[linux-2.6-block.git] / fs / cifs / cifs_unicode.c

/*
 *   fs/cifs/cifs_unicode.c
 *
 *   Copyright (c) International Business Machines  Corp., 2000,2009
 *   Modified by Steve French (sfrench@us.ibm.com)
 *
 *   This program is free software;  you can redistribute it and/or modify
 *   it under the terms of the GNU General Public License as published by
 *   the Free Software Foundation; either version 2 of the License, or
 *   (at your option) any later version.
 *
 *   This program is distributed in the hope that it will be useful,
 *   but WITHOUT ANY WARRANTY;  without even the implied warranty of
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
 *   the GNU General Public License for more details.
 *
 *   You should have received a copy of the GNU General Public License
 *   along with this program;  if not, write to the Free Software
 *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
 */
#include <linux/fs.h>
#include <linux/slab.h>
#include "cifs_fs_sb.h"
#include "cifs_unicode.h"
#include "cifs_uniupr.h"
#include "cifspdu.h"
#include "cifsglob.h"
#include "cifs_debug.h"

int cifs_remap(struct cifs_sb_info *cifs_sb)
{
	int map_type;

	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR)
		map_type = SFM_MAP_UNI_RSVD;
	else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR)
		map_type = SFU_MAP_UNI_RSVD;
	else
		map_type = NO_MAP_UNI_RSVD;

	return map_type;
}

/* Convert character using the SFU - "Services for Unix" remapping range */
static bool
convert_sfu_char(const __u16 src_char, char *target)
{
	/*
	 * BB: Cannot handle remapping UNI_SLASH until all the calls to
	 *     build_path_from_dentry are modified, as they use slash as
	 *     separator.
	 */
	switch (src_char) {
	case UNI_COLON:
		*target = ':';
		break;
	case UNI_ASTERISK:
		*target = '*';
		break;
	case UNI_QUESTION:
		*target = '?';
		break;
	case UNI_PIPE:
		*target = '|';
		break;
	case UNI_GRTRTHAN:
		*target = '>';
		break;
	case UNI_LESSTHAN:
		*target = '<';
		break;
	default:
		return false;
	}
	return true;
}

/* Convert character using the SFM - "Services for Mac" remapping range */
static bool
convert_sfm_char(const __u16 src_char, char *target)
{
	switch (src_char) {
	case SFM_COLON:
		*target = ':';
		break;
	case SFM_ASTERISK:
		*target = '*';
		break;
	case SFM_QUESTION:
		*target = '?';
		break;
	case SFM_PIPE:
		*target = '|';
		break;
	case SFM_GRTRTHAN:
		*target = '>';
		break;
	case SFM_LESSTHAN:
		*target = '<';
		break;
	case SFM_SLASH:
		*target = '\\';
		break;
	case SFM_SPACE:
		*target = ' ';
		break;
	case SFM_PERIOD:
		*target = '.';
		break;
	default:
		return false;
	}
	return true;
}


/*
 * cifs_mapchar - convert a host-endian char to proper char in codepage
 * @target - where converted character should be copied
 * @src_char - 2 byte host-endian source character
 * @cp - codepage to which character should be converted
 * @map_type - How should the 7 NTFS/SMB reserved characters be mapped to UCS2?
 *
 * This function handles the conversion of a single character. It is the
 * responsibility of the caller to ensure that the target buffer is large
 * enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE).
 */
static int
cifs_mapchar(char *target, const __u16 *from, const struct nls_table *cp,
	     int maptype)
{
	int len = 1;
	__u16 src_char;

	src_char = *from;

	if ((maptype == SFM_MAP_UNI_RSVD) && convert_sfm_char(src_char, target))
		return len;
	else if ((maptype == SFU_MAP_UNI_RSVD) &&
		  convert_sfu_char(src_char, target))
		return len;

	/* if character not one of seven in special remap set */
	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
	if (len <= 0)
		goto surrogate_pair;

	return len;

surrogate_pair:
	/* convert SURROGATE_PAIR and IVS */
	if (strcmp(cp->charset, "utf8"))
		goto unknown;
	len = utf16s_to_utf8s(from, 3, UTF16_LITTLE_ENDIAN, target, 6);
	if (len <= 0)
		goto unknown;
	return len;

unknown:
	*target = '?';
	len = 1;
	return len;
}

/*
 * cifs_from_utf16 - convert utf16le string to local charset
 * @to - destination buffer
 * @from - source buffer
 * @tolen - destination buffer size (in bytes)
 * @fromlen - source buffer size (in bytes)
 * @codepage - codepage to which characters should be converted
 * @mapchar - should characters be remapped according to the mapchars option?
 *
 * Convert a little-endian utf16le string (as sent by the server) to a string
 * in the provided codepage. The tolen and fromlen parameters are to ensure
 * that the code doesn't walk off of the end of the buffer (which is always
 * a danger if the alignment of the source buffer is off). The destination
 * string is always properly null terminated and fits in the destination
 * buffer. Returns the length of the destination string in bytes (including
 * null terminator).
 *
 * Note that some windows versions actually send multiword UTF-16 characters
 * instead of straight UTF16-2. The linux nls routines however aren't able to
 * deal with those characters properly. In the event that we get some of
 * those characters, they won't be translated properly.
 */
int
cifs_from_utf16(char *to, const __le16 *from, int tolen, int fromlen,
		const struct nls_table *codepage, int map_type)
{
	int i, charlen, safelen;
	int outlen = 0;
	int nullsize = nls_nullsize(codepage);
	int fromwords = fromlen / 2;
	char tmp[NLS_MAX_CHARSET_SIZE];
	__u16 ftmp[3];		/* ftmp[3] = 3array x 2bytes = 6bytes UTF-16 */

	/*
	 * because the chars can be of varying widths, we need to take care
	 * not to overflow the destination buffer when we get close to the
	 * end of it. Until we get to this offset, we don't need to check
	 * for overflow however.
	 */
	safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize);

	for (i = 0; i < fromwords; i++) {
		ftmp[0] = get_unaligned_le16(&from[i]);
		if (ftmp[0] == 0)
			break;
		if (i + 1 < fromwords)
			ftmp[1] = get_unaligned_le16(&from[i + 1]);
		else
			ftmp[1] = 0;
		if (i + 2 < fromwords)
			ftmp[2] = get_unaligned_le16(&from[i + 2]);
		else
			ftmp[2] = 0;

		/*
		 * check to see if converting this character might make the
		 * conversion bleed into the null terminator
		 */
		if (outlen >= safelen) {
			charlen = cifs_mapchar(tmp, ftmp, codepage, map_type);
			if ((outlen + charlen) > (tolen - nullsize))
				break;
		}

		/* put converted char into 'to' buffer */
		charlen = cifs_mapchar(&to[outlen], ftmp, codepage, map_type);
		outlen += charlen;

		/* charlen (=bytes of UTF-8 for 1 character)
		 * 4bytes UTF-8(surrogate pair) is charlen=4
		 *   (4bytes UTF-16 code)
		 * 7-8bytes UTF-8(IVS) is charlen=3+4 or 4+4
		 *   (2 UTF-8 pairs divided to 2 UTF-16 pairs) */
		if (charlen == 4)
			i++;
		else if (charlen >= 5)
			/* 5-6bytes UTF-8 */
			i += 2;
	}

	/* properly null-terminate string */
	for (i = 0; i < nullsize; i++)
		to[outlen++] = 0;

	return outlen;
}

/*
 * NAME:	cifs_strtoUTF16()
 *
 * FUNCTION:	Convert character string to unicode string
 *
 */
int
cifs_strtoUTF16(__le16 *to, const char *from, int len,
	      const struct nls_table *codepage)
{
	int charlen;
	int i;
	wchar_t wchar_to; /* needed to quiet sparse */

	/* special case for utf8 to handle no plane0 chars */
	if (!strcmp(codepage->charset, "utf8")) {
		/*
		 * convert utf8 -> utf16, we assume we have enough space
		 * as caller should have assumed conversion does not overflow
		 * in destination len is length in wchar_t units (16bits)
		 */
		i  = utf8s_to_utf16s(from, len, UTF16_LITTLE_ENDIAN,
				       (wchar_t *) to, len);

		/* if success terminate and exit */
		if (i >= 0)
			goto success;
		/*
		 * if fails fall back to UCS encoding as this
		 * function should not return negative values
		 * currently can fail only if source contains
		 * invalid encoded characters
		 */
	}

	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
		charlen = codepage->char2uni(from, len, &wchar_to);
		if (charlen < 1) {
			cifs_dbg(VFS, "strtoUTF16: char2uni of 0x%x returned %d\n",
				 *from, charlen);
			/* A question mark */
			wchar_to = 0x003f;
			charlen = 1;
		}
		put_unaligned_le16(wchar_to, &to[i]);
	}

success:
	put_unaligned_le16(0, &to[i]);
	return i;
}

/*
 * cifs_utf16_bytes - how long will a string be after conversion?
 * @utf16 - pointer to input string
 * @maxbytes - don't go past this many bytes of input string
 * @codepage - destination codepage
 *
 * Walk a utf16le string and return the number of bytes that the string will
 * be after being converted to the given charset, not including any null
 * termination required. Don't walk past maxbytes in the source buffer.
 */
int
cifs_utf16_bytes(const __le16 *from, int maxbytes,
		const struct nls_table *codepage)
{
	int i;
	int charlen, outlen = 0;
	int maxwords = maxbytes / 2;
	char tmp[NLS_MAX_CHARSET_SIZE];
	__u16 ftmp[3];

	for (i = 0; i < maxwords; i++) {
		ftmp[0] = get_unaligned_le16(&from[i]);
		if (ftmp[0] == 0)
			break;
		if (i + 1 < maxwords)
			ftmp[1] = get_unaligned_le16(&from[i + 1]);
		else
			ftmp[1] = 0;
		if (i + 2 < maxwords)
			ftmp[2] = get_unaligned_le16(&from[i + 2]);
		else
			ftmp[2] = 0;

		charlen = cifs_mapchar(tmp, ftmp, codepage, NO_MAP_UNI_RSVD);
		outlen += charlen;
	}

	return outlen;
}

/*
 * cifs_strndup_from_utf16 - copy a string from wire format to the local
 * codepage
 * @src - source string
 * @maxlen - don't walk past this many bytes in the source string
 * @is_unicode - is this a unicode string?
 * @codepage - destination codepage
 *
 * Take a string given by the server, convert it to the local codepage and
 * put it in a new buffer. Returns a pointer to the new string or NULL on
 * error.
 */
char *
cifs_strndup_from_utf16(const char *src, const int maxlen,
			const bool is_unicode, const struct nls_table *codepage)
{
	int len;
	char *dst;

	if (is_unicode) {
		len = cifs_utf16_bytes((__le16 *) src, maxlen, codepage);
		len += nls_nullsize(codepage);
		dst = kmalloc(len, GFP_KERNEL);
		if (!dst)
			return NULL;
		cifs_from_utf16(dst, (__le16 *) src, len, maxlen, codepage,
			       NO_MAP_UNI_RSVD);
	} else {
		len = strnlen(src, maxlen);
		len++;
		dst = kmalloc(len, GFP_KERNEL);
		if (!dst)
			return NULL;
		strlcpy(dst, src, len);
	}

	return dst;
}

static __le16 convert_to_sfu_char(char src_char)
{
	__le16 dest_char;

	switch (src_char) {
	case ':':
		dest_char = cpu_to_le16(UNI_COLON);
		break;
	case '*':
		dest_char = cpu_to_le16(UNI_ASTERISK);
		break;
	case '?':
		dest_char = cpu_to_le16(UNI_QUESTION);
		break;
	case '<':
		dest_char = cpu_to_le16(UNI_LESSTHAN);
		break;
	case '>':
		dest_char = cpu_to_le16(UNI_GRTRTHAN);
		break;
	case '|':
		dest_char = cpu_to_le16(UNI_PIPE);
		break;
	default:
		dest_char = 0;
	}

	return dest_char;
}

static __le16 convert_to_sfm_char(char src_char, bool end_of_string)
{
	__le16 dest_char;

	switch (src_char) {
	case ':':
		dest_char = cpu_to_le16(SFM_COLON);
		break;
	case '*':
		dest_char = cpu_to_le16(SFM_ASTERISK);
		break;
	case '?':
		dest_char = cpu_to_le16(SFM_QUESTION);
		break;
	case '<':
		dest_char = cpu_to_le16(SFM_LESSTHAN);
		break;
	case '>':
		dest_char = cpu_to_le16(SFM_GRTRTHAN);
		break;
	case '|':
		dest_char = cpu_to_le16(SFM_PIPE);
		break;
	case '.':
		if (end_of_string)
			dest_char = cpu_to_le16(SFM_PERIOD);
		else
			dest_char = 0;
		break;
	case ' ':
		if (end_of_string)
			dest_char = cpu_to_le16(SFM_SPACE);
		else
			dest_char = 0;
		break;
	default:
		dest_char = 0;
	}

	return dest_char;
}

/*
 * Convert 16 bit Unicode pathname to wire format from string in current code
 * page. Conversion may involve remapping up the six characters that are
 * only legal in POSIX-like OS (if they are present in the string). Path
 * names are little endian 16 bit Unicode on the wire
 */
int
cifsConvertToUTF16(__le16 *target, const char *source, int srclen,
		 const struct nls_table *cp, int map_chars)
{
	int i, charlen;
	int j = 0;
	char src_char;
	__le16 dst_char;
	wchar_t tmp;
	wchar_t *wchar_to;	/* UTF-16 */
	int ret;
	unicode_t u;

	if (map_chars == NO_MAP_UNI_RSVD)
		return cifs_strtoUTF16(target, source, PATH_MAX, cp);

	wchar_to = kzalloc(6, GFP_KERNEL);

	for (i = 0; i < srclen; j++) {
		src_char = source[i];
		charlen = 1;

		/* check if end of string */
		if (src_char == 0)
			goto ctoUTF16_out;

		/* see if we must remap this char */
		if (map_chars == SFU_MAP_UNI_RSVD)
			dst_char = convert_to_sfu_char(src_char);
		else if (map_chars == SFM_MAP_UNI_RSVD) {
			bool end_of_string;

			if (i == srclen - 1)
				end_of_string = true;
			else
				end_of_string = false;

			dst_char = convert_to_sfm_char(src_char, end_of_string);
		} else
			dst_char = 0;
		/*
		 * FIXME: We can not handle remapping backslash (UNI_SLASH)
		 * until all the calls to build_path_from_dentry are modified,
		 * as they use backslash as separator.
		 */
		if (dst_char == 0) {
			charlen = cp->char2uni(source + i, srclen - i, &tmp);
			dst_char = cpu_to_le16(tmp);

			/*
			 * if no match, use question mark, which at least in
			 * some cases serves as wild card
			 */
			if (charlen > 0)
				goto ctoUTF16;

			/* convert SURROGATE_PAIR */
			if (strcmp(cp->charset, "utf8") || !wchar_to)
				goto unknown;
			if (*(source + i) & 0x80) {
				charlen = utf8_to_utf32(source + i, 6, &u);
				if (charlen < 0)
					goto unknown;
			} else
				goto unknown;
			ret  = utf8s_to_utf16s(source + i, charlen,
					       UTF16_LITTLE_ENDIAN,
					       wchar_to, 6);
			if (ret < 0)
				goto unknown;

			i += charlen;
			dst_char = cpu_to_le16(*wchar_to);
			if (charlen <= 3)
				/* 1-3bytes UTF-8 to 2bytes UTF-16 */
				put_unaligned(dst_char, &target[j]);
			else if (charlen == 4) {
				/* 4bytes UTF-8(surrogate pair) to 4bytes UTF-16
				 * 7-8bytes UTF-8(IVS) divided to 2 UTF-16
				 *   (charlen=3+4 or 4+4) */
				put_unaligned(dst_char, &target[j]);
				dst_char = cpu_to_le16(*(wchar_to + 1));
				j++;
				put_unaligned(dst_char, &target[j]);
			} else if (charlen >= 5) {
				/* 5-6bytes UTF-8 to 6bytes UTF-16 */
				put_unaligned(dst_char, &target[j]);
				dst_char = cpu_to_le16(*(wchar_to + 1));
				j++;
				put_unaligned(dst_char, &target[j]);
				dst_char = cpu_to_le16(*(wchar_to + 2));
				j++;
				put_unaligned(dst_char, &target[j]);
			}
			continue;

unknown:
			dst_char = cpu_to_le16(0x003f);
			charlen = 1;
		}

ctoUTF16:
		/*
		 * character may take more than one byte in the source string,
		 * but will take exactly two bytes in the target string
		 */
		i += charlen;
		put_unaligned(dst_char, &target[j]);
	}

ctoUTF16_out:
	put_unaligned(0, &target[j]); /* Null terminate target unicode string */
	kfree(wchar_to);
	return j;
}

#ifdef CONFIG_CIFS_SMB2
/*
 * cifs_local_to_utf16_bytes - how long will a string be after conversion?
 * @from - pointer to input string
 * @maxbytes - don't go past this many bytes of input string
 * @codepage - source codepage
 *
 * Walk a string and return the number of bytes that the string will
 * be after being converted to the given charset, not including any null
 * termination required. Don't walk past maxbytes in the source buffer.
 */

static int
cifs_local_to_utf16_bytes(const char *from, int len,
			  const struct nls_table *codepage)
{
	int charlen;
	int i;
	wchar_t wchar_to;

	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
		charlen = codepage->char2uni(from, len, &wchar_to);
		/* Failed conversion defaults to a question mark */
		if (charlen < 1)
			charlen = 1;
	}
	return 2 * i; /* UTF16 characters are two bytes */
}

/*
 * cifs_strndup_to_utf16 - copy a string to wire format from the local codepage
 * @src - source string
 * @maxlen - don't walk past this many bytes in the source string
 * @utf16_len - the length of the allocated string in bytes (including null)
 * @cp - source codepage
 * @remap - map special chars
 *
 * Take a string convert it from the local codepage to UTF16 and
 * put it in a new buffer. Returns a pointer to the new string or NULL on
 * error.
 */
__le16 *
cifs_strndup_to_utf16(const char *src, const int maxlen, int *utf16_len,
		      const struct nls_table *cp, int remap)
{
	int len;
	__le16 *dst;

	len = cifs_local_to_utf16_bytes(src, maxlen, cp);
	len += 2; /* NULL */
	dst = kmalloc(len, GFP_KERNEL);
	if (!dst) {
		*utf16_len = 0;
		return NULL;
	}
	cifsConvertToUTF16(dst, src, strlen(src), cp, remap);
	*utf16_len = len;
	return dst;
}
#endif /* CONFIG_CIFS_SMB2 */
Commit	Line	Data
1da177e4 LT	1	/*
	2	* fs/cifs/cifs_unicode.c
	3	*
d185cda7	4	* Copyright (c) International Business Machines Corp., 2000,2009
1da177e4 LT	5	* Modified by Steve French (sfrench@us.ibm.com)
	6	*
	7	* This program is free software; you can redistribute it and/or modify
	8	* it under the terms of the GNU General Public License as published by
221601c3	9	* the Free Software Foundation; either version 2 of the License, or
1da177e4	10	* (at your option) any later version.
221601c3	11	*
1da177e4 LT	12	* This program is distributed in the hope that it will be useful,
	13	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	14	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
	15	* the GNU General Public License for more details.
	16	*
	17	* You should have received a copy of the GNU General Public License
221601c3	18	* along with this program; if not, write to the Free Software
1da177e4 LT	19	* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
	20	*/
	21	#include <linux/fs.h>
5a0e3ad6	22	#include <linux/slab.h>
2baa2682	23	#include "cifs_fs_sb.h"
1da177e4 LT	24	#include "cifs_unicode.h"
	25	#include "cifs_uniupr.h"
	26	#include "cifspdu.h"
3979877e	27	#include "cifsglob.h"
1da177e4 LT	28	#include "cifs_debug.h"
1da177e4 LT	29
2baa2682 SF	30	int cifs_remap(struct cifs_sb_info *cifs_sb)
	31	{
	32	int map_type;
	33
	34	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR)
	35	map_type = SFM_MAP_UNI_RSVD;
	36	else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR)
	37	map_type = SFU_MAP_UNI_RSVD;
	38	else
	39	map_type = NO_MAP_UNI_RSVD;
	40
	41	return map_type;
	42	}
	43
b693855f SF	44	/* Convert character using the SFU - "Services for Unix" remapping range */
	45	static bool
	46	convert_sfu_char(const __u16 src_char, char *target)
7fabf0c9	47	{
7fabf0c9 JL	48	/*
	49	* BB: Cannot handle remapping UNI_SLASH until all the calls to
	50	* build_path_from_dentry are modified, as they use slash as
	51	* separator.
	52	*/
ba2dbf30	53	switch (src_char) {
7fabf0c9 JL	54	case UNI_COLON:
	55	*target = ':';
	56	break;
581ade4d	57	case UNI_ASTERISK:
7fabf0c9 JL	58	target = '';
	59	break;
	60	case UNI_QUESTION:
	61	*target = '?';
	62	break;
	63	case UNI_PIPE:
	64	*target = '\|';
	65	break;
	66	case UNI_GRTRTHAN:
	67	*target = '>';
	68	break;
	69	case UNI_LESSTHAN:
	70	*target = '<';
	71	break;
	72	default:
b693855f	73	return false;
7fabf0c9	74	}
b693855f SF	75	return true;
b693855f SF	76	}
7fabf0c9	77
b693855f SF	78	/* Convert character using the SFM - "Services for Mac" remapping range */
	79	static bool
	80	convert_sfm_char(const __u16 src_char, char *target)
	81	{
	82	switch (src_char) {
	83	case SFM_COLON:
	84	*target = ':';
	85	break;
	86	case SFM_ASTERISK:
	87	target = '';
	88	break;
	89	case SFM_QUESTION:
	90	*target = '?';
	91	break;
	92	case SFM_PIPE:
	93	*target = '\|';
	94	break;
	95	case SFM_GRTRTHAN:
	96	*target = '>';
	97	break;
	98	case SFM_LESSTHAN:
	99	*target = '<';
	100	break;
	101	case SFM_SLASH:
	102	*target = '\\';
	103	break;
45e8a258 SF	104	case SFM_SPACE:
	105	*target = ' ';
	106	break;
	107	case SFM_PERIOD:
	108	*target = '.';
	109	break;
b693855f SF	110	default:
	111	return false;
	112	}
	113	return true;
	114	}
	115
	116
	117	/*
	118	* cifs_mapchar - convert a host-endian char to proper char in codepage
	119	* @target - where converted character should be copied
	120	* @src_char - 2 byte host-endian source character
	121	* @cp - codepage to which character should be converted
	122	* @map_type - How should the 7 NTFS/SMB reserved characters be mapped to UCS2?
	123	*
	124	* This function handles the conversion of a single character. It is the
	125	* responsibility of the caller to ensure that the target buffer is large
	126	* enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE).
	127	*/
	128	static int
b2910307	129	cifs_mapchar(char target, const __u16 from, const struct nls_table *cp,
b693855f SF	130	int maptype)
	131	{
	132	int len = 1;
b2910307 NA	133	__u16 src_char;
	134
	135	src_char = *from;
b693855f SF	136
	137	if ((maptype == SFM_MAP_UNI_RSVD) && convert_sfm_char(src_char, target))
	138	return len;
	139	else if ((maptype == SFU_MAP_UNI_RSVD) &&
	140	convert_sfu_char(src_char, target))
	141	return len;
7fabf0c9	142
b693855f	143	/* if character not one of seven in special remap set */
ba2dbf30	144	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
b2910307 NA	145	if (len <= 0)
	146	goto surrogate_pair;
	147
	148	return len;
	149
	150	surrogate_pair:
	151	/* convert SURROGATE_PAIR and IVS */
	152	if (strcmp(cp->charset, "utf8"))
	153	goto unknown;
	154	len = utf16s_to_utf8s(from, 3, UTF16_LITTLE_ENDIAN, target, 6);
	155	if (len <= 0)
	156	goto unknown;
	157	return len;
	158
	159	unknown:
	160	*target = '?';
	161	len = 1;
b693855f	162	return len;
7fabf0c9 JL	163	}
	164
	165	/*
acbbb76a	166	* cifs_from_utf16 - convert utf16le string to local charset
7fabf0c9 JL	167	* @to - destination buffer
	168	* @from - source buffer
	169	* @tolen - destination buffer size (in bytes)
	170	* @fromlen - source buffer size (in bytes)
	171	* @codepage - codepage to which characters should be converted
	172	* @mapchar - should characters be remapped according to the mapchars option?
	173	*
acbbb76a	174	* Convert a little-endian utf16le string (as sent by the server) to a string
7fabf0c9 JL	175	* in the provided codepage. The tolen and fromlen parameters are to ensure
	176	* that the code doesn't walk off of the end of the buffer (which is always
	177	* a danger if the alignment of the source buffer is off). The destination
	178	* string is always properly null terminated and fits in the destination
	179	* buffer. Returns the length of the destination string in bytes (including
	180	* null terminator).
	181	*
	182	* Note that some windows versions actually send multiword UTF-16 characters
acbbb76a	183	* instead of straight UTF16-2. The linux nls routines however aren't able to
7fabf0c9 JL	184	* deal with those characters properly. In the event that we get some of
	185	* those characters, they won't be translated properly.
	186	*/
	187	int
acbbb76a	188	cifs_from_utf16(char to, const __le16 from, int tolen, int fromlen,
b693855f	189	const struct nls_table *codepage, int map_type)
7fabf0c9 JL	190	{
	191	int i, charlen, safelen;
	192	int outlen = 0;
	193	int nullsize = nls_nullsize(codepage);
	194	int fromwords = fromlen / 2;
	195	char tmp[NLS_MAX_CHARSET_SIZE];
b2910307	196	__u16 ftmp[3]; /* ftmp[3] = 3array x 2bytes = 6bytes UTF-16 */
7fabf0c9 JL	197
	198	/*
	199	* because the chars can be of varying widths, we need to take care
	200	* not to overflow the destination buffer when we get close to the
	201	* end of it. Until we get to this offset, we don't need to check
	202	* for overflow however.
	203	*/
	204	safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize);
	205
ba2dbf30	206	for (i = 0; i < fromwords; i++) {
b2910307 NA	207	ftmp[0] = get_unaligned_le16(&from[i]);
b2910307 NA	208	if (ftmp[0] == 0)
ba2dbf30	209	break;
b2910307 NA	210	if (i + 1 < fromwords)
	211	ftmp[1] = get_unaligned_le16(&from[i + 1]);
	212	else
	213	ftmp[1] = 0;
	214	if (i + 2 < fromwords)
	215	ftmp[2] = get_unaligned_le16(&from[i + 2]);
	216	else
	217	ftmp[2] = 0;
ba2dbf30	218
7fabf0c9 JL	219	/*
	220	* check to see if converting this character might make the
	221	* conversion bleed into the null terminator
	222	*/
	223	if (outlen >= safelen) {
b693855f	224	charlen = cifs_mapchar(tmp, ftmp, codepage, map_type);
7fabf0c9 JL	225	if ((outlen + charlen) > (tolen - nullsize))
	226	break;
	227	}
	228
	229	/* put converted char into 'to' buffer */
b693855f	230	charlen = cifs_mapchar(&to[outlen], ftmp, codepage, map_type);
7fabf0c9	231	outlen += charlen;
b2910307 NA	232
	233	/* charlen (=bytes of UTF-8 for 1 character)
	234	* 4bytes UTF-8(surrogate pair) is charlen=4
	235	* (4bytes UTF-16 code)
	236	* 7-8bytes UTF-8(IVS) is charlen=3+4 or 4+4
	237	* (2 UTF-8 pairs divided to 2 UTF-16 pairs) */
	238	if (charlen == 4)
	239	i++;
	240	else if (charlen >= 5)
	241	/* 5-6bytes UTF-8 */
	242	i += 2;
7fabf0c9 JL	243	}
	244
	245	/* properly null-terminate string */
	246	for (i = 0; i < nullsize; i++)
	247	to[outlen++] = 0;
	248
	249	return outlen;
	250	}
	251
1da177e4	252	/*
acbbb76a	253	* NAME: cifs_strtoUTF16()
1da177e4 LT	254	*
	255	* FUNCTION: Convert character string to unicode string
	256	*
	257	*/
	258	int
acbbb76a	259	cifs_strtoUTF16(__le16 to, const char from, int len,
1da177e4 LT	260	const struct nls_table *codepage)
	261	{
	262	int charlen;
	263	int i;
ba2dbf30	264	wchar_t wchar_to; /* needed to quiet sparse */
1da177e4	265
fd3ba42c FZ	266	/* special case for utf8 to handle no plane0 chars */
	267	if (!strcmp(codepage->charset, "utf8")) {
	268	/*
	269	* convert utf8 -> utf16, we assume we have enough space
	270	* as caller should have assumed conversion does not overflow
	271	* in destination len is length in wchar_t units (16bits)
	272	*/
	273	i = utf8s_to_utf16s(from, len, UTF16_LITTLE_ENDIAN,
	274	(wchar_t *) to, len);
	275
	276	/* if success terminate and exit */
	277	if (i >= 0)
	278	goto success;
	279	/*
	280	* if fails fall back to UCS encoding as this
	281	* function should not return negative values
	282	* currently can fail only if source contains
	283	* invalid encoded characters
	284	*/
	285	}
	286
1da177e4	287	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
ba2dbf30	288	charlen = codepage->char2uni(from, len, &wchar_to);
1da177e4	289	if (charlen < 1) {
f96637be JP	290	cifs_dbg(VFS, "strtoUTF16: char2uni of 0x%x returned %d\n",
f96637be JP	291	*from, charlen);
69114089	292	/* A question mark */
ba2dbf30	293	wchar_to = 0x003f;
1da177e4	294	charlen = 1;
ba2dbf30 JL	295	}
ba2dbf30 JL	296	put_unaligned_le16(wchar_to, &to[i]);
1da177e4 LT	297	}
1da177e4 LT	298
fd3ba42c	299	success:
ba2dbf30	300	put_unaligned_le16(0, &to[i]);
1da177e4 LT	301	return i;
	302	}
	303
b2910307 NA	304	/*
	305	* cifs_utf16_bytes - how long will a string be after conversion?
	306	* @utf16 - pointer to input string
	307	* @maxbytes - don't go past this many bytes of input string
	308	* @codepage - destination codepage
	309	*
	310	* Walk a utf16le string and return the number of bytes that the string will
	311	* be after being converted to the given charset, not including any null
	312	* termination required. Don't walk past maxbytes in the source buffer.
	313	*/
	314	int
	315	cifs_utf16_bytes(const __le16 *from, int maxbytes,
	316	const struct nls_table *codepage)
	317	{
	318	int i;
	319	int charlen, outlen = 0;
	320	int maxwords = maxbytes / 2;
	321	char tmp[NLS_MAX_CHARSET_SIZE];
	322	__u16 ftmp[3];
	323
	324	for (i = 0; i < maxwords; i++) {
	325	ftmp[0] = get_unaligned_le16(&from[i]);
	326	if (ftmp[0] == 0)
	327	break;
	328	if (i + 1 < maxwords)
	329	ftmp[1] = get_unaligned_le16(&from[i + 1]);
	330	else
	331	ftmp[1] = 0;
	332	if (i + 2 < maxwords)
	333	ftmp[2] = get_unaligned_le16(&from[i + 2]);
	334	else
	335	ftmp[2] = 0;
	336
	337	charlen = cifs_mapchar(tmp, ftmp, codepage, NO_MAP_UNI_RSVD);
	338	outlen += charlen;
	339	}
	340
	341	return outlen;
	342	}
	343
066ce689	344	/*
acbbb76a SF	345	* cifs_strndup_from_utf16 - copy a string from wire format to the local
acbbb76a SF	346	* codepage
066ce689 JL	347	* @src - source string
	348	* @maxlen - don't walk past this many bytes in the source string
	349	* @is_unicode - is this a unicode string?
	350	* @codepage - destination codepage
	351	*
	352	* Take a string given by the server, convert it to the local codepage and
	353	* put it in a new buffer. Returns a pointer to the new string or NULL on
	354	* error.
	355	*/
	356	char *
acbbb76a SF	357	cifs_strndup_from_utf16(const char *src, const int maxlen,
acbbb76a SF	358	const bool is_unicode, const struct nls_table *codepage)
066ce689 JL	359	{
	360	int len;
	361	char *dst;
	362
	363	if (is_unicode) {
acbbb76a	364	len = cifs_utf16_bytes((__le16 *) src, maxlen, codepage);
066ce689 JL	365	len += nls_nullsize(codepage);
	366	dst = kmalloc(len, GFP_KERNEL);
	367	if (!dst)
	368	return NULL;
acbbb76a	369	cifs_from_utf16(dst, (__le16 *) src, len, maxlen, codepage,
b693855f	370	NO_MAP_UNI_RSVD);
066ce689 JL	371	} else {
	372	len = strnlen(src, maxlen);
	373	len++;
	374	dst = kmalloc(len, GFP_KERNEL);
	375	if (!dst)
	376	return NULL;
	377	strlcpy(dst, src, len);
	378	}
	379
	380	return dst;
	381	}
	382
a4153cb1 SF	383	static __le16 convert_to_sfu_char(char src_char)
	384	{
	385	__le16 dest_char;
	386
	387	switch (src_char) {
	388	case ':':
	389	dest_char = cpu_to_le16(UNI_COLON);
	390	break;
	391	case '*':
	392	dest_char = cpu_to_le16(UNI_ASTERISK);
	393	break;
	394	case '?':
	395	dest_char = cpu_to_le16(UNI_QUESTION);
	396	break;
	397	case '<':
	398	dest_char = cpu_to_le16(UNI_LESSTHAN);
	399	break;
	400	case '>':
	401	dest_char = cpu_to_le16(UNI_GRTRTHAN);
	402	break;
	403	case '\|':
	404	dest_char = cpu_to_le16(UNI_PIPE);
	405	break;
	406	default:
	407	dest_char = 0;
	408	}
	409
	410	return dest_char;
	411	}
	412
45e8a258	413	static __le16 convert_to_sfm_char(char src_char, bool end_of_string)
a4153cb1 SF	414	{
	415	__le16 dest_char;
	416
	417	switch (src_char) {
	418	case ':':
	419	dest_char = cpu_to_le16(SFM_COLON);
	420	break;
	421	case '*':
	422	dest_char = cpu_to_le16(SFM_ASTERISK);
	423	break;
	424	case '?':
	425	dest_char = cpu_to_le16(SFM_QUESTION);
	426	break;
	427	case '<':
	428	dest_char = cpu_to_le16(SFM_LESSTHAN);
	429	break;
	430	case '>':
	431	dest_char = cpu_to_le16(SFM_GRTRTHAN);
	432	break;
	433	case '\|':
	434	dest_char = cpu_to_le16(SFM_PIPE);
	435	break;
45e8a258 SF	436	case '.':
	437	if (end_of_string)
	438	dest_char = cpu_to_le16(SFM_PERIOD);
	439	else
	440	dest_char = 0;
	441	break;
	442	case ' ':
	443	if (end_of_string)
	444	dest_char = cpu_to_le16(SFM_SPACE);
	445	else
	446	dest_char = 0;
	447	break;
a4153cb1 SF	448	default:
	449	dest_char = 0;
	450	}
	451
	452	return dest_char;
	453	}
	454
84cdf74e JL	455	/*
	456	* Convert 16 bit Unicode pathname to wire format from string in current code
	457	* page. Conversion may involve remapping up the six characters that are
	458	* only legal in POSIX-like OS (if they are present in the string). Path
	459	* names are little endian 16 bit Unicode on the wire
	460	*/
	461	int
acbbb76a	462	cifsConvertToUTF16(__le16 target, const char source, int srclen,
a4153cb1	463	const struct nls_table *cp, int map_chars)
84cdf74e	464	{
ce36d9ab SF	465	int i, charlen;
ce36d9ab SF	466	int j = 0;
84cdf74e	467	char src_char;
581ade4d JL	468	__le16 dst_char;
581ade4d JL	469	wchar_t tmp;
b2910307 NA	470	wchar_t wchar_to; / UTF-16 */
	471	int ret;
	472	unicode_t u;
84cdf74e	473
a4153cb1	474	if (map_chars == NO_MAP_UNI_RSVD)
acbbb76a	475	return cifs_strtoUTF16(target, source, PATH_MAX, cp);
84cdf74e	476
b2910307 NA	477	wchar_to = kzalloc(6, GFP_KERNEL);
b2910307 NA	478
ce36d9ab	479	for (i = 0; i < srclen; j++) {
84cdf74e	480	src_char = source[i];
11379b5e	481	charlen = 1;
a4153cb1 SF	482
	483	/* check if end of string */
	484	if (src_char == 0)
acbbb76a	485	goto ctoUTF16_out;
a4153cb1 SF	486
	487	/* see if we must remap this char */
	488	if (map_chars == SFU_MAP_UNI_RSVD)
	489	dst_char = convert_to_sfu_char(src_char);
45e8a258 SF	490	else if (map_chars == SFM_MAP_UNI_RSVD) {
	491	bool end_of_string;
	492
	493	if (i == srclen - 1)
	494	end_of_string = true;
	495	else
	496	end_of_string = false;
	497
	498	dst_char = convert_to_sfm_char(src_char, end_of_string);
	499	} else
a4153cb1	500	dst_char = 0;
84cdf74e JL	501	/*
	502	* FIXME: We can not handle remapping backslash (UNI_SLASH)
	503	* until all the calls to build_path_from_dentry are modified,
	504	* as they use backslash as separator.
	505	*/
a4153cb1	506	if (dst_char == 0) {
581ade4d JL	507	charlen = cp->char2uni(source + i, srclen - i, &tmp);
	508	dst_char = cpu_to_le16(tmp);
	509
84cdf74e JL	510	/*
	511	* if no match, use question mark, which at least in
	512	* some cases serves as wild card
	513	*/
b2910307 NA	514	if (charlen > 0)
	515	goto ctoUTF16;
	516
	517	/* convert SURROGATE_PAIR */
	518	if (strcmp(cp->charset, "utf8") \|\| !wchar_to)
	519	goto unknown;
	520	if (*(source + i) & 0x80) {
	521	charlen = utf8_to_utf32(source + i, 6, &u);
	522	if (charlen < 0)
	523	goto unknown;
	524	} else
	525	goto unknown;
	526	ret = utf8s_to_utf16s(source + i, charlen,
	527	UTF16_LITTLE_ENDIAN,
	528	wchar_to, 6);
	529	if (ret < 0)
	530	goto unknown;
	531
	532	i += charlen;
	533	dst_char = cpu_to_le16(*wchar_to);
	534	if (charlen <= 3)
	535	/* 1-3bytes UTF-8 to 2bytes UTF-16 */
	536	put_unaligned(dst_char, &target[j]);
	537	else if (charlen == 4) {
	538	/* 4bytes UTF-8(surrogate pair) to 4bytes UTF-16
	539	* 7-8bytes UTF-8(IVS) divided to 2 UTF-16
	540	* (charlen=3+4 or 4+4) */
	541	put_unaligned(dst_char, &target[j]);
	542	dst_char = cpu_to_le16(*(wchar_to + 1));
	543	j++;
	544	put_unaligned(dst_char, &target[j]);
	545	} else if (charlen >= 5) {
	546	/* 5-6bytes UTF-8 to 6bytes UTF-16 */
	547	put_unaligned(dst_char, &target[j]);
	548	dst_char = cpu_to_le16(*(wchar_to + 1));
	549	j++;
	550	put_unaligned(dst_char, &target[j]);
	551	dst_char = cpu_to_le16(*(wchar_to + 2));
	552	j++;
	553	put_unaligned(dst_char, &target[j]);
84cdf74e	554	}
b2910307 NA	555	continue;
	556
	557	unknown:
	558	dst_char = cpu_to_le16(0x003f);
	559	charlen = 1;
84cdf74e	560	}
b2910307 NA	561
b2910307 NA	562	ctoUTF16:
11379b5e JL	563	/*
	564	* character may take more than one byte in the source string,
	565	* but will take exactly two bytes in the target string
	566	*/
	567	i += charlen;
581ade4d	568	put_unaligned(dst_char, &target[j]);
84cdf74e JL	569	}
84cdf74e JL	570
acbbb76a	571	ctoUTF16_out:
ce36d9ab	572	put_unaligned(0, &target[j]); /* Null terminate target unicode string */
b2910307	573	kfree(wchar_to);
c73f6939	574	return j;
84cdf74e	575	}
2503a0db PS	576
	577	#ifdef CONFIG_CIFS_SMB2
	578	/*
	579	* cifs_local_to_utf16_bytes - how long will a string be after conversion?
	580	* @from - pointer to input string
	581	* @maxbytes - don't go past this many bytes of input string
	582	* @codepage - source codepage
	583	*
	584	* Walk a string and return the number of bytes that the string will
	585	* be after being converted to the given charset, not including any null
	586	* termination required. Don't walk past maxbytes in the source buffer.
	587	*/
	588
	589	static int
	590	cifs_local_to_utf16_bytes(const char *from, int len,
	591	const struct nls_table *codepage)
	592	{
	593	int charlen;
	594	int i;
	595	wchar_t wchar_to;
	596
	597	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
	598	charlen = codepage->char2uni(from, len, &wchar_to);
	599	/* Failed conversion defaults to a question mark */
	600	if (charlen < 1)
	601	charlen = 1;
	602	}
	603	return 2 * i; /* UTF16 characters are two bytes */
	604	}
	605
	606	/*
	607	* cifs_strndup_to_utf16 - copy a string to wire format from the local codepage
	608	* @src - source string
	609	* @maxlen - don't walk past this many bytes in the source string
	610	* @utf16_len - the length of the allocated string in bytes (including null)
	611	* @cp - source codepage
	612	* @remap - map special chars
	613	*
	614	* Take a string convert it from the local codepage to UTF16 and
	615	* put it in a new buffer. Returns a pointer to the new string or NULL on
	616	* error.
	617	*/
	618	__le16 *
	619	cifs_strndup_to_utf16(const char src, const int maxlen, int utf16_len,
	620	const struct nls_table *cp, int remap)
	621	{
	622	int len;
	623	__le16 *dst;
	624
	625	len = cifs_local_to_utf16_bytes(src, maxlen, cp);
	626	len += 2; /* NULL */
	627	dst = kmalloc(len, GFP_KERNEL);
	628	if (!dst) {
	629	*utf16_len = 0;
	630	return NULL;
	631	}
	632	cifsConvertToUTF16(dst, src, strlen(src), cp, remap);
	633	*utf16_len = len;
	634	return dst;
	635	}
	636	#endif /* CONFIG_CIFS_SMB2 */