ldq $17, 136($27)
+	ldq $18, 144($27)
+	ldq $19, 152($27)
+	ldq $20, 160($27)
+	ldq $21, 168($27)
+	ldq $22, 176($27)
+	ldq $23, 184($27)
+	ldq $24, 192($27)
+	ldq $25, 200($27)
+	ldq $26, 208($27)
+	ldq $28, 224($27)
+	ldq $29, 232($27)
+	ldq $30, 240($27)
+	ldq $27, (__entry - __original_registers)($27)
+	jsr $31, ($27)
diff --git a/util/mkelfImage/kunzip_src/arch/i386/include/stddef.h b/util/mkelfImage/kunzip_src/arch/i386/include/stddef.h
new file mode 100644
index 0000000000..88a3b48957
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/i386/include/stddef.h
@@ -0,0 +1,15 @@
+#ifndef I386_STDDEF_H
+#define I386_STDDEF_H
+
+typedef long ptrdiff_t;
+typedef unsigned long size_t;
+typedef long ssize_t;
+
+typedef int wchar_t;
+typedef unsigned int wint_t;
+
+#define NULL 0
+
+#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
+
+#endif I386_STDDEF_H
diff --git a/util/mkelfImage/kunzip_src/arch/i386/include/stdint.h b/util/mkelfImage/kunzip_src/arch/i386/include/stdint.h
new file mode 100644
index 0000000000..58d7519cde
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/i386/include/stdint.h
@@ -0,0 +1,52 @@
+#ifndef I386_STDINT_H
+#define I386_STDINT_H
+
+/* Exact integral types */
+typedef unsigned char      uint8_t;
+typedef signed char        int8_t; 
+
+typedef unsigned short     uint16_t;
+typedef signed short       int16_t;
+
+typedef unsigned int       uint32_t;
+typedef signed int         int32_t;
+
+typedef unsigned long long uint64_t;
+typedef signed long long   int64_t;
+
+/* Small types */
+typedef unsigned char      uint_least8_t;
+typedef signed char        int_least8_t; 
+
+typedef unsigned short     uint_least16_t;
+typedef signed short       int_least16_t;
+
+typedef unsigned int       uint_least32_t;
+typedef signed int         int_least32_t;
+
+typedef unsigned long long uint_least64_t;
+typedef signed long long   int_least64_t;
+
+/* Fast Types */
+typedef unsigned char      uint_fast8_t;
+typedef signed char        int_fast8_t; 
+
+typedef unsigned int       uint_fast16_t;
+typedef signed int         int_fast16_t;
+
+typedef unsigned int       uint_fast32_t;
+typedef signed int         int_fast32_t;
+
+typedef unsigned long long uint_fast64_t;
+typedef signed long long   int_fast64_t;
+
+/* Types for `void *' pointers.  */
+typedef int                intptr_t;
+typedef unsigned int       uintptr_t;
+
+/* Largest integral types */
+typedef long long int      intmax_t;
+typedef unsigned long long uintmax_t;
+
+
+#endif /* I386_STDINT_H */
diff --git a/util/mkelfImage/kunzip_src/arch/i386/lib/Makefile b/util/mkelfImage/kunzip_src/arch/i386/lib/Makefile
new file mode 100644
index 0000000000..df2474d210
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/i386/lib/Makefile
@@ -0,0 +1 @@
+ARCH_OPTIONS=
diff --git a/util/mkelfImage/kunzip_src/arch/i386/lib/kunzip.lds b/util/mkelfImage/kunzip_src/arch/i386/lib/kunzip.lds
new file mode 100644
index 0000000000..0e649005cf
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/i386/lib/kunzip.lds
@@ -0,0 +1,50 @@
+PAGE_SIZE = 4096;
+BASIC_ALIGN = 8;
+OUTPUT_FORMAT("elf32-i386")
+ENTRY(__start)
+SECTIONS
+{
+	. = PAGE_SIZE;
+	_start = .;
+	/*
+	 * First we place the code and read only data (typically const declared).
+	 * This get placed in rom.
+	 */
+	.text : {
+		_text = .;
+		*(.text)
+		_etext = .;
+		_rodata = .;
+		*(.rodata);
+		_erodata = .;
+	}
+	/* Global data */
+	.data : {
+		_data = .;
+		*(.data) 
+		CONSTRUCTORS
+		*(.got)
+		*(.sdata)
+		_edata = .;
+	}
+	
+	/* Important align _bss so bss may be zeroed with quadword access */
+	. = ALIGN(BASIC_ALIGN);
+	.bss : {
+		_bss = .;
+		*(.sbss) 
+		*(.scommon)
+		*(.bss)
+		*(COMMON)
+		*(.heap)
+		*(.stack)
+		/* Important align _ebss so bss may be zeroed with quadword access */
+		. = ALIGN(BASIC_ALIGN);
+		_ebss = .;
+	}
+	_end = .;
+
+	/DISCARD/ : {
+		*(*)
+	}
+}
diff --git a/util/mkelfImage/kunzip_src/arch/i386/lib/start.S b/util/mkelfImage/kunzip_src/arch/i386/lib/start.S
new file mode 100644
index 0000000000..fec7c9da66
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/arch/i386/lib/start.S
@@ -0,0 +1,36 @@
+.text
+
+__original_registers:
+	.long 0, 0, 0, 0, 0, 0 , 0 , 0
+__entry:
+	.long entry
+
+.globl __start
+__start:
+	movl %eax, 0+__original_registers
+	movl %ebx, 4+__original_registers
+	movl %ecx, 8+__original_registers
+	movl %edx, 12+__original_registers
+	movl %esi, 16+__original_registers
+	movl %edi, 20+__original_registers
+	movl %esp, 24+__original_registers
+	movl %ebp, 28+__original_registers
+
+__normal_start:
+	movl $_estack, %esp
+	call kunzip
+
+.globl jmp_to_program_entry
+jmp_to_program_entry:
+	movl 4(%esp), %eax
+	movl %eax, __entry
+	movl 0+__original_registers, %eax
+	movl 4+__original_registers, %ebx
+	movl 8+__original_registers, %ecx
+	movl 12+__original_registers, %edx
+	movl 16+__original_registers, %esi
+	movl 20+__original_registers, %edi
+	movl 24+__original_registers, %esp
+	movl 28+__original_registers, %ebp
+	jmp *__entry
+	
\ No newline at end of file
diff --git a/util/mkelfImage/kunzip_src/include/stdarg.h b/util/mkelfImage/kunzip_src/include/stdarg.h
new file mode 100644
index 0000000000..24f3383198
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/include/stdarg.h
@@ -0,0 +1,205 @@
+/* stdarg.h for GNU.
+   Note that the type used in va_arg is supposed to match the
+   actual type **after default promotions**.
+   Thus, va_arg (..., short) is not valid.  */
+
+#ifndef _STDARG_H
+#ifndef _ANSI_STDARG_H_
+#ifndef __need___va_list
+#define _STDARG_H
+#define _ANSI_STDARG_H_
+#endif /* not __need___va_list */
+#undef __need___va_list
+
+#ifdef __clipper__
+#include "va-clipper.h"
+#else
+#ifdef __m88k__
+#include "va-m88k.h"
+#else
+#ifdef __i860__
+#include "va-i860.h"
+#else
+#ifdef __hppa__
+#include "va-pa.h"
+#else
+#ifdef __mips__
+#include "va-mips.h"
+#else
+#ifdef __sparc__
+#include "va-sparc.h"
+#else
+#ifdef __i960__
+#include "va-i960.h"
+#else
+#ifdef __alpha__
+#include "va-alpha.h"
+#else
+#if defined (__H8300__) || defined (__H8300H__) || defined (__H8300S__)
+#include "va-h8300.h"
+#else
+#if defined (__PPC__) && (defined (_CALL_SYSV) || defined (_WIN32))
+#include "va-ppc.h"
+#else
+#ifdef __arc__
+#include "va-arc.h"
+#else
+#ifdef __M32R__
+#include "va-m32r.h"
+#else
+#ifdef __sh__
+#include "va-sh.h"
+#else
+#ifdef __mn10300__
+#include "va-mn10300.h"
+#else
+#ifdef __mn10200__
+#include "va-mn10200.h"
+#else
+#ifdef __v850__
+#include "va-v850.h"
+#else
+
+/* Define __gnuc_va_list.  */
+
+#ifndef __GNUC_VA_LIST
+#define __GNUC_VA_LIST
+#if defined(__svr4__) || defined(_AIX) || defined(_M_UNIX) || defined(__NetBSD__)
+typedef char *__gnuc_va_list;
+#else
+typedef void *__gnuc_va_list;
+#endif
+#endif
+
+/* Define the standard macros for the user,
+   if this invocation was from the user program.  */
+#ifdef _STDARG_H
+
+/* Amount of space required in an argument list for an arg of type TYPE.
+   TYPE may alternatively be an expression whose type is used.  */
+
+#if defined(sysV68)
+#define __va_rounded_size(TYPE)  \
+  (((sizeof (TYPE) + sizeof (short) - 1) / sizeof (short)) * sizeof (short))
+#else
+#define __va_rounded_size(TYPE)  \
+  (((sizeof (TYPE) + sizeof (int) - 1) / sizeof (int)) * sizeof (int))
+#endif
+
+#define va_start(AP, LASTARG) 						\
+ (AP = ((__gnuc_va_list) __builtin_next_arg (LASTARG)))
+
+#undef va_end
+void va_end (__gnuc_va_list);		/* Defined in libgcc.a */
+#define va_end(AP)	((void)0)
+
+/* We cast to void * and then to TYPE * because this avoids
+   a warning about increasing the alignment requirement.  */
+
+#if (defined (__arm__) && ! defined (__ARMEB__)) || defined (__i386__) || defined (__i860__) || defined (__ns32000__) || defined (__vax__)
+/* This is for little-endian machines; small args are padded upward.  */
+#define va_arg(AP, TYPE)						\
+ (AP = (__gnuc_va_list) ((char *) (AP) + __va_rounded_size (TYPE)),	\
+  *((TYPE *) (void *) ((char *) (AP) - __va_rounded_size (TYPE))))
+#else /* big-endian */
+/* This is for big-endian machines; small args are padded downward.  */
+#define va_arg(AP, TYPE)						\
+ (AP = (__gnuc_va_list) ((char *) (AP) + __va_rounded_size (TYPE)),	\
+  *((TYPE *) (void *) ((char *) (AP)					\
+		       - ((sizeof (TYPE) < __va_rounded_size (char)	\
+			   ? sizeof (TYPE) : __va_rounded_size (TYPE))))))
+#endif /* big-endian */
+
+/* Copy __gnuc_va_list into another variable of this type.  */
+#define __va_copy(dest, src) (dest) = (src)
+
+#endif /* _STDARG_H */
+
+#endif /* not v850 */
+#endif /* not mn10200 */
+#endif /* not mn10300 */
+#endif /* not sh */
+#endif /* not m32r */
+#endif /* not arc */
+#endif /* not powerpc with V.4 calling sequence */
+#endif /* not h8300 */
+#endif /* not alpha */
+#endif /* not i960 */
+#endif /* not sparc */
+#endif /* not mips */
+#endif /* not hppa */
+#endif /* not i860 */
+#endif /* not m88k */
+#endif /* not clipper */
+
+#ifdef _STDARG_H
+/* Define va_list, if desired, from __gnuc_va_list. */
+/* We deliberately do not define va_list when called from
+   stdio.h, because ANSI C says that stdio.h is not supposed to define
+   va_list.  stdio.h needs to have access to that data type, 
+   but must not use that name.  It should use the name __gnuc_va_list,
+   which is safe because it is reserved for the implementation.  */
+
+#ifdef _HIDDEN_VA_LIST  /* On OSF1, this means varargs.h is "half-loaded".  */
+#undef _VA_LIST
+#endif
+
+#ifdef _BSD_VA_LIST
+#undef _BSD_VA_LIST
+#endif
+
+#if defined(__svr4__) || (defined(_SCO_DS) && !defined(__VA_LIST))
+/* SVR4.2 uses _VA_LIST for an internal alias for va_list,
+   so we must avoid testing it and setting it here.
+   SVR4 uses _VA_LIST as a flag in stdarg.h, but we should
+   have no conflict with that.  */
+#ifndef _VA_LIST_
+#define _VA_LIST_
+#ifdef __i860__
+#ifndef _VA_LIST
+#define _VA_LIST va_list
+#endif
+#endif /* __i860__ */
+typedef __gnuc_va_list va_list;
+#ifdef _SCO_DS
+#define __VA_LIST
+#endif
+#endif /* _VA_LIST_ */
+#else /* not __svr4__ || _SCO_DS */
+
+/* The macro _VA_LIST_ is the same thing used by this file in Ultrix.
+   But on BSD NET2 we must not test or define or undef it.
+   (Note that the comments in NET 2's ansi.h
+   are incorrect for _VA_LIST_--see stdio.h!)  */
+#if !defined (_VA_LIST_) || defined (__BSD_NET2__) || defined (____386BSD____) || defined (__bsdi__) || defined (__sequent__) || defined (__FreeBSD__) || defined(WINNT)
+/* The macro _VA_LIST_DEFINED is used in Windows NT 3.5  */
+#ifndef _VA_LIST_DEFINED
+/* The macro _VA_LIST is used in SCO Unix 3.2.  */
+#ifndef _VA_LIST
+/* The macro _VA_LIST_T_H is used in the Bull dpx2  */
+#ifndef _VA_LIST_T_H
+typedef __gnuc_va_list va_list;
+#endif /* not _VA_LIST_T_H */
+#endif /* not _VA_LIST */
+#endif /* not _VA_LIST_DEFINED */
+#if !(defined (__BSD_NET2__) || defined (____386BSD____) || defined (__bsdi__) || defined (__sequent__) || defined (__FreeBSD__))
+#define _VA_LIST_
+#endif
+#ifndef _VA_LIST
+#define _VA_LIST
+#endif
+#ifndef _VA_LIST_DEFINED
+#define _VA_LIST_DEFINED
+#endif
+#ifndef _VA_LIST_T_H
+#define _VA_LIST_T_H
+#endif
+
+#endif /* not _VA_LIST_, except on certain systems */
+
+#endif /* not __svr4__ */
+
+#endif /* _STDARG_H */
+
+#endif /* not _ANSI_STDARG_H_ */
+#endif /* not _STDARG_H */
diff --git a/util/mkelfImage/kunzip_src/include/stdlib.h b/util/mkelfImage/kunzip_src/include/stdlib.h
new file mode 100644
index 0000000000..eb67d20fe7
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/include/stdlib.h
@@ -0,0 +1,14 @@
+#ifndef STDLIB_H
+#define STDLIB_H
+
+#include <stddef.h>
+
+extern void *malloc(size_t size);
+void free(void *ptr);
+
+/* Extensions to malloc... */
+typedef size_t malloc_mark_t;
+void malloc_mark(malloc_mark_t *place);
+void malloc_release(malloc_mark_t *place);
+
+#endif /* STDLIB_H */
diff --git a/util/mkelfImage/kunzip_src/include/string.h b/util/mkelfImage/kunzip_src/include/string.h
new file mode 100644
index 0000000000..dbaa1ecd17
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/include/string.h
@@ -0,0 +1,31 @@
+#ifndef STRING_H
+#define STRING_H
+
+#include <stddef.h>
+
+// yes, linux has fancy ones. We don't care. This stuff gets used 
+// hardly at all. And the pain of including those files is just too high.
+
+//extern inline void strcpy(char *dst, char *src) {while (*src) *dst++ = *src++;}
+
+//extern inline int strlen(char *src) { int i = 0; while (*src++) i++; return i;}
+
+static inline size_t strnlen(const char *src, size_t max) { 
+  int i = 0; 
+  if (max<0) {
+    while (*src++) 
+      i++; 
+    return i;
+  }
+  else {
+    while ((*src++) && (i < max)) 
+      i++; 
+    return i;
+  }
+}
+
+extern void *memcpy(void *dest, const void *src, size_t n);
+extern void *memset(void *s, int c, size_t n);
+extern int memcmp(const void *s1, const void *s2, size_t n);
+
+#endif /* STRING_H */
diff --git a/util/mkelfImage/kunzip_src/include/types.h b/util/mkelfImage/kunzip_src/include/types.h
new file mode 100644
index 0000000000..4b9dadfe55
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/include/types.h
@@ -0,0 +1,18 @@
+#ifndef TYPES_H
+#define TYPES_H
+
+#include <stdint.h>
+#include <stddef.h>
+
+typedef uint8_t  u8;
+typedef int8_t   s8;
+typedef uint16_t u16;
+typedef int16_t  s16;
+typedef uint32_t u32;
+typedef int32_t  s32;
+typedef uint64_t u64;
+typedef int64_t  s64;
+
+/* FIXME is BITS_PER_LONG needed? */
+
+#endif
diff --git a/util/mkelfImage/kunzip_src/lib/inflate.c b/util/mkelfImage/kunzip_src/lib/inflate.c
new file mode 100644
index 0000000000..894a98cd27
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/lib/inflate.c
@@ -0,0 +1,1178 @@
+#define DEBG(x)  
+#define DEBG1(x)  
+/* Taken from /usr/src/linux/lib/inflate.c [unmodified]
+   Used for start32, 1/11/2000
+   James Hendricks, Dale Webster */
+
+/* inflate.c -- Not copyrighted 1992 by Mark Adler
+   version c10p1, 10 January 1993 */
+
+/* 
+ * Adapted for booting Linux by Hannu Savolainen 1993
+ * based on gzip-1.0.3 
+ *
+ * Nicolas Pitre <nico@cam.org>, 1999/04/14 :
+ *   Little mods for all variable to reside either into rodata or bss segments
+ *   by marking constant variables with 'const' and initializing all the others
+ *   at run-time only.  This allows for the kernel uncompressor to run
+ *   directly from Flash or ROM memory on embeded systems.
+ */
+
+/*
+   Inflate deflated (PKZIP's method 8 compressed) data.  The compression
+   method searches for as much of the current string of bytes (up to a
+   length of 258) in the previous 32 K bytes.  If it doesn't find any
+   matches (of at least length 3), it codes the next byte.  Otherwise, it
+   codes the length of the matched string and its distance backwards from
+   the current position.  There is a single Huffman code that codes both
+   single bytes (called "literals") and match lengths.  A second Huffman
+   code codes the distance information, which follows a length code.  Each
+   length or distance code actually represents a base value and a number
+   of "extra" (sometimes zero) bits to get to add to the base value.  At
+   the end of each deflated block is a special end-of-block (EOB) literal/
+   length code.  The decoding process is basically: get a literal/length
+   code; if EOB then done; if a literal, emit the decoded byte; if a
+   length then get the distance and emit the referred-to bytes from the
+   sliding window of previously emitted data.
+
+   There are (currently) three kinds of inflate blocks: stored, fixed, and
+   dynamic.  The compressor deals with some chunk of data at a time, and
+   decides which method to use on a chunk-by-chunk basis.  A chunk might
+   typically be 32 K or 64 K.  If the chunk is incompressible, then the
+   "stored" method is used.  In this case, the bytes are simply stored as
+   is, eight bits per byte, with none of the above coding.  The bytes are
+   preceded by a count, since there is no longer an EOB code.
+
+   If the data is compressible, then either the fixed or dynamic methods
+   are used.  In the dynamic method, the compressed data is preceded by
+   an encoding of the literal/length and distance Huffman codes that are
+   to be used to decode this block.  The representation is itself Huffman
+   coded, and so is preceded by a description of that code.  These code
+   descriptions take up a little space, and so for small blocks, there is
+   a predefined set of codes, called the fixed codes.  The fixed method is
+   used if the block codes up smaller that way (usually for quite small
+   chunks), otherwise the dynamic method is used.  In the latter case, the
+   codes are customized to the probabilities in the current block, and so
+   can code it much better than the pre-determined fixed codes.
+ 
+   The Huffman codes themselves are decoded using a multi-level table
+   lookup, in order to maximize the speed of decoding plus the speed of
+   building the decoding tables.  See the comments below that precede the
+   lbits and dbits tuning parameters.
+ */
+
+
+/*
+   Notes beyond the 1.93a appnote.txt:
+
+   1. Distance pointers never point before the beginning of the output
+      stream.
+   2. Distance pointers can point back across blocks, up to 32k away.
+   3. There is an implied maximum of 7 bits for the bit length table and
+      15 bits for the actual data.
+   4. If only one code exists, then it is encoded using one bit.  (Zero
+      would be more efficient, but perhaps a little confusing.)  If two
+      codes exist, they are coded using one bit each (0 and 1).
+   5. There is no way of sending zero distance codes--a dummy must be
+      sent if there are none.  (History: a pre 2.0 version of PKZIP would
+      store blocks with no distance codes, but this was discovered to be
+      too harsh a criterion.)  Valid only for 1.93a.  2.04c does allow
+      zero distance codes, which is sent as one code of zero bits in
+      length.
+   6. There are up to 286 literal/length codes.  Code 256 represents the
+      end-of-block.  Note however that the static length tree defines
+      288 codes just to fill out the Huffman codes.  Codes 286 and 287
+      cannot be used though, since there is no length base or extra bits
+      defined for them.  Similarly, there are up to 30 distance codes.
+      However, static trees define 32 codes (all 5 bits) to fill out the
+      Huffman codes, but the last two had better not show up in the data.
+   7. Unzip can check dynamic Huffman blocks for complete code sets.
+      The exception is that a single code would not be complete (see #4).
+   8. The five bits following the block type is really the number of
+      literal codes sent minus 257.
+   9. Length codes 8,16,16 are interpreted as 13 length codes of 8 bits
+      (1+6+6).  Therefore, to output three times the length, you output
+      three codes (1+1+1), whereas to output four times the same length,
+      you only need two codes (1+3).  Hmm.
+  10. In the tree reconstruction algorithm, Code = Code + Increment
+      only if BitLength(i) is not zero.  (Pretty obvious.)
+  11. Correction: 4 Bits: # of Bit Length codes - 4     (4 - 19)
+  12. Note: length code 284 can represent 227-258, but length code 285
+      really is 258.  The last length deserves its own, short code
+      since it gets used a lot in very redundant files.  The length
+      258 is special since 258 - 3 (the min match length) is 255.
+  13. The literal/length and distance code bit lengths are read as a
+      single stream of lengths.  It is possible (and advantageous) for
+      a repeat code (16, 17, or 18) to go across the boundary between
+      the two sets of lengths.
+ */
+
+#ifdef RCSID
+static char rcsid[] = "#Id: inflate.c,v 0.14 1993/06/10 13:27:04 jloup Exp #";
+#endif
+
+#ifndef STATIC
+
+#if defined(STDC_HEADERS) || defined(HAVE_STDLIB_H)
+#  include <sys/types.h>
+#  include <stdlib.h>
+#endif
+
+#include "gzip.h"
+#define STATIC
+#endif /* !STATIC */
+	
+#define slide window
+
+/* Huffman code lookup table entry--this entry is four bytes for machines
+   that have 16-bit pointers (e.g. PC's in the small or medium model).
+   Valid extra bits are 0..13.  e == 15 is EOB (end of block), e == 16
+   means that v is a literal, 16 < e < 32 means that v is a pointer to
+   the next table, which codes e - 16 bits, and lastly e == 99 indicates
+   an unused code.  If a code with e == 99 is looked up, this implies an
+   error in the data. */
+struct huft {
+  uch e;                /* number of extra bits or operation */
+  uch b;                /* number of bits in this code or subcode */
+  union {
+    ush n;              /* literal, length base, or distance base */
+    struct huft *t;     /* pointer to next level of table */
+  } v;
+};
+
+
+/* Function prototypes */
+STATIC int huft_build OF((unsigned *, unsigned, unsigned, 
+		const ush *, const ush *, struct huft **, int *));
+STATIC int huft_free OF((struct huft *));
+STATIC int inflate_codes OF((struct huft *, struct huft *, int, int));
+STATIC int inflate_stored OF((void));
+STATIC int inflate_fixed OF((void));
+STATIC int inflate_dynamic OF((void));
+STATIC int inflate_block OF((int *));
+STATIC int inflate OF((void));
+
+
+/* The inflate algorithm uses a sliding 32 K byte window on the uncompressed
+   stream to find repeated byte strings.  This is implemented here as a
+   circular buffer.  The index is updated simply by incrementing and then
+   ANDing with 0x7fff (32K-1). */
+/* It is left to other modules to supply the 32 K area.  It is assumed
+   to be usable as if it were declared "uch slide[32768];" or as just
+   "uch *slide;" and then malloc'ed in the latter case.  The definition
+   must be in unzip.h, included above. */
+/* unsigned wp;             current position in slide */
+#define wp outcnt
+#define flush_output(w) (wp=(w),flush_window())
+
+/* Tables for deflate from PKZIP's appnote.txt. */
+static const unsigned border[] = {    /* Order of the bit length code lengths */
+        16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15};
+static const ush cplens[] = {         /* Copy lengths for literal codes 257..285 */
+        3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31,
+        35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0};
+        /* note: see note #13 above about the 258 in this list. */
+static const ush cplext[] = {         /* Extra bits for literal codes 257..285 */
+        0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2,
+        3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0, 99, 99}; /* 99==invalid */
+static const ush cpdist[] = {         /* Copy offsets for distance codes 0..29 */
+        1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193,
+        257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145,
+        8193, 12289, 16385, 24577};
+static const ush cpdext[] = {         /* Extra bits for distance codes */
+        0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6,
+        7, 7, 8, 8, 9, 9, 10, 10, 11, 11,
+        12, 12, 13, 13};
+
+
+
+/* Macros for inflate() bit peeking and grabbing.
+   The usage is:
+   
+        NEEDBITS(j)
+        x = b & mask_bits[j];
+        DUMPBITS(j)
+
+   where NEEDBITS makes sure that b has at least j bits in it, and
+   DUMPBITS removes the bits from b.  The macros use the variable k
+   for the number of bits in b.  Normally, b and k are register
+   variables for speed, and are initialized at the beginning of a
+   routine that uses these macros from a global bit buffer and count.
+
+   If we assume that EOB will be the longest code, then we will never
+   ask for bits with NEEDBITS that are beyond the end of the stream.
+   So, NEEDBITS should not read any more bytes than are needed to
+   meet the request.  Then no bytes need to be "returned" to the buffer
+   at the end of the last block.
+
+   However, this assumption is not true for fixed blocks--the EOB code
+   is 7 bits, but the other literal/length codes can be 8 or 9 bits.
+   (The EOB code is shorter than other codes because fixed blocks are
+   generally short.  So, while a block always has an EOB, many other
+   literal/length codes have a significantly lower probability of
+   showing up at all.)  However, by making the first table have a
+   lookup of seven bits, the EOB code will be found in that first
+   lookup, and so will not require that too many bits be pulled from
+   the stream.
+ */
+
+STATIC ulg bb;                         /* bit buffer */
+STATIC unsigned bk;                    /* bits in bit buffer */
+
+STATIC const ush mask_bits[] = {
+    0x0000,
+    0x0001, 0x0003, 0x0007, 0x000f, 0x001f, 0x003f, 0x007f, 0x00ff,
+    0x01ff, 0x03ff, 0x07ff, 0x0fff, 0x1fff, 0x3fff, 0x7fff, 0xffff
+};
+
+#define NEXTBYTE()  (uch)get_byte()
+#define NEEDBITS(n) {while(k<(n)){b|=((ulg)NEXTBYTE())<<k;k+=8;}}
+#define DUMPBITS(n) {b>>=(n);k-=(n);}
+
+
+/*
+   Huffman code decoding is performed using a multi-level table lookup.
+   The fastest way to decode is to simply build a lookup table whose
+   size is determined by the longest code.  However, the time it takes
+   to build this table can also be a factor if the data being decoded
+   is not very long.  The most common codes are necessarily the
+   shortest codes, so those codes dominate the decoding time, and hence
+   the speed.  The idea is you can have a shorter table that decodes the
+   shorter, more probable codes, and then point to subsidiary tables for
+   the longer codes.  The time it costs to decode the longer codes is
+   then traded against the time it takes to make longer tables.
+
+   This results of this trade are in the variables lbits and dbits
+   below.  lbits is the number of bits the first level table for literal/
+   length codes can decode in one step, and dbits is the same thing for
+   the distance codes.  Subsequent tables are also less than or equal to
+   those sizes.  These values may be adjusted either when all of the
+   codes are shorter than that, in which case the longest code length in
+   bits is used, or when the shortest code is *longer* than the requested
+   table size, in which case the length of the shortest code in bits is
+   used.
+
+   There are two different values for the two tables, since they code a
+   different number of possibilities each.  The literal/length table
+   codes 286 possible values, or in a flat code, a little over eight
+   bits.  The distance table codes 30 possible values, or a little less
+   than five bits, flat.  The optimum values for speed end up being
+   about one bit more than those, so lbits is 8+1 and dbits is 5+1.
+   The optimum values may differ though from machine to machine, and
+   possibly even between compilers.  Your mileage may vary.
+ */
+
+
+STATIC const int lbits = 9;          /* bits in base literal/length lookup table */
+STATIC const int dbits = 6;          /* bits in base distance lookup table */
+
+
+/* If BMAX needs to be larger than 16, then h and x[] should be ulg. */
+#define BMAX 16         /* maximum bit length of any code (16 for explode) */
+#define N_MAX 288       /* maximum number of codes in any set */
+
+
+STATIC unsigned hufts;         /* track memory usage */
+
+
+STATIC int huft_build(b, n, s, d, e, t, m)
+unsigned *b;            /* code lengths in bits (all assumed <= BMAX) */
+unsigned n;             /* number of codes (assumed <= N_MAX) */
+unsigned s;             /* number of simple-valued codes (0..s-1) */
+const ush *d;                 /* list of base values for non-simple codes */
+const ush *e;                 /* list of extra bits for non-simple codes */
+struct huft **t;        /* result: starting table */
+int *m;                 /* maximum lookup bits, returns actual */
+/* Given a list of code lengths and a maximum table size, make a set of
+   tables to decode that set of codes.  Return zero on success, one if
+   the given code set is incomplete (the tables are still built in this
+   case), two if the input is invalid (all zero length codes or an
+   oversubscribed set of lengths), and three if not enough memory. */
+{
+  unsigned a;                   /* counter for codes of length k */
+  unsigned c[BMAX+1];           /* bit length count table */
+  unsigned f;                   /* i repeats in table every f entries */
+  int g;                        /* maximum code length */
+  int h;                        /* table level */
+  register unsigned i;          /* counter, current code */
+  register unsigned j;          /* counter */
+  register int k;               /* number of bits in current code */
+  int l;                        /* bits per table (returned in m) */
+  register unsigned *p;         /* pointer into c[], b[], or v[] */
+  register struct huft *q;      /* points to current table */
+  struct huft r;                /* table entry for structure assignment */
+  struct huft *u[BMAX];         /* table stack */
+  unsigned v[N_MAX];            /* values in order of bit length */
+  register int w;               /* bits before this table == (l * h) */
+  unsigned x[BMAX+1];           /* bit offsets, then code stack */
+  unsigned *xp;                 /* pointer into x */
+  int y;                        /* number of dummy codes added */
+  unsigned z;                   /* number of entries in current table */
+
+DEBG("huft1 ");
+
+  /* Generate counts for each bit length */
+  memzero(c, sizeof(c));
+  p = b;  i = n;
+  do {
+    Tracecv(*p, (stderr, (n-i >= ' ' && n-i <= '~' ? "%c %d\n" : "0x%x %d\n"), 
+	    n-i, *p));
+    c[*p]++;                    /* assume all entries <= BMAX */
+    p++;                      /* Can't combine with above line (Solaris bug) */
+  } while (--i);
+  if (c[0] == n)                /* null input--all zero length codes */
+  {
+    *t = (struct huft *)NULL;
+    *m = 0;
+    return 0;
+  }
+
+DEBG("huft2 ");
+
+  /* Find minimum and maximum length, bound *m by those */
+  l = *m;
+  for (j = 1; j <= BMAX; j++)
+    if (c[j])
+      break;
+  k = j;                        /* minimum code length */
+  if ((unsigned)l < j)
+    l = j;
+  for (i = BMAX; i; i--)
+    if (c[i])
+      break;
+  g = i;                        /* maximum code length */
+  if ((unsigned)l > i)
+    l = i;
+  *m = l;
+
+DEBG("huft3 ");
+
+  /* Adjust last length count to fill out codes, if needed */
+  for (y = 1 << j; j < i; j++, y <<= 1)
+    if ((y -= c[j]) < 0)
+      return 2;                 /* bad input: more codes than bits */
+  if ((y -= c[i]) < 0)
+    return 2;
+  c[i] += y;
+
+DEBG("huft4 ");
+
+  /* Generate starting offsets into the value table for each length */
+  x[1] = j = 0;
+  p = c + 1;  xp = x + 2;
+  while (--i) {                 /* note that i == g from above */
+    *xp++ = (j += *p++);
+  }
+
+DEBG("huft5 ");
+
+  /* Make a table of values in order of bit lengths */
+  p = b;  i = 0;
+  do {
+    if ((j = *p++) != 0)
+      v[x[j]++] = i;
+  } while (++i < n);
+
+DEBG("h6 ");
+
+  /* Generate the Huffman codes and for each, make the table entries */
+  x[0] = i = 0;                 /* first Huffman code is zero */
+  p = v;                        /* grab values in bit order */
+  h = -1;                       /* no tables yet--level -1 */
+  w = -l;                       /* bits decoded == (l * h) */
+  u[0] = (struct huft *)NULL;   /* just to keep compilers happy */
+  q = (struct huft *)NULL;      /* ditto */
+  z = 0;                        /* ditto */
+DEBG("h6a ");
+
+  /* go through the bit lengths (k already is bits in shortest code) */
+  for (; k <= g; k++)
+  {
+DEBG("h6b ");
+    a = c[k];
+    while (a--)
+    {
+DEBG("h6b1 ");
+      /* here i is the Huffman code of length k bits for value *p */
+      /* make tables up to required level */
+      while (k > w + l)
+      {
+DEBG1("1 ");
+        h++;
+        w += l;                 /* previous table always l bits */
+
+        /* compute minimum size table less than or equal to l bits */
+        z = (z = g - w) > (unsigned)l ? l : z;  /* upper limit on table size */
+        if ((f = 1 << (j = k - w)) > a + 1)     /* try a k-w bit table */
+        {                       /* too few codes for k-w bit table */
+DEBG1("2 ");
+          f -= a + 1;           /* deduct codes from patterns left */
+          xp = c + k;
+          while (++j < z)       /* try smaller tables up to z bits */
+          {
+            if ((f <<= 1) <= *++xp)
+              break;            /* enough codes to use up j bits */
+            f -= *xp;           /* else deduct codes from patterns */
+          }
+        }
+DEBG1("3 ");
+        z = 1 << j;             /* table entries for j-bit table */
+
+        /* allocate and link in new table */
+        if ((q = (struct huft *)malloc((z + 1)*sizeof(struct huft))) ==
+            (struct huft *)NULL)
+        {
+          if (h)
+            huft_free(u[0]);
+          return 3;             /* not enough memory */
+        }
+DEBG1("4 ");
+        hufts += z + 1;         /* track memory usage */
+        *t = q + 1;             /* link to list for huft_free() */
+        *(t = &(q->v.t)) = (struct huft *)NULL;
+        u[h] = ++q;             /* table starts after link */
+
+DEBG1("5 ");
+        /* connect to last table, if there is one */
+        if (h)
+        {
+          x[h] = i;             /* save pattern for backing up */
+          r.b = (uch)l;         /* bits to dump before this table */
+          r.e = (uch)(16 + j);  /* bits in this table */
+          r.v.t = q;            /* pointer to this table */
+          j = i >> (w - l);     /* (get around Turbo C bug) */
+          u[h-1][j] = r;        /* connect to last table */
+        }
+DEBG1("6 ");
+      }
+DEBG("h6c ");
+
+      /* set up table entry in r */
+      r.b = (uch)(k - w);
+      if (p >= v + n)
+        r.e = 99;               /* out of values--invalid code */
+      else if (*p < s)
+      {
+        r.e = (uch)(*p < 256 ? 16 : 15);    /* 256 is end-of-block code */
+        r.v.n = (ush)(*p);             /* simple code is just the value */
+	p++;                           /* one compiler does not like *p++ */
+      }
+      else
+      {
+        r.e = (uch)e[*p - s];   /* non-simple--look up in lists */
+        r.v.n = d[*p++ - s];
+      }
+DEBG("h6d ");
+
+      /* fill code-like entries with r */
+      f = 1 << (k - w);
+      for (j = i >> w; j < z; j += f)
+        q[j] = r;
+
+      /* backwards increment the k-bit code i */
+      for (j = 1 << (k - 1); i & j; j >>= 1)
+        i ^= j;
+      i ^= j;
+
+      /* backup over finished tables */
+      while ((i & ((1 << w) - 1)) != x[h])
+      {
+        h--;                    /* don't need to update q */
+        w -= l;
+      }
+DEBG("h6e ");
+    }
+DEBG("h6f ");
+  }
+
+DEBG("huft7 ");
+
+  /* Return true (1) if we were given an incomplete table */
+  return y != 0 && g != 1;
+}
+
+
+
+STATIC int huft_free(t)
+struct huft *t;         /* table to free */
+/* Free the malloc'ed tables built by huft_build(), which makes a linked
+   list of the tables it made, with the links in a dummy first entry of
+   each table. */
+{
+  register struct huft *p, *q;
+
+
+  /* Go through linked list, freeing from the malloced (t[-1]) address. */
+  p = t;
+  while (p != (struct huft *)NULL)
+  {
+    q = (--p)->v.t;
+    free((char*)p);
+    p = q;
+  } 
+  return 0;
+}
+
+
+STATIC int inflate_codes(tl, td, bl, bd)
+struct huft *tl, *td;   /* literal/length and distance decoder tables */
+int bl, bd;             /* number of bits decoded by tl[] and td[] */
+/* inflate (decompress) the codes in a deflated (compressed) block.
+   Return an error code or zero if it all goes ok. */
+{
+  register unsigned e;  /* table entry flag/number of extra bits */
+  unsigned n, d;        /* length and index for copy */
+  unsigned w;           /* current window position */
+  struct huft *t;       /* pointer to table entry */
+  unsigned ml, md;      /* masks for bl and bd bits */
+  register ulg b;       /* bit buffer */
+  register unsigned k;  /* number of bits in bit buffer */
+
+
+  /* make local copies of globals */
+  b = bb;                       /* initialize bit buffer */
+  k = bk;
+  w = wp;                       /* initialize window position */
+
+  /* inflate the coded data */
+  ml = mask_bits[bl];           /* precompute masks for speed */
+  md = mask_bits[bd];
+  for (;;)                      /* do until end of block */
+  {
+    NEEDBITS((unsigned)bl)
+    if ((e = (t = tl + ((unsigned)b & ml))->e) > 16)
+      do {
+        if (e == 99)
+          return 1;
+        DUMPBITS(t->b)
+        e -= 16;
+        NEEDBITS(e)
+      } while ((e = (t = t->v.t + ((unsigned)b & mask_bits[e]))->e) > 16);
+    DUMPBITS(t->b)
+    if (e == 16)                /* then it's a literal */
+    {
+      slide[w++] = (uch)t->v.n;
+      Tracevv((stderr, "%c", slide[w-1]));
+      if (w == WSIZE)
+      {
+        flush_output(w);
+        w = 0;
+      }
+    }
+    else                        /* it's an EOB or a length */
+    {
+      /* exit if end of block */
+      if (e == 15)
+        break;
+
+      /* get length of block to copy */
+      NEEDBITS(e)
+      n = t->v.n + ((unsigned)b & mask_bits[e]);
+      DUMPBITS(e);
+
+      /* decode distance of block to copy */
+      NEEDBITS((unsigned)bd)
+      if ((e = (t = td + ((unsigned)b & md))->e) > 16)
+        do {
+          if (e == 99)
+            return 1;
+          DUMPBITS(t->b)
+          e -= 16;
+          NEEDBITS(e)
+        } while ((e = (t = t->v.t + ((unsigned)b & mask_bits[e]))->e) > 16);
+      DUMPBITS(t->b)
+      NEEDBITS(e)
+      d = w - t->v.n - ((unsigned)b & mask_bits[e]);
+      DUMPBITS(e)
+      Tracevv((stderr,"\\[%d,%d]", w-d, n));
+
+      /* do the copy */
+      do {
+        n -= (e = (e = WSIZE - ((d &= WSIZE-1) > w ? d : w)) > n ? n : e);
+#if !defined(NOMEMCPY) && !defined(DEBUG)
+        if (w - d >= e)         /* (this test assumes unsigned comparison) */
+        {
+          memcpy(slide + w, slide + d, e);
+          w += e;
+          d += e;
+        }
+        else                      /* do it slow to avoid memcpy() overlap */
+#endif /* !NOMEMCPY */
+          do {
+            slide[w++] = slide[d++];
+	    Tracevv((stderr, "%c", slide[w-1]));
+          } while (--e);
+        if (w == WSIZE)
+        {
+          flush_output(w);
+          w = 0;
+        }
+      } while (n);
+    }
+  }
+
+
+  /* restore the globals from the locals */
+  wp = w;                       /* restore global window pointer */
+  bb = b;                       /* restore global bit buffer */
+  bk = k;
+
+  /* done */
+  return 0;
+}
+
+
+
+STATIC int inflate_stored()
+/* "decompress" an inflated type 0 (stored) block. */
+{
+  unsigned n;           /* number of bytes in block */
+  unsigned w;           /* current window position */
+  register ulg b;       /* bit buffer */
+  register unsigned k;  /* number of bits in bit buffer */
+
+DEBG("<stor");
+
+  /* make local copies of globals */
+  b = bb;                       /* initialize bit buffer */
+  k = bk;
+  w = wp;                       /* initialize window position */
+
+
+  /* go to byte boundary */
+  n = k & 7;
+  DUMPBITS(n);
+
+
+  /* get the length and its complement */
+  NEEDBITS(16)
+  n = ((unsigned)b & 0xffff);
+  DUMPBITS(16)
+  NEEDBITS(16)
+  if (n != (unsigned)((~b) & 0xffff))
+    return 1;                   /* error in compressed data */
+  DUMPBITS(16)
+
+
+  /* read and output the compressed data */
+  while (n--)
+  {
+    NEEDBITS(8)
+    slide[w++] = (uch)b;
+    if (w == WSIZE)
+    {
+      flush_output(w);
+      w = 0;
+    }
+    DUMPBITS(8)
+  }
+
+
+  /* restore the globals from the locals */
+  wp = w;                       /* restore global window pointer */
+  bb = b;                       /* restore global bit buffer */
+  bk = k;
+
+  DEBG(">");
+  return 0;
+}
+
+
+
+STATIC int inflate_fixed()
+/* decompress an inflated type 1 (fixed Huffman codes) block.  We should
+   either replace this with a custom decoder, or at least precompute the
+   Huffman tables. */
+{
+  int i;                /* temporary variable */
+  struct huft *tl;      /* literal/length code table */
+  struct huft *td;      /* distance code table */
+  int bl;               /* lookup bits for tl */
+  int bd;               /* lookup bits for td */
+  unsigned l[288];      /* length list for huft_build */
+
+DEBG("<fix");
+
+  /* set up literal table */
+  for (i = 0; i < 144; i++)
+    l[i] = 8;
+  for (; i < 256; i++)
+    l[i] = 9;
+  for (; i < 280; i++)
+    l[i] = 7;
+  for (; i < 288; i++)          /* make a complete, but wrong code set */
+    l[i] = 8;
+  bl = 7;
+  if ((i = huft_build(l, 288, 257, cplens, cplext, &tl, &bl)) != 0)
+    return i;
+
+
+  /* set up distance table */
+  for (i = 0; i < 30; i++)      /* make an incomplete code set */
+    l[i] = 5;
+  bd = 5;
+  if ((i = huft_build(l, 30, 0, cpdist, cpdext, &td, &bd)) > 1)
+  {
+    huft_free(tl);
+
+    DEBG(">");
+    return i;
+  }
+
+
+  /* decompress until an end-of-block code */
+  if (inflate_codes(tl, td, bl, bd))
+    return 1;
+
+
+  /* free the decoding tables, return */
+  huft_free(tl);
+  huft_free(td);
+  return 0;
+}
+
+
+
+STATIC int inflate_dynamic()
+/* decompress an inflated type 2 (dynamic Huffman codes) block. */
+{
+  int i;                /* temporary variables */
+  unsigned j;
+  unsigned l;           /* last length */
+  unsigned m;           /* mask for bit lengths table */
+  unsigned n;           /* number of lengths to get */
+  struct huft *tl;      /* literal/length code table */
+  struct huft *td;      /* distance code table */
+  int bl;               /* lookup bits for tl */
+  int bd;               /* lookup bits for td */
+  unsigned nb;          /* number of bit length codes */
+  unsigned nl;          /* number of literal/length codes */
+  unsigned nd;          /* number of distance codes */
+#ifdef PKZIP_BUG_WORKAROUND
+  unsigned ll[288+32];  /* literal/length and distance code lengths */
+#else
+  unsigned ll[286+30];  /* literal/length and distance code lengths */
+#endif
+  register ulg b;       /* bit buffer */
+  register unsigned k;  /* number of bits in bit buffer */
+
+DEBG("<dyn");
+
+  /* make local bit buffer */
+  b = bb;
+  k = bk;
+
+
+  /* read in table lengths */
+  NEEDBITS(5)
+  nl = 257 + ((unsigned)b & 0x1f);      /* number of literal/length codes */
+  DUMPBITS(5)
+  NEEDBITS(5)
+  nd = 1 + ((unsigned)b & 0x1f);        /* number of distance codes */
+  DUMPBITS(5)
+  NEEDBITS(4)
+  nb = 4 + ((unsigned)b & 0xf);         /* number of bit length codes */
+  DUMPBITS(4)
+#ifdef PKZIP_BUG_WORKAROUND
+  if (nl > 288 || nd > 32)
+#else
+  if (nl > 286 || nd > 30)
+#endif
+    return 1;                   /* bad lengths */
+
+DEBG("dyn1 ");
+
+  /* read in bit-length-code lengths */
+  for (j = 0; j < nb; j++)
+  {
+    NEEDBITS(3)
+    ll[border[j]] = (unsigned)b & 7;
+    DUMPBITS(3)
+  }
+  for (; j < 19; j++)
+    ll[border[j]] = 0;
+
+DEBG("dyn2 ");
+
+  /* build decoding table for trees--single level, 7 bit lookup */
+  bl = 7;
+  if ((i = huft_build(ll, 19, 19, NULL, NULL, &tl, &bl)) != 0)
+  {
+    if (i == 1)
+      huft_free(tl);
+    return i;                   /* incomplete code set */
+  }
+
+DEBG("dyn3 ");
+
+  /* read in literal and distance code lengths */
+  n = nl + nd;
+  m = mask_bits[bl];
+  i = l = 0;
+  while ((unsigned)i < n)
+  {
+    NEEDBITS((unsigned)bl)
+    j = (td = tl + ((unsigned)b & m))->b;
+    DUMPBITS(j)
+    j = td->v.n;
+    if (j < 16)                 /* length of code in bits (0..15) */
+      ll[i++] = l = j;          /* save last length in l */
+    else if (j == 16)           /* repeat last length 3 to 6 times */
+    {
+      NEEDBITS(2)
+      j = 3 + ((unsigned)b & 3);
+      DUMPBITS(2)
+      if ((unsigned)i + j > n)
+        return 1;
+      while (j--)
+        ll[i++] = l;
+    }
+    else if (j == 17)           /* 3 to 10 zero length codes */
+    {
+      NEEDBITS(3)
+      j = 3 + ((unsigned)b & 7);
+      DUMPBITS(3)
+      if ((unsigned)i + j > n)
+        return 1;
+      while (j--)
+        ll[i++] = 0;
+      l = 0;
+    }
+    else                        /* j == 18: 11 to 138 zero length codes */
+    {
+      NEEDBITS(7)
+      j = 11 + ((unsigned)b & 0x7f);
+      DUMPBITS(7)
+      if ((unsigned)i + j > n)
+        return 1;
+      while (j--)
+        ll[i++] = 0;
+      l = 0;
+    }
+  }
+
+DEBG("dyn4 ");
+
+  /* free decoding table for trees */
+  huft_free(tl);
+
+DEBG("dyn5 ");
+
+  /* restore the global bit buffer */
+  bb = b;
+  bk = k;
+
+DEBG("dyn5a ");
+
+  /* build the decoding tables for literal/length and distance codes */
+  bl = lbits;
+  if ((i = huft_build(ll, nl, 257, cplens, cplext, &tl, &bl)) != 0)
+  {
+DEBG("dyn5b ");
+    if (i == 1) {
+      error(" incomplete literal tree\n");
+      huft_free(tl);
+    }
+    return i;                   /* incomplete code set */
+  }
+DEBG("dyn5c ");
+  bd = dbits;
+  if ((i = huft_build(ll + nl, nd, 0, cpdist, cpdext, &td, &bd)) != 0)
+  {
+DEBG("dyn5d ");
+    if (i == 1) {
+      error(" incomplete distance tree\n");
+#ifdef PKZIP_BUG_WORKAROUND
+      i = 0;
+    }
+#else
+      huft_free(td);
+    }
+    huft_free(tl);
+    return i;                   /* incomplete code set */
+#endif
+  }
+
+DEBG("dyn6 ");
+
+  /* decompress until an end-of-block code */
+  if (inflate_codes(tl, td, bl, bd))
+    return 1;
+
+DEBG("dyn7 ");
+
+  /* free the decoding tables, return */
+  huft_free(tl);
+  huft_free(td);
+
+  DEBG(">");
+  return 0;
+}
+
+
+
+STATIC int inflate_block(e)
+int *e;                 /* last block flag */
+/* decompress an inflated block */
+{
+  unsigned t;           /* block type */
+  register ulg b;       /* bit buffer */
+  register unsigned k;  /* number of bits in bit buffer */
+
+  DEBG("<blk");
+
+  /* make local bit buffer */
+  b = bb;
+  k = bk;
+
+
+  /* read in last block bit */
+  NEEDBITS(1)
+  *e = (int)b & 1;
+  DUMPBITS(1)
+
+
+  /* read in block type */
+  NEEDBITS(2)
+  t = (unsigned)b & 3;
+  DUMPBITS(2)
+
+
+  /* restore the global bit buffer */
+  bb = b;
+  bk = k;
+
+  /* inflate that block type */
+  if (t == 2)
+    return inflate_dynamic();
+  if (t == 0)
+    return inflate_stored();
+  if (t == 1)
+    return inflate_fixed();
+
+  DEBG(">");
+
+  /* bad block type */
+  return 2;
+}
+
+
+
+STATIC int inflate()
+/* decompress an inflated entry */
+{
+  int e;                /* last block flag */
+  int r;                /* result code */
+  unsigned h;           /* maximum struct huft's malloc'ed */
+  malloc_mark_t mark;
+
+  /* initialize window, bit buffer */
+  wp = 0;
+  bk = 0;
+  bb = 0;
+
+
+  /* decompress until the last block */
+  h = 0;
+  do {
+    hufts = 0;
+    malloc_mark(&mark);
+    if ((r = inflate_block(&e)) != 0) {
+      malloc_release(&mark);	    
+      return r;
+    }
+    malloc_release(&mark);
+    if (hufts > h)
+      h = hufts;
+  } while (!e);
+
+  /* Undo too much lookahead. The next read will be byte aligned so we
+   * can discard unused bits in the last meaningful byte.
+   */
+  while (bk >= 8) {
+    bk -= 8;
+    inptr--;
+  }
+
+  /* flush out slide */
+  flush_output(wp);
+
+
+  /* return success */
+  DBG(("<%u> ", h));
+  return 0;
+}
+
+/**********************************************************************
+ *
+ * The following are support routines for inflate.c
+ *
+ **********************************************************************/
+
+static ulg crc_32_tab[256];
+static ulg crc;		/* initialized in makecrc() so it'll reside in bss */
+#define CRC_VALUE (crc ^ 0xffffffffL)
+
+/*
+ * Code to compute the CRC-32 table. Borrowed from 
+ * gzip-1.0.3/makecrc.c.
+ */
+
+static void
+makecrc(void)
+{
+/* Not copyrighted 1990 Mark Adler	*/
+
+  unsigned long c;      /* crc shift register */
+  unsigned long e;      /* polynomial exclusive-or pattern */
+  int i;                /* counter for all possible eight bit values */
+  int k;                /* byte being shifted into crc apparatus */
+
+  /* terms of polynomial defining this crc (except x^32): */
+  static const int p[] = {0,1,2,4,5,7,8,10,11,12,16,22,23,26};
+
+  /* Make exclusive-or pattern from polynomial */
+  e = 0;
+  for (i = 0; i < sizeof(p)/sizeof(int); i++)
+    e |= 1L << (31 - p[i]);
+
+  crc_32_tab[0] = 0;
+
+  for (i = 1; i < 256; i++)
+  {
+    c = 0;
+    for (k = i | 256; k != 1; k >>= 1)
+    {
+      c = c & 1 ? (c >> 1) ^ e : c >> 1;
+      if (k & 1)
+        c ^= e;
+    }
+    crc_32_tab[i] = c;
+  }
+
+  /* this is initialized here so this code could reside in ROM */
+  crc = (ulg)0xffffffffL; /* shift register contents */
+}
+
+/* gzip flag byte */
+#define ASCII_FLAG   0x01 /* bit 0 set: file probably ASCII text */
+#define CONTINUATION 0x02 /* bit 1 set: continuation of multi-part gzip file */
+#define EXTRA_FIELD  0x04 /* bit 2 set: extra field present */
+#define ORIG_NAME    0x08 /* bit 3 set: original file name present */
+#define COMMENT      0x10 /* bit 4 set: file comment present */
+#define ENCRYPTED    0x20 /* bit 5 set: file is encrypted */
+#define RESERVED     0xC0 /* bit 6,7:   reserved */
+
+/*
+ * Do the uncompression!
+ */
+int gunzip(void)
+{
+    uch flags;
+    unsigned char magic[2]; /* magic header */
+    char method;
+    ulg orig_crc = 0;       /* original crc */
+    ulg orig_len = 0;       /* original uncompressed length */
+    int res;
+
+    magic[0] = (unsigned char)get_byte();
+    magic[1] = (unsigned char)get_byte();
+    method = (unsigned char)get_byte();
+
+    if (magic[0] != 037 ||
+	((magic[1] != 0213) && (magic[1] != 0236))) {
+	    error("bad gzip magic numbers");
+	    return -1;
+    }
+
+    /* We only support method #8, DEFLATED */
+    if (method != 8)  {
+	    error("internal error, invalid method");
+	    return -1;
+    }
+
+    flags  = (uch)get_byte();
+    if ((flags & ENCRYPTED) != 0) {
+	    error("Input is encrypted\n");
+	    return -1;
+    }
+    if ((flags & CONTINUATION) != 0) {
+	    error("Multi part input\n");
+	    return -1;
+    }
+    if ((flags & RESERVED) != 0) {
+	    error("Input has invalid flags\n");
+	    return -1;
+    }
+    (ulg)get_byte();	/* Get timestamp */
+    ((ulg)get_byte()) << 8;
+    ((ulg)get_byte()) << 16;
+    ((ulg)get_byte()) << 24;
+
+    (void)get_byte();  /* Ignore extra flags for the moment */
+    (void)get_byte();  /* Ignore OS type for the moment */
+
+    if ((flags & EXTRA_FIELD) != 0) {
+	    unsigned len = (unsigned)get_byte();
+	    len |= ((unsigned)get_byte())<<8;
+	    while (len--) (void)get_byte();
+    }
+
+    /* Get original file name if it was truncated */
+    if ((flags & ORIG_NAME) != 0) {
+	    /* Discard the old name */
+	    while (get_byte() != 0) /* null */ ;
+    } 
+
+    /* Discard file comment if any */
+    if ((flags & COMMENT) != 0) {
+	    while (get_byte() != 0) /* null */ ;
+    }
+
+    /* Decompress */
+    if ((res = inflate())) {
+	    switch (res) {
+	    case 0:
+		    break;
+	    case 1:
+		    error("invalid compressed format (err=1)");
+		    break;
+	    case 2:
+		    error("invalid compressed format (err=2)");
+		    break;
+	    case 3:
+		    error("out of memory");
+		    break;
+	    default:
+		    error("invalid compressed format (other)");
+	    }
+	    return -1;
+    }
+	    
+    /* Get the crc and original length */
+    /* crc32  (see algorithm.doc)
+     * uncompressed input size modulo 2^32
+     */
+    orig_crc = (ulg) get_byte();
+    orig_crc |= (ulg) get_byte() << 8;
+    orig_crc |= (ulg) get_byte() << 16;
+    orig_crc |= (ulg) get_byte() << 24;
+    
+    orig_len = (ulg) get_byte();
+    orig_len |= (ulg) get_byte() << 8;
+    orig_len |= (ulg) get_byte() << 16;
+    orig_len |= (ulg) get_byte() << 24;
+    
+    /* Validate decompression */
+    if (orig_crc != CRC_VALUE) {
+	    error("crc error");
+	    return -1;
+    }
+    if (orig_len != bytes_out) {
+	    error("length error");
+	    return -1;
+    }
+    return 0;
+}
+
+
diff --git a/util/mkelfImage/kunzip_src/lib/kunzip.c b/util/mkelfImage/kunzip_src/lib/kunzip.c
new file mode 100644
index 0000000000..4bfe504f70
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/lib/kunzip.c
@@ -0,0 +1,170 @@
+#include <string.h>
+#include <stdlib.h>
+
+/*
+ * gzip support routine declartions..
+ * =========================================================
+ */
+
+#ifdef DEBUG
+#  define Trace(x) 
+#  define Tracev(x)
+#  define Tracevv(x)
+#  define Tracec(c,x)
+#  define Tracecv(c,x)
+#  define DBG(x) printf x
+#else
+#  define Trace(x)
+#  define Tracev(x)
+#  define Tracevv(x)
+#  define Tracec(c,x)
+#  define Tracecv(c,x)
+#  define DBG(x)
+#endif
+
+void error(char *str)
+{
+	DBG(("%s\n", str));
+}
+
+static unsigned char *inbuf;	/* input buffer */
+static unsigned int insize;	/* valid bytes in inbuf */
+static unsigned int inptr;	/* index of next byte to be processed in inbuf */
+
+#if !defined(DEBUG)
+#define get_byte()  (inptr < insize ? inbuf[inptr++] : 0)
+#else
+static unsigned char get_byte(void)
+{
+	static int count;
+	unsigned char byte = (inptr < insize ? inbuf[inptr++] : 0);
+#if 0
+	printf("%02x ", byte);
+	if ((++count & 0x0f) == 0) {
+		printf("\n");
+	}
+#endif
+	return byte;
+}
+
+#endif
+
+static void flush_window(void);
+
+static long bytes_out;		/* total bytes compressed */
+static unsigned outcnt;		/* bytes in output buffer */
+
+#define WSIZE 0x8000		/* Window size must be at least 32k, and a power of two */
+static unsigned char window[WSIZE];	/* Sliding window buffer */
+
+/*
+ * gzip declarations
+ */
+
+#define OF(args)  args
+#define STATIC static
+
+
+#define memzero(s, n)     memset ((s), 0, (n))
+
+typedef unsigned char uch;
+typedef unsigned short ush;
+typedef unsigned long ulg;
+
+
+
+#include "inflate.c"
+
+
+/* Variables that gunzip doesn't need to see... */
+static unsigned char *output_ptr;
+static unsigned long end_offset;
+static struct unzip_region {
+	unsigned long start;
+	unsigned long end_offset;
+} unzip_region;
+
+/* Data provided by the header */
+extern unsigned char zipped_data[];
+extern unsigned char zipped_data_end[];
+extern unsigned char entry;
+/* Assembly language routines */
+extern void jmp_to_program_entry(void *);
+
+/* ===========================================================================
+ * Write the output window window[0..outcnt-1] and update crc and bytes_out.
+ * (Used for the decompressed data only.)
+ */
+static void flush_window(void)
+{
+	ulg c = crc;		/* temporary variable */
+	unsigned n;
+	unsigned long limit;
+	uch *in, *out, ch;
+
+	limit = outcnt;
+
+
+	n = 0;
+	in = window;
+	while (n < outcnt) {
+		limit = end_offset - bytes_out +n;
+		if (limit > outcnt) {
+			limit = outcnt;
+		}
+		out = output_ptr;
+		DBG(("flush 0x%08lx start 0x%08lx limit 0x%08lx\n", 
+			(unsigned long) out, (unsigned long)n, limit));
+		for (; n < limit; n++) {
+			ch = *out++ = *in++;
+			c = crc_32_tab[((int) c ^ ch) & 0xff] ^ (c >> 8);
+		}
+		crc = c;
+		bytes_out += (out - output_ptr);
+		output_ptr = out;
+		if (bytes_out == end_offset) {
+			if (output_ptr == (unsigned char *)(&unzip_region+1)) {
+				output_ptr = (unsigned char *)(unzip_region.start);
+				end_offset = unzip_region.end_offset;
+			} else {
+				output_ptr = (unsigned char *)&unzip_region;
+				end_offset += sizeof(unzip_region);
+			}
+		}
+	}
+	outcnt = 0;
+}
+
+
+void gunzip_setup(void)
+{
+	DBG(("gunzip_setup\n"));
+	outcnt = 0;
+	bytes_out = 0;
+
+	end_offset = sizeof(unzip_region);
+	output_ptr = (unsigned char *)&unzip_region;
+
+	inbuf = &zipped_data[0];
+	insize = zipped_data_end - zipped_data;
+	inptr = 0;
+
+	makecrc();
+	DBG(("gunzip_setup_done\n"));
+}
+
+
+int kunzip(int argc, char **argv)
+{
+	DBG(("kunzip\n"));
+	gunzip_setup();
+	DBG(("pre_gunzip\n"));
+	if (gunzip() != 0) {
+		error("gunzip failed");
+		while(1) {}
+		return -1;
+	}
+	DBG(("pre_jmp_to_program_entry: %p\n", &entry ));
+	jmp_to_program_entry(&entry);
+	return 0;
+}
diff --git a/util/mkelfImage/kunzip_src/lib/malloc.c b/util/mkelfImage/kunzip_src/lib/malloc.c
new file mode 100644
index 0000000000..8e43e0c226
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/lib/malloc.c
@@ -0,0 +1,42 @@
+#include <stdlib.h>
+
+extern unsigned char _heap, _eheap;
+static size_t free_mem_ptr = (size_t)&_heap;		/* Start of heap */
+static size_t free_mem_end_ptr = (size_t)&_eheap;	/* End of heap */
+
+
+void malloc_mark(malloc_mark_t *place)
+{
+	*place = free_mem_ptr;
+}
+
+void malloc_release(malloc_mark_t *ptr)
+{
+	free_mem_ptr = *ptr;
+}
+
+void *malloc(size_t size)
+{
+	void *p;
+
+	if (size < 0)
+		error("Error! malloc: Size < 0");
+	if (free_mem_ptr <= 0)
+		error("Error! malloc: Free_mem_ptr <= 0");
+
+	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
+
+	p = (void *) free_mem_ptr;
+	free_mem_ptr += size;
+
+	if (free_mem_ptr >= free_mem_end_ptr)
+		error("Error! malloc: Free_mem_ptr >= free_mem_end_ptr");
+
+
+	return p;
+}
+
+void free(void *where)
+{
+	/* Don't care */
+}
diff --git a/util/mkelfImage/kunzip_src/lib/memcmp.c b/util/mkelfImage/kunzip_src/lib/memcmp.c
new file mode 100644
index 0000000000..46f13a41bd
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/lib/memcmp.c
@@ -0,0 +1,17 @@
+#include <string.h>
+
+int memcmp(const void *src1, const void *src2, size_t bytes)
+{
+	const unsigned char *s1, *s2;
+	int result;
+	s1 = src1;
+	s2 = src2;
+	result = 0;
+	while((bytes > 0) && (result == 0)) {
+		result = *s1 - *s2;
+		bytes--;
+		s1++;
+		s2++;
+	}
+	return result;
+}
diff --git a/util/mkelfImage/kunzip_src/lib/memcpy.c b/util/mkelfImage/kunzip_src/lib/memcpy.c
new file mode 100644
index 0000000000..ad8e8bd3f0
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/lib/memcpy.c
@@ -0,0 +1,11 @@
+#include <string.h>
+void *memcpy(void *__dest, __const void *__src, size_t __n)
+{
+	int i;
+	char *d = (char *) __dest, *s = (char *) __src;
+
+	for (i = 0; i < __n; i++)
+		d[i] = s[i];
+
+	return __dest;
+}
diff --git a/util/mkelfImage/kunzip_src/lib/memset.c b/util/mkelfImage/kunzip_src/lib/memset.c
new file mode 100644
index 0000000000..c1bb4f841f
--- /dev/null
+++ b/util/mkelfImage/kunzip_src/lib/memset.c
@@ -0,0 +1,12 @@
+#include <string.h>
+
+void *memset(void *s, int c, size_t n)
+{
+	int i;
+	char *ss = (char *) s;
+
+	for (i = 0; i < n; i++)
+		ss[i] = c;
+
+	return s;
+}