6603632 PSARC/2007/458 User land UTF-8 text preparation functions

6603633 PSARC/2007/517 Uconv functions at libc

--HG--
rename : usr/src/uts/common/os/u8_textprep.c => usr/src/common/unicode/u8_textprep.c
rename : usr/src/uts/common/os/uconv.c => usr/src/common/unicode/uconv.c

13 files changed, 198 insertions, 96 deletions

diff --git a/usr/src/uts/common/os/u8_textprep.c b/usr/src/common/unicode/u8_textprep.c
index 7bb2312e92..37a01b83b1 100644
--- a/usr/src/uts/common/os/u8_textprep.c
+++ b/usr/src/common/unicode/u8_textprep.c
@@ -27,14 +27,16 @@
 
 
 /*
- * UTF-8 text preparation functions (PSARC/2007/149).
+ * UTF-8 text preparation functions (PSARC/2007/149, PSARC/2007/458).
  *
  * Man pages: u8_textprep_open(9F), u8_textprep_buf(9F), u8_textprep_close(9F),
- * u8_textprep_str(9F), u8_strcmp(9F), and u8_validate(9F).
+ * u8_textprep_str(9F), u8_strcmp(9F), and u8_validate(9F). See also
+ * the section 3C man pages.
  * Interface stability: Committed.
  */
 
 #include <sys/types.h>
+#ifdef	_KERNEL
 #include <sys/param.h>
 #include <sys/sysmacros.h>
 #include <sys/systm.h>
@@ -42,6 +44,10 @@
 #include <sys/kmem.h>
 #include <sys/ddi.h>
 #include <sys/sunddi.h>
+#else
+#include <sys/u8_textprep.h>
+#include <strings.h>
+#endif	/* _KERNEL */
 #include <sys/byteorder.h>
 #include <sys/errno.h>
 #include <sys/u8_textprep_data.h>
diff --git a/usr/src/uts/common/os/uconv.c b/usr/src/common/unicode/uconv.c
index 678928976d..f4e634a952 100644
--- a/usr/src/uts/common/os/uconv.c
+++ b/usr/src/common/unicode/uconv.c
@@ -2,9 +2,8 @@
  * CDDL HEADER START
  *
  * The contents of this file are subject to the terms of the
- * Common Development and Distribution License, Version 1.0 only
- * (the "License").  You may not use this file except in compliance
- * with the License.
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
  *
  * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
  * or http://www.opensolaris.org/os/licensing.
@@ -20,7 +19,7 @@
  * CDDL HEADER END
  */
 /*
- * Copyright 2006 Sun Microsystems, Inc.  All rights reserved.
+ * Copyright 2007 Sun Microsystems, Inc.  All rights reserved.
  * Use is subject to license terms.
  */
 
@@ -28,17 +27,24 @@
 
 /*
  * Unicode encoding conversion functions among UTF-8, UTF-16, and UTF-32.
- * Man page: /shared/sac/PSARC/2005/446/materials/uconv_functions.9f
- * Interface stability: Consolidation Private
+ * (PSARC/2005/446, PSARC/2007/038, PSARC/2007/517)
+ * Man pages: uconv_u16tou32(9F), uconv_u16tou8(9F), uconv_u32tou16(9F),
+ * uconv_u32tou8(9F), uconv_u8tou16(9F), and uconv_u8tou32(9F). See also
+ * the section 3C man pages.
+ * Interface stability: Committed
  */
 
 #include <sys/types.h>
+#ifdef	_KERNEL
 #include <sys/param.h>
 #include <sys/sysmacros.h>
 #include <sys/systm.h>
 #include <sys/debug.h>
 #include <sys/kmem.h>
 #include <sys/sunddi.h>
+#else
+#include <sys/u8_textprep.h>
+#endif	/* _KERNEL */
 #include <sys/byteorder.h>
 #include <sys/errno.h>
 
@@ -355,7 +361,7 @@ uconv_u16tou32(const uint16_t *u16s, size_t *utf16len,
 	 */
 	if (*utf16len > 0 && *utf32len > 0 && (flag & UCONV_OUT_EMIT_BOM))
 		u32s[u32l++] = (outendian) ? UCONV_BOM_NORMAL :
-			UCONV_BOM_SWAPPED_32;
+		    UCONV_BOM_SWAPPED_32;
 
 	/*
 	 * Do conversion; if encounter a surrogate pair, assemble high and
@@ -378,8 +384,8 @@ uconv_u16tou32(const uint16_t *u16s, size_t *utf16len,
 			if (! hi)
 				return (EILSEQ);
 			lo = (((hi - UCONV_U16_HI_MIN) * UCONV_U16_BIT_SHIFT +
-				lo - UCONV_U16_LO_MIN) & UCONV_U16_BIT_MASK)
-				+ UCONV_U16_START;
+			    lo - UCONV_U16_LO_MIN) & UCONV_U16_BIT_MASK)
+			    + UCONV_U16_START;
 			hi = 0;
 		} else if (hi) {
 			return (EILSEQ);
@@ -457,8 +463,8 @@ uconv_u16tou8(const uint16_t *u16s, size_t *utf16len,
 			if (! hi)
 				return (EILSEQ);
 			lo = (((hi - UCONV_U16_HI_MIN) * UCONV_U16_BIT_SHIFT +
-				lo - UCONV_U16_LO_MIN) & UCONV_U16_BIT_MASK)
-				+ UCONV_U16_START;
+			    lo - UCONV_U16_LO_MIN) & UCONV_U16_BIT_MASK)
+			    + UCONV_U16_START;
 			hi = 0;
 		} else if (hi) {
 			return (EILSEQ);
@@ -538,7 +544,7 @@ uconv_u32tou16(const uint32_t *u32s, size_t *utf32len,
 
 	if (*utf32len > 0 && *utf16len > 0 && (flag & UCONV_OUT_EMIT_BOM))
 		u16s[u16l++] = (outendian) ? UCONV_BOM_NORMAL :
-			UCONV_BOM_SWAPPED;
+		    UCONV_BOM_SWAPPED;
 
 	for (; u32l < *utf32len; u32l++) {
 		if (u32s[u32l] == 0 && do_not_ignore_null)
@@ -560,9 +566,9 @@ uconv_u32tou16(const uint32_t *u32s, size_t *utf32len,
 		 */
 		if (hi >= UCONV_U16_START) {
 			lo = ((hi - UCONV_U16_START) % UCONV_U16_BIT_SHIFT) +
-				UCONV_U16_LO_MIN;
+			    UCONV_U16_LO_MIN;
 			hi = ((hi - UCONV_U16_START) / UCONV_U16_BIT_SHIFT) +
-				UCONV_U16_HI_MIN;
+			    UCONV_U16_HI_MIN;
 
 			if ((u16l + 1) >= *utf16len)
 				return (E2BIG);
@@ -578,7 +584,7 @@ uconv_u32tou16(const uint32_t *u32s, size_t *utf32len,
 			if (u16l >= *utf16len)
 				return (E2BIG);
 			u16s[u16l++] = (outendian) ? (uint16_t)hi :
-				BSWAP_16(((uint16_t)hi));
+			    BSWAP_16(((uint16_t)hi));
 		}
 	}
 
@@ -686,7 +692,7 @@ uconv_u8tou16(const uchar_t *u8s, size_t *utf8len,
 
 	if (*utf8len > 0 && *utf16len > 0 && (flag & UCONV_OUT_EMIT_BOM))
 		u16s[u16l++] = (outendian) ? UCONV_BOM_NORMAL :
-			UCONV_BOM_SWAPPED;
+		    UCONV_BOM_SWAPPED;
 
 	for (; u8l < *utf8len; ) {
 		if (u8s[u8l] == 0 && do_not_ignore_null)
@@ -728,19 +734,19 @@ uconv_u8tou16(const uchar_t *u8s, size_t *utf8len,
 						return (EILSEQ);
 					first_b = 0;
 				} else if (lo < UCONV_U8_BYTE_MIN ||
-					lo > UCONV_U8_BYTE_MAX) {
+				    lo > UCONV_U8_BYTE_MAX) {
 					return (EILSEQ);
 				}
 				hi = (hi << UCONV_U8_BIT_SHIFT) |
-					(lo & UCONV_U8_BIT_MASK);
+				    (lo & UCONV_U8_BIT_MASK);
 			}
 		}
 
 		if (hi >= UCONV_U16_START) {
 			lo = ((hi - UCONV_U16_START) % UCONV_U16_BIT_SHIFT) +
-				UCONV_U16_LO_MIN;
+			    UCONV_U16_LO_MIN;
 			hi = ((hi - UCONV_U16_START) / UCONV_U16_BIT_SHIFT) +
-				UCONV_U16_HI_MIN;
+			    UCONV_U16_HI_MIN;
 
 			if ((u16l + 1) >= *utf16len)
 				return (E2BIG);
@@ -757,7 +763,7 @@ uconv_u8tou16(const uchar_t *u8s, size_t *utf8len,
 				return (E2BIG);
 
 			u16s[u16l++] = (outendian) ? (uint16_t)hi :
-				BSWAP_16(((uint16_t)hi));
+			    BSWAP_16(((uint16_t)hi));
 		}
 	}
 
@@ -797,7 +803,7 @@ uconv_u8tou32(const uchar_t *u8s, size_t *utf8len,
 
 	if (*utf8len > 0 && *utf32len > 0 && (flag & UCONV_OUT_EMIT_BOM))
 		u32s[u32l++] = (outendian) ? UCONV_BOM_NORMAL :
-			UCONV_BOM_SWAPPED_32;
+		    UCONV_BOM_SWAPPED_32;
 
 	for (; u8l < *utf8len; ) {
 		if (u8s[u8l] == 0 && do_not_ignore_null)
@@ -824,11 +830,11 @@ uconv_u8tou32(const uchar_t *u8s, size_t *utf8len,
 						return (EILSEQ);
 					first_b = 0;
 				} else if (c < UCONV_U8_BYTE_MIN ||
-					c > UCONV_U8_BYTE_MAX) {
+				    c > UCONV_U8_BYTE_MAX) {
 					return (EILSEQ);
 				}
 				hi = (hi << UCONV_U8_BIT_SHIFT) |
-					(c & UCONV_U8_BIT_MASK);
+				    (c & UCONV_U8_BIT_MASK);
 			}
 		}
 
diff --git a/usr/src/lib/libc/Makefile.targ b/usr/src/lib/libc/Makefile.targ
index 38d73741c3..b8237246a4 100644
--- a/usr/src/lib/libc/Makefile.targ
+++ b/usr/src/lib/libc/Makefile.targ
@@ -276,6 +276,10 @@ $(DTRACEOBJS:%=pics/%): $(SRC)/common/dtrace/$$(@F:.o=.c)
 	$(COMPILE.c) -o $@ $(SRC)/common/dtrace/$(@F:.o=.c)
 	$(POST_PROCESS_O)
 
+$(UNICODEOBJS:%=pics/%): $(SRC)/common/unicode/$$(@F:.o=.c)
+	$(COMPILE.c) -o $@ $(SRC)/common/unicode/$(@F:.o=.c)
+	$(POST_PROCESS_O)
+
 # DTrace rules
 pics/%.o: ../port/threads/%.d $(THREADSOBJS:%=pics/%)
 	$(COMPILE.d) -C -xlazyload -s $< -o $@ $(THREADSOBJS:%=pics/%)
diff --git a/usr/src/lib/libc/amd64/Makefile b/usr/src/lib/libc/amd64/Makefile
index 3f4cd4fea2..0e2d924805 100644
--- a/usr/src/lib/libc/amd64/Makefile
+++ b/usr/src/lib/libc/amd64/Makefile
@@ -762,6 +762,10 @@ THREADSMACHOBJS=		\
 THREADSASMOBJS=			\
 	asm_subr.o
 
+UNICODEOBJS=			\
+	u8_textprep.o		\
+	uconv.o
+
 UNWINDMACHOBJS=			\
 	call_frame_inst.o	\
 	eh_frame.o		\
@@ -866,6 +870,7 @@ MOSTOBJS=			\
 	$(THREADSOBJS)		\
 	$(THREADSMACHOBJS)	\
 	$(THREADSASMOBJS)	\
+	$(UNICODEOBJS)		\
 	$(UNWINDMACHOBJS)	\
 	$(UNWINDASMOBJS)	\
 	$(COMSYSOBJS)		\
@@ -974,6 +979,7 @@ SRCS=							\
 	$(TPOOLOBJS:%.o=../port/tpool/%.c)		\
 	$(THREADSOBJS:%.o=../port/threads/%.c)		\
 	$(THREADSMACHOBJS:%.o=threads/%.c)		\
+	$(UNICODEOBJS:%.o=$(SRC)/common/unicode/%.c)	\
 	$(UNWINDMACHOBJS:%.o=unwind/%.c)		\
 	$(FPOBJS:%.o=fp/%.c)				\
 	$(I386FPOBJS:%.o=../i386/fp/%.c)		\
diff --git a/usr/src/lib/libc/i386/Makefile.com b/usr/src/lib/libc/i386/Makefile.com
index cbc0843f80..6fe1a1c5cc 100644
--- a/usr/src/lib/libc/i386/Makefile.com
+++ b/usr/src/lib/libc/i386/Makefile.com
@@ -804,6 +804,10 @@ THREADSMACHOBJS=		\
 THREADSASMOBJS=			\
 	asm_subr.o
 
+UNICODEOBJS=			\
+	u8_textprep.o		\
+	uconv.o
+
 UNWINDMACHOBJS=			\
 	unwind.o
 
@@ -904,6 +908,7 @@ MOSTOBJS=			\
 	$(THREADSOBJS)		\
 	$(THREADSMACHOBJS)	\
 	$(THREADSASMOBJS)	\
+	$(UNICODEOBJS)		\
 	$(UNWINDMACHOBJS)	\
 	$(UNWINDASMOBJS)	\
 	$(COMSYSOBJS)		\
@@ -1035,6 +1040,7 @@ SRCS=							\
 	$(TPOOLOBJS:%.o=../port/tpool/%.c)		\
 	$(THREADSOBJS:%.o=../port/threads/%.c)		\
 	$(THREADSMACHOBJS:%.o=../$(MACH)/threads/%.c)	\
+	$(UNICODEOBJS:%.o=$(SRC)/common/unicode/%.c)	\
 	$(UNWINDMACHOBJS:%.o=../port/unwind/%.c)	\
 	$(FPOBJS:%.o=../$(MACH)/fp/%.c)			\
 	$(LIBCBASE)/gen/ecvt.c				\
diff --git a/usr/src/lib/libc/port/mapfile-vers b/usr/src/lib/libc/port/mapfile-vers
index 5b8612a19c..161df2efb8 100644
--- a/usr/src/lib/libc/port/mapfile-vers
+++ b/usr/src/lib/libc/port/mapfile-vers
@@ -123,6 +123,15 @@ SUNW_1.23 {		# SunOS 5.11 (Solaris 11)
 	timer_getoverrun;
 	timer_gettime;
 	timer_settime;
+	u8_strcmp;
+	u8_textprep_str;
+	u8_validate;
+	uconv_u16tou32;
+	uconv_u16tou8;
+	uconv_u32tou16;
+	uconv_u32tou8;
+	uconv_u8tou16;
+	uconv_u8tou32;
 	uucopy;
 	uucopystr;
 	vforkx;
diff --git a/usr/src/lib/libc/sparc/Makefile b/usr/src/lib/libc/sparc/Makefile
index ce8627d963..c67463652b 100644
--- a/usr/src/lib/libc/sparc/Makefile
+++ b/usr/src/lib/libc/sparc/Makefile
@@ -830,6 +830,10 @@ THREADSMACHOBJS=		\
 THREADSASMOBJS=			\
 	asm_subr.o
 
+UNICODEOBJS=			\
+	u8_textprep.o		\
+	uconv.o
+
 UNWINDMACHOBJS=			\
 	unwind.o
 
@@ -932,6 +936,7 @@ MOSTOBJS=			\
 	$(THREADSOBJS)		\
 	$(THREADSMACHOBJS)	\
 	$(THREADSASMOBJS)	\
+	$(UNICODEOBJS)		\
 	$(UNWINDMACHOBJS)	\
 	$(UNWINDASMOBJS)	\
 	$(COMSYSOBJS)		\
@@ -1053,6 +1058,7 @@ SRCS=							\
 	$(TPOOLOBJS:%.o=../port/tpool/%.c)		\
 	$(THREADSOBJS:%.o=../port/threads/%.c)		\
 	$(THREADSMACHOBJS:%.o=../$(MACH)/threads/%.c)	\
+	$(UNICODEOBJS:%.o=$(SRC)/common/unicode/%.c)	\
 	$(UNWINDMACHOBJS:%.o=../port/unwind/%.c)	\
 	$(FPOBJS:%.o=../$(MACH)/fp/%.c)			\
 	$(LIBCBASE)/crt/_ftou.c				\
diff --git a/usr/src/lib/libc/sparcv9/Makefile b/usr/src/lib/libc/sparcv9/Makefile
index a2a6084b87..c114126bfe 100644
--- a/usr/src/lib/libc/sparcv9/Makefile
+++ b/usr/src/lib/libc/sparcv9/Makefile
@@ -777,6 +777,10 @@ THREADSMACHOBJS=		\
 THREADSASMOBJS=			\
 	asm_subr.o
 
+UNICODEOBJS=			\
+	u8_textprep.o		\
+	uconv.o
+
 UNWINDMACHOBJS=			\
 	unwind.o
 
@@ -875,6 +879,7 @@ MOSTOBJS=			\
 	$(THREADSOBJS)		\
 	$(THREADSMACHOBJS)	\
 	$(THREADSASMOBJS)	\
+	$(UNICODEOBJS)		\
 	$(UNWINDMACHOBJS)	\
 	$(UNWINDASMOBJS)	\
 	$(COMSYSOBJS)		\
@@ -986,6 +991,7 @@ SRCS=							\
 	$(TPOOLOBJS:%.o=../port/tpool/%.c)		\
 	$(THREADSOBJS:%.o=../port/threads/%.c)		\
 	$(THREADSMACHOBJS:%.o=../$(MACH)/threads/%.c)	\
+	$(UNICODEOBJS:%.o=$(SRC)/common/unicode/%.c)	\
 	$(UNWINDMACHOBJS:%.o=../port/unwind/%.c)	\
 	$(FPOBJS:%.o=../$(MACH)/fp/%.c)			\
 	$(FPOBJS64:%.o=$(LIBCBASE)/fp/%.c)		\
diff --git a/usr/src/pkgdefs/SUNWhea/prototype_com b/usr/src/pkgdefs/SUNWhea/prototype_com
index 74727e566a..a726754cda 100644
--- a/usr/src/pkgdefs/SUNWhea/prototype_com
+++ b/usr/src/pkgdefs/SUNWhea/prototype_com
@@ -1293,6 +1293,7 @@ f none usr/include/sys/tuneable.h 644 root bin
 f none usr/include/sys/turnstile.h 644 root bin
 f none usr/include/sys/types.h 644 root bin
 f none usr/include/sys/types32.h 644 root bin
+f none usr/include/sys/u8_textprep.h 644 root bin
 f none usr/include/sys/uadmin.h 644 root bin
 f none usr/include/sys/ucontext.h 644 root bin
 f none usr/include/sys/uio.h 644 root bin
diff --git a/usr/src/uts/common/Makefile.rules b/usr/src/uts/common/Makefile.rules
index d9e8413118..2d10397e7c 100644
--- a/usr/src/uts/common/Makefile.rules
+++ b/usr/src/uts/common/Makefile.rules
@@ -951,6 +951,10 @@ $(OBJS_DIR)/%.o:		$(COMMONBASE)/util/%.c
 	$(COMPILE.c) -o $@ $<
 	$(CTFCONVERT_O)
 
+$(OBJS_DIR)/%.o:		$(COMMONBASE)/unicode/%.c
+	$(COMPILE.c) -o $@ $<
+	$(CTFCONVERT_O)
+
 $(OBJS_DIR)/%.o:		$(UTSBASE)/common/vm/%.c
 	$(COMPILE.c) -o $@ $<
 	$(CTFCONVERT_O)
@@ -1673,6 +1677,9 @@ $(LINTS_DIR)/%.ln:		$(COMMONBASE)/tsol/%.c
 $(LINTS_DIR)/%.ln:		$(COMMONBASE)/util/%.c
 	@($(LHEAD) $(LINT.c) $< $(LTAIL))
 
+$(LINTS_DIR)/%.ln:		$(COMMONBASE)/unicode/%.c
+	@($(LHEAD) $(LINT.c) $< $(LTAIL))
+
 $(LINTS_DIR)/%.ln:		$(UTSBASE)/common/vm/%.c
 	@($(LHEAD) $(LINT.c) $< $(LTAIL))
 
diff --git a/usr/src/uts/common/sys/Makefile b/usr/src/uts/common/sys/Makefile
index a2d180667b..b5bf354dcb 100644
--- a/usr/src/uts/common/sys/Makefile
+++ b/usr/src/uts/common/sys/Makefile
@@ -534,6 +534,7 @@ CHKHDRS=			\
 	turnstile.h		\
 	types.h			\
 	types32.h		\
+	u8_textprep.h		\
 	u8_textprep_data.h	\
 	uadmin.h		\
 	ucred.h			\
diff --git a/usr/src/uts/common/sys/sunddi.h b/usr/src/uts/common/sys/sunddi.h
index 126b1b5e4a..69b62b20c4 100644
--- a/usr/src/uts/common/sys/sunddi.h
+++ b/usr/src/uts/common/sys/sunddi.h
@@ -59,6 +59,7 @@
 #ifdef _KERNEL
 #include <sys/ddi_obsolete.h>
 #endif
+#include <sys/u8_textprep.h>
 
 #ifdef	__cplusplus
 extern "C" {
@@ -441,76 +442,6 @@ extern int ddi_strtol(const char *, char **, int, long *);
 extern int ddi_strtoul(const char *, char **, int, unsigned long *);
 
 /*
- * Unicode encoding conversion functions and their macros.
- */
-#define	UCONV_IN_BIG_ENDIAN		0x0001
-#define	UCONV_OUT_BIG_ENDIAN		0x0002
-#define	UCONV_IN_SYSTEM_ENDIAN		0x0004
-#define	UCONV_OUT_SYSTEM_ENDIAN		0x0008
-#define	UCONV_IN_LITTLE_ENDIAN		0x0010
-#define	UCONV_OUT_LITTLE_ENDIAN		0x0020
-#define	UCONV_IGNORE_NULL		0x0040
-#define	UCONV_IN_ACCEPT_BOM		0x0080
-#define	UCONV_OUT_EMIT_BOM		0x0100
-
-extern int uconv_u16tou32(const uint16_t *, size_t *, uint32_t *, size_t *,
-	int);
-extern int uconv_u16tou8(const uint16_t *, size_t *, uchar_t *, size_t *, int);
-extern int uconv_u32tou16(const uint32_t *, size_t *, uint16_t *, size_t *,
-	int);
-extern int uconv_u32tou8(const uint32_t *, size_t *, uchar_t *, size_t *, int);
-extern int uconv_u8tou16(const uchar_t *, size_t *, uint16_t *, size_t *, int);
-extern int uconv_u8tou32(const uchar_t *, size_t *, uint32_t *, size_t *, int);
-
-/*
- * UTF-8 text preparation functions and their macros.
- *
- * Among the macros defined, U8_CANON_DECOMP, U8_COMPAT_DECOMP, and
- * U8_CANON_COMP are not public interfaces and must not be used directly
- * at the flag input argument.
- */
-#define	U8_STRCMP_CS			(0x00000001)
-#define	U8_STRCMP_CI_UPPER		(0x00000002)
-#define	U8_STRCMP_CI_LOWER		(0x00000004)
-
-#define	U8_CANON_DECOMP			(0x00000010)
-#define	U8_COMPAT_DECOMP		(0x00000020)
-#define	U8_CANON_COMP			(0x00000040)
-
-#define	U8_STRCMP_NFD			(U8_CANON_DECOMP)
-#define	U8_STRCMP_NFC			(U8_CANON_DECOMP | U8_CANON_COMP)
-#define	U8_STRCMP_NFKD			(U8_COMPAT_DECOMP)
-#define	U8_STRCMP_NFKC			(U8_COMPAT_DECOMP | U8_CANON_COMP)
-
-#define	U8_TEXTPREP_TOUPPER		(U8_STRCMP_CI_UPPER)
-#define	U8_TEXTPREP_TOLOWER		(U8_STRCMP_CI_LOWER)
-
-#define	U8_TEXTPREP_NFD			(U8_STRCMP_NFD)
-#define	U8_TEXTPREP_NFC			(U8_STRCMP_NFC)
-#define	U8_TEXTPREP_NFKD		(U8_STRCMP_NFKD)
-#define	U8_TEXTPREP_NFKC		(U8_STRCMP_NFKC)
-
-#define	U8_TEXTPREP_IGNORE_NULL		(0x00010000)
-#define	U8_TEXTPREP_IGNORE_INVALID	(0x00020000)
-#define	U8_TEXTPREP_NOWAIT		(0x00040000)
-
-#define	U8_UNICODE_320			(0)
-#define	U8_UNICODE_500			(1)
-#define	U8_UNICODE_LATEST		(U8_UNICODE_500)
-
-#define	U8_VALIDATE_ENTIRE		(0x00100000)
-#define	U8_VALIDATE_CHECK_ADDITIONAL	(0x00200000)
-#define	U8_VALIDATE_UCS2_RANGE		(0x00400000)
-
-#define	U8_ILLEGAL_CHAR			(-1)
-#define	U8_OUT_OF_RANGE_CHAR		(-2)
-
-extern int u8_validate(char *, size_t, char **, int, int *);
-extern int u8_strcmp(const char *, const char *, size_t, int, size_t, int *);
-extern size_t u8_textprep_str(char *, size_t *, char *, size_t *, int, size_t,
-	int *);
-
-/*
  * ddi_map_regs
  *
  *	Map in the register set given by rnumber.
diff --git a/usr/src/uts/common/sys/u8_textprep.h b/usr/src/uts/common/sys/u8_textprep.h
new file mode 100644
index 0000000000..e30f064b2d
--- /dev/null
+++ b/usr/src/uts/common/sys/u8_textprep.h
@@ -0,0 +1,113 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+ * or http://www.opensolaris.org/os/licensing.
+ * See the License for the specific language governing permissions
+ * and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+/*
+ * Copyright 2007 Sun Microsystems, Inc.  All rights reserved.
+ * Use is subject to license terms.
+ */
+
+#ifndef	_SYS_U8_TEXTPREP_H
+#define	_SYS_U8_TEXTPREP_H
+
+#pragma ident	"%Z%%M%	%I%	%E% SMI"
+
+#include <sys/isa_defs.h>
+#include <sys/types.h>
+#include <sys/errno.h>
+
+#ifdef	__cplusplus
+extern "C" {
+#endif
+
+/*
+ * Unicode encoding conversion functions and their macros.
+ */
+#define	UCONV_IN_BIG_ENDIAN		0x0001
+#define	UCONV_OUT_BIG_ENDIAN		0x0002
+#define	UCONV_IN_SYSTEM_ENDIAN		0x0004
+#define	UCONV_OUT_SYSTEM_ENDIAN		0x0008
+#define	UCONV_IN_LITTLE_ENDIAN		0x0010
+#define	UCONV_OUT_LITTLE_ENDIAN		0x0020
+#define	UCONV_IGNORE_NULL		0x0040
+#define	UCONV_IN_ACCEPT_BOM		0x0080
+#define	UCONV_OUT_EMIT_BOM		0x0100
+
+extern int uconv_u16tou32(const uint16_t *, size_t *, uint32_t *, size_t *,
+	int);
+extern int uconv_u16tou8(const uint16_t *, size_t *, uchar_t *, size_t *, int);
+extern int uconv_u32tou16(const uint32_t *, size_t *, uint16_t *, size_t *,
+	int);
+extern int uconv_u32tou8(const uint32_t *, size_t *, uchar_t *, size_t *, int);
+extern int uconv_u8tou16(const uchar_t *, size_t *, uint16_t *, size_t *, int);
+extern int uconv_u8tou32(const uchar_t *, size_t *, uint32_t *, size_t *, int);
+
+/*
+ * UTF-8 text preparation functions and their macros.
+ *
+ * Among the macros defined, U8_CANON_DECOMP, U8_COMPAT_DECOMP, and
+ * U8_CANON_COMP are not public interfaces and must not be used directly
+ * at the flag input argument.
+ */
+#define	U8_STRCMP_CS			(0x00000001)
+#define	U8_STRCMP_CI_UPPER		(0x00000002)
+#define	U8_STRCMP_CI_LOWER		(0x00000004)
+
+#define	U8_CANON_DECOMP			(0x00000010)
+#define	U8_COMPAT_DECOMP		(0x00000020)
+#define	U8_CANON_COMP			(0x00000040)
+
+#define	U8_STRCMP_NFD			(U8_CANON_DECOMP)
+#define	U8_STRCMP_NFC			(U8_CANON_DECOMP | U8_CANON_COMP)
+#define	U8_STRCMP_NFKD			(U8_COMPAT_DECOMP)
+#define	U8_STRCMP_NFKC			(U8_COMPAT_DECOMP | U8_CANON_COMP)
+
+#define	U8_TEXTPREP_TOUPPER		(U8_STRCMP_CI_UPPER)
+#define	U8_TEXTPREP_TOLOWER		(U8_STRCMP_CI_LOWER)
+
+#define	U8_TEXTPREP_NFD			(U8_STRCMP_NFD)
+#define	U8_TEXTPREP_NFC			(U8_STRCMP_NFC)
+#define	U8_TEXTPREP_NFKD		(U8_STRCMP_NFKD)
+#define	U8_TEXTPREP_NFKC		(U8_STRCMP_NFKC)
+
+#define	U8_TEXTPREP_IGNORE_NULL		(0x00010000)
+#define	U8_TEXTPREP_IGNORE_INVALID	(0x00020000)
+#define	U8_TEXTPREP_NOWAIT		(0x00040000)
+
+#define	U8_UNICODE_320			(0)
+#define	U8_UNICODE_500			(1)
+#define	U8_UNICODE_LATEST		(U8_UNICODE_500)
+
+#define	U8_VALIDATE_ENTIRE		(0x00100000)
+#define	U8_VALIDATE_CHECK_ADDITIONAL	(0x00200000)
+#define	U8_VALIDATE_UCS2_RANGE		(0x00400000)
+
+#define	U8_ILLEGAL_CHAR			(-1)
+#define	U8_OUT_OF_RANGE_CHAR		(-2)
+
+extern int u8_validate(char *, size_t, char **, int, int *);
+extern int u8_strcmp(const char *, const char *, size_t, int, size_t, int *);
+extern size_t u8_textprep_str(char *, size_t *, char *, size_t *, int, size_t,
+	int *);
+
+#ifdef	__cplusplus
+}
+#endif
+
+#endif	/* _SYS_U8_TEXTPREP_H */