[linux-block.git] / fs / smb / client / cifs_unicode.c

// SPDX-License-Identifier: GPL-2.0-or-later
/*
 *
 *   Copyright (c) International Business Machines  Corp., 2000,2009
 *   Modified by Steve French (sfrench@us.ibm.com)
 */
#include <linux/fs.h>
#include <linux/slab.h>
#include "cifs_fs_sb.h"
#include "cifs_unicode.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)
{
	if (src_char >= 0xF001 && src_char <= 0xF01F) {
		*target = src_char - 0xF000;
		return true;
	}
	switch (src_char) {
	case SFM_COLON:
		*target = ':';
		break;
	case SFM_DOUBLEQUOTE:
		*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_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 {
		dst = kstrndup(src, maxlen, GFP_KERNEL);
	}

	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;

	if (src_char >= 0x01 && src_char <= 0x1F) {
		dest_char = cpu_to_le16(src_char + 0xF000);
		return dest_char;
	}
	switch (src_char) {
	case ':':
		dest_char = cpu_to_le16(SFM_COLON);
		break;
	case '"':
		dest_char = cpu_to_le16(SFM_DOUBLEQUOTE);
		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;

			/**
			 * Remap spaces and periods found at the end of every
			 * component of the path. The special cases of '.' and
			 * '..' do not need to be dealt with explicitly because
			 * they are addressed in namei.c:link_path_walk().
			 **/
			if ((i == srclen - 1) || (source[i+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;
}

/*
 * 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;
}
Commit	Line	Data
1a59d1b8	1	// SPDX-License-Identifier: GPL-2.0-or-later
1da177e4	2	/*
1da177e4	3	*
d185cda7	4	* Copyright (c) International Business Machines Corp., 2000,2009
1da177e4	5	* Modified by Steve French (sfrench@us.ibm.com)
1da177e4 LT	6	*/
1da177e4 LT	7	#include <linux/fs.h>
5a0e3ad6	8	#include <linux/slab.h>
2baa2682	9	#include "cifs_fs_sb.h"
1da177e4	10	#include "cifs_unicode.h"
1da177e4	11	#include "cifspdu.h"
3979877e	12	#include "cifsglob.h"
1da177e4 LT	13	#include "cifs_debug.h"
1da177e4 LT	14
2baa2682 SF	15	int cifs_remap(struct cifs_sb_info *cifs_sb)
	16	{
	17	int map_type;
	18
	19	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR)
	20	map_type = SFM_MAP_UNI_RSVD;
	21	else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR)
	22	map_type = SFU_MAP_UNI_RSVD;
	23	else
	24	map_type = NO_MAP_UNI_RSVD;
	25
	26	return map_type;
	27	}
	28
b693855f SF	29	/* Convert character using the SFU - "Services for Unix" remapping range */
	30	static bool
	31	convert_sfu_char(const __u16 src_char, char *target)
7fabf0c9	32	{
7fabf0c9 JL	33	/*
	34	* BB: Cannot handle remapping UNI_SLASH until all the calls to
	35	* build_path_from_dentry are modified, as they use slash as
	36	* separator.
	37	*/
ba2dbf30	38	switch (src_char) {
7fabf0c9 JL	39	case UNI_COLON:
	40	*target = ':';
	41	break;
581ade4d	42	case UNI_ASTERISK:
7fabf0c9 JL	43	target = '';
	44	break;
	45	case UNI_QUESTION:
	46	*target = '?';
	47	break;
	48	case UNI_PIPE:
	49	*target = '\|';
	50	break;
	51	case UNI_GRTRTHAN:
	52	*target = '>';
	53	break;
	54	case UNI_LESSTHAN:
	55	*target = '<';
	56	break;
	57	default:
b693855f	58	return false;
7fabf0c9	59	}
b693855f SF	60	return true;
b693855f SF	61	}
7fabf0c9	62
b693855f SF	63	/* Convert character using the SFM - "Services for Mac" remapping range */
	64	static bool
	65	convert_sfm_char(const __u16 src_char, char *target)
	66	{
7e46f090 BJ	67	if (src_char >= 0xF001 && src_char <= 0xF01F) {
	68	*target = src_char - 0xF000;
	69	return true;
	70	}
b693855f SF	71	switch (src_char) {
	72	case SFM_COLON:
	73	*target = ':';
	74	break;
85435d7a BJ	75	case SFM_DOUBLEQUOTE:
	76	*target = '"';
	77	break;
b693855f SF	78	case SFM_ASTERISK:
	79	target = '';
	80	break;
	81	case SFM_QUESTION:
	82	*target = '?';
	83	break;
	84	case SFM_PIPE:
	85	*target = '\|';
	86	break;
	87	case SFM_GRTRTHAN:
	88	*target = '>';
	89	break;
	90	case SFM_LESSTHAN:
	91	*target = '<';
	92	break;
45e8a258 SF	93	case SFM_SPACE:
	94	*target = ' ';
	95	break;
	96	case SFM_PERIOD:
	97	*target = '.';
	98	break;
b693855f SF	99	default:
	100	return false;
	101	}
	102	return true;
	103	}
	104
	105
	106	/*
	107	* cifs_mapchar - convert a host-endian char to proper char in codepage
	108	* @target - where converted character should be copied
	109	* @src_char - 2 byte host-endian source character
	110	* @cp - codepage to which character should be converted
	111	* @map_type - How should the 7 NTFS/SMB reserved characters be mapped to UCS2?
	112	*
	113	* This function handles the conversion of a single character. It is the
	114	* responsibility of the caller to ensure that the target buffer is large
	115	* enough to hold the result of the conversion (at least NLS_MAX_CHARSET_SIZE).
	116	*/
	117	static int
b2910307	118	cifs_mapchar(char target, const __u16 from, const struct nls_table *cp,
b693855f SF	119	int maptype)
	120	{
	121	int len = 1;
b2910307 NA	122	__u16 src_char;
	123
	124	src_char = *from;
b693855f SF	125
	126	if ((maptype == SFM_MAP_UNI_RSVD) && convert_sfm_char(src_char, target))
	127	return len;
	128	else if ((maptype == SFU_MAP_UNI_RSVD) &&
	129	convert_sfu_char(src_char, target))
	130	return len;
7fabf0c9	131
b693855f	132	/* if character not one of seven in special remap set */
ba2dbf30	133	len = cp->uni2char(src_char, target, NLS_MAX_CHARSET_SIZE);
b2910307 NA	134	if (len <= 0)
	135	goto surrogate_pair;
	136
	137	return len;
	138
	139	surrogate_pair:
	140	/* convert SURROGATE_PAIR and IVS */
	141	if (strcmp(cp->charset, "utf8"))
	142	goto unknown;
	143	len = utf16s_to_utf8s(from, 3, UTF16_LITTLE_ENDIAN, target, 6);
	144	if (len <= 0)
	145	goto unknown;
	146	return len;
	147
	148	unknown:
	149	*target = '?';
	150	len = 1;
b693855f	151	return len;
7fabf0c9 JL	152	}
	153
	154	/*
acbbb76a	155	* cifs_from_utf16 - convert utf16le string to local charset
7fabf0c9 JL	156	* @to - destination buffer
	157	* @from - source buffer
	158	* @tolen - destination buffer size (in bytes)
	159	* @fromlen - source buffer size (in bytes)
	160	* @codepage - codepage to which characters should be converted
	161	* @mapchar - should characters be remapped according to the mapchars option?
	162	*
acbbb76a	163	* Convert a little-endian utf16le string (as sent by the server) to a string
7fabf0c9 JL	164	* in the provided codepage. The tolen and fromlen parameters are to ensure
	165	* that the code doesn't walk off of the end of the buffer (which is always
	166	* a danger if the alignment of the source buffer is off). The destination
	167	* string is always properly null terminated and fits in the destination
	168	* buffer. Returns the length of the destination string in bytes (including
	169	* null terminator).
	170	*
	171	* Note that some windows versions actually send multiword UTF-16 characters
acbbb76a	172	* instead of straight UTF16-2. The linux nls routines however aren't able to
7fabf0c9 JL	173	* deal with those characters properly. In the event that we get some of
	174	* those characters, they won't be translated properly.
	175	*/
	176	int
acbbb76a	177	cifs_from_utf16(char to, const __le16 from, int tolen, int fromlen,
b693855f	178	const struct nls_table *codepage, int map_type)
7fabf0c9 JL	179	{
	180	int i, charlen, safelen;
	181	int outlen = 0;
	182	int nullsize = nls_nullsize(codepage);
	183	int fromwords = fromlen / 2;
	184	char tmp[NLS_MAX_CHARSET_SIZE];
b2910307	185	__u16 ftmp[3]; /* ftmp[3] = 3array x 2bytes = 6bytes UTF-16 */
7fabf0c9 JL	186
	187	/*
	188	* because the chars can be of varying widths, we need to take care
	189	* not to overflow the destination buffer when we get close to the
	190	* end of it. Until we get to this offset, we don't need to check
	191	* for overflow however.
	192	*/
	193	safelen = tolen - (NLS_MAX_CHARSET_SIZE + nullsize);
	194
ba2dbf30	195	for (i = 0; i < fromwords; i++) {
b2910307 NA	196	ftmp[0] = get_unaligned_le16(&from[i]);
b2910307 NA	197	if (ftmp[0] == 0)
ba2dbf30	198	break;
b2910307 NA	199	if (i + 1 < fromwords)
	200	ftmp[1] = get_unaligned_le16(&from[i + 1]);
	201	else
	202	ftmp[1] = 0;
	203	if (i + 2 < fromwords)
	204	ftmp[2] = get_unaligned_le16(&from[i + 2]);
	205	else
	206	ftmp[2] = 0;
ba2dbf30	207
7fabf0c9 JL	208	/*
	209	* check to see if converting this character might make the
	210	* conversion bleed into the null terminator
	211	*/
	212	if (outlen >= safelen) {
b693855f	213	charlen = cifs_mapchar(tmp, ftmp, codepage, map_type);
7fabf0c9 JL	214	if ((outlen + charlen) > (tolen - nullsize))
	215	break;
	216	}
	217
	218	/* put converted char into 'to' buffer */
b693855f	219	charlen = cifs_mapchar(&to[outlen], ftmp, codepage, map_type);
7fabf0c9	220	outlen += charlen;
b2910307 NA	221
	222	/* charlen (=bytes of UTF-8 for 1 character)
	223	* 4bytes UTF-8(surrogate pair) is charlen=4
	224	* (4bytes UTF-16 code)
	225	* 7-8bytes UTF-8(IVS) is charlen=3+4 or 4+4
	226	* (2 UTF-8 pairs divided to 2 UTF-16 pairs) */
	227	if (charlen == 4)
	228	i++;
	229	else if (charlen >= 5)
	230	/* 5-6bytes UTF-8 */
	231	i += 2;
7fabf0c9 JL	232	}
	233
	234	/* properly null-terminate string */
	235	for (i = 0; i < nullsize; i++)
	236	to[outlen++] = 0;
	237
	238	return outlen;
	239	}
	240
1da177e4	241	/*
acbbb76a	242	* NAME: cifs_strtoUTF16()
1da177e4 LT	243	*
	244	* FUNCTION: Convert character string to unicode string
	245	*
	246	*/
	247	int
acbbb76a	248	cifs_strtoUTF16(__le16 to, const char from, int len,
1da177e4 LT	249	const struct nls_table *codepage)
	250	{
	251	int charlen;
	252	int i;
ba2dbf30	253	wchar_t wchar_to; /* needed to quiet sparse */
1da177e4	254
fd3ba42c FZ	255	/* special case for utf8 to handle no plane0 chars */
	256	if (!strcmp(codepage->charset, "utf8")) {
	257	/*
	258	* convert utf8 -> utf16, we assume we have enough space
	259	* as caller should have assumed conversion does not overflow
	260	* in destination len is length in wchar_t units (16bits)
	261	*/
	262	i = utf8s_to_utf16s(from, len, UTF16_LITTLE_ENDIAN,
	263	(wchar_t *) to, len);
	264
	265	/* if success terminate and exit */
	266	if (i >= 0)
	267	goto success;
	268	/*
	269	* if fails fall back to UCS encoding as this
	270	* function should not return negative values
	271	* currently can fail only if source contains
	272	* invalid encoded characters
	273	*/
	274	}
	275
1da177e4	276	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
ba2dbf30	277	charlen = codepage->char2uni(from, len, &wchar_to);
1da177e4	278	if (charlen < 1) {
f96637be JP	279	cifs_dbg(VFS, "strtoUTF16: char2uni of 0x%x returned %d\n",
f96637be JP	280	*from, charlen);
69114089	281	/* A question mark */
ba2dbf30	282	wchar_to = 0x003f;
1da177e4	283	charlen = 1;
ba2dbf30 JL	284	}
ba2dbf30 JL	285	put_unaligned_le16(wchar_to, &to[i]);
1da177e4 LT	286	}
1da177e4 LT	287
fd3ba42c	288	success:
ba2dbf30	289	put_unaligned_le16(0, &to[i]);
1da177e4 LT	290	return i;
	291	}
	292
b2910307 NA	293	/*
	294	* cifs_utf16_bytes - how long will a string be after conversion?
	295	* @utf16 - pointer to input string
	296	* @maxbytes - don't go past this many bytes of input string
	297	* @codepage - destination codepage
	298	*
	299	* Walk a utf16le string and return the number of bytes that the string will
	300	* be after being converted to the given charset, not including any null
	301	* termination required. Don't walk past maxbytes in the source buffer.
	302	*/
	303	int
	304	cifs_utf16_bytes(const __le16 *from, int maxbytes,
	305	const struct nls_table *codepage)
	306	{
	307	int i;
	308	int charlen, outlen = 0;
	309	int maxwords = maxbytes / 2;
	310	char tmp[NLS_MAX_CHARSET_SIZE];
	311	__u16 ftmp[3];
	312
	313	for (i = 0; i < maxwords; i++) {
	314	ftmp[0] = get_unaligned_le16(&from[i]);
	315	if (ftmp[0] == 0)
	316	break;
	317	if (i + 1 < maxwords)
	318	ftmp[1] = get_unaligned_le16(&from[i + 1]);
	319	else
	320	ftmp[1] = 0;
	321	if (i + 2 < maxwords)
	322	ftmp[2] = get_unaligned_le16(&from[i + 2]);
	323	else
	324	ftmp[2] = 0;
	325
	326	charlen = cifs_mapchar(tmp, ftmp, codepage, NO_MAP_UNI_RSVD);
	327	outlen += charlen;
	328	}
	329
	330	return outlen;
	331	}
	332
066ce689	333	/*
acbbb76a SF	334	* cifs_strndup_from_utf16 - copy a string from wire format to the local
acbbb76a SF	335	* codepage
066ce689 JL	336	* @src - source string
	337	* @maxlen - don't walk past this many bytes in the source string
	338	* @is_unicode - is this a unicode string?
	339	* @codepage - destination codepage
	340	*
	341	* Take a string given by the server, convert it to the local codepage and
	342	* put it in a new buffer. Returns a pointer to the new string or NULL on
	343	* error.
	344	*/
	345	char *
acbbb76a SF	346	cifs_strndup_from_utf16(const char *src, const int maxlen,
acbbb76a SF	347	const bool is_unicode, const struct nls_table *codepage)
066ce689 JL	348	{
	349	int len;
	350	char *dst;
	351
	352	if (is_unicode) {
acbbb76a	353	len = cifs_utf16_bytes((__le16 *) src, maxlen, codepage);
066ce689 JL	354	len += nls_nullsize(codepage);
	355	dst = kmalloc(len, GFP_KERNEL);
	356	if (!dst)
	357	return NULL;
acbbb76a	358	cifs_from_utf16(dst, (__le16 *) src, len, maxlen, codepage,
f980d055	359	NO_MAP_UNI_RSVD);
066ce689	360	} else {
f980d055	361	dst = kstrndup(src, maxlen, GFP_KERNEL);
066ce689 JL	362	}
	363
	364	return dst;
	365	}
	366
a4153cb1 SF	367	static __le16 convert_to_sfu_char(char src_char)
	368	{
	369	__le16 dest_char;
	370
	371	switch (src_char) {
	372	case ':':
	373	dest_char = cpu_to_le16(UNI_COLON);
	374	break;
	375	case '*':
	376	dest_char = cpu_to_le16(UNI_ASTERISK);
	377	break;
	378	case '?':
	379	dest_char = cpu_to_le16(UNI_QUESTION);
	380	break;
	381	case '<':
	382	dest_char = cpu_to_le16(UNI_LESSTHAN);
	383	break;
	384	case '>':
	385	dest_char = cpu_to_le16(UNI_GRTRTHAN);
	386	break;
	387	case '\|':
	388	dest_char = cpu_to_le16(UNI_PIPE);
	389	break;
	390	default:
	391	dest_char = 0;
	392	}
	393
	394	return dest_char;
	395	}
	396
45e8a258	397	static __le16 convert_to_sfm_char(char src_char, bool end_of_string)
a4153cb1 SF	398	{
	399	__le16 dest_char;
	400
7e46f090 BJ	401	if (src_char >= 0x01 && src_char <= 0x1F) {
	402	dest_char = cpu_to_le16(src_char + 0xF000);
	403	return dest_char;
	404	}
a4153cb1 SF	405	switch (src_char) {
	406	case ':':
	407	dest_char = cpu_to_le16(SFM_COLON);
	408	break;
85435d7a BJ	409	case '"':
	410	dest_char = cpu_to_le16(SFM_DOUBLEQUOTE);
	411	break;
a4153cb1 SF	412	case '*':
	413	dest_char = cpu_to_le16(SFM_ASTERISK);
	414	break;
	415	case '?':
	416	dest_char = cpu_to_le16(SFM_QUESTION);
	417	break;
	418	case '<':
	419	dest_char = cpu_to_le16(SFM_LESSTHAN);
	420	break;
	421	case '>':
	422	dest_char = cpu_to_le16(SFM_GRTRTHAN);
	423	break;
	424	case '\|':
	425	dest_char = cpu_to_le16(SFM_PIPE);
	426	break;
45e8a258 SF	427	case '.':
	428	if (end_of_string)
	429	dest_char = cpu_to_le16(SFM_PERIOD);
	430	else
	431	dest_char = 0;
	432	break;
	433	case ' ':
	434	if (end_of_string)
	435	dest_char = cpu_to_le16(SFM_SPACE);
	436	else
	437	dest_char = 0;
	438	break;
a4153cb1 SF	439	default:
	440	dest_char = 0;
	441	}
	442
	443	return dest_char;
	444	}
	445
84cdf74e JL	446	/*
	447	* Convert 16 bit Unicode pathname to wire format from string in current code
	448	* page. Conversion may involve remapping up the six characters that are
	449	* only legal in POSIX-like OS (if they are present in the string). Path
	450	* names are little endian 16 bit Unicode on the wire
	451	*/
	452	int
acbbb76a	453	cifsConvertToUTF16(__le16 target, const char source, int srclen,
a4153cb1	454	const struct nls_table *cp, int map_chars)
84cdf74e	455	{
ce36d9ab SF	456	int i, charlen;
ce36d9ab SF	457	int j = 0;
84cdf74e	458	char src_char;
581ade4d JL	459	__le16 dst_char;
581ade4d JL	460	wchar_t tmp;
b2910307 NA	461	wchar_t wchar_to; / UTF-16 */
	462	int ret;
	463	unicode_t u;
84cdf74e	464
a4153cb1	465	if (map_chars == NO_MAP_UNI_RSVD)
acbbb76a	466	return cifs_strtoUTF16(target, source, PATH_MAX, cp);
84cdf74e	467
b2910307 NA	468	wchar_to = kzalloc(6, GFP_KERNEL);
b2910307 NA	469
ce36d9ab	470	for (i = 0; i < srclen; j++) {
84cdf74e	471	src_char = source[i];
11379b5e	472	charlen = 1;
a4153cb1 SF	473
	474	/* check if end of string */
	475	if (src_char == 0)
acbbb76a	476	goto ctoUTF16_out;
a4153cb1 SF	477
	478	/* see if we must remap this char */
	479	if (map_chars == SFU_MAP_UNI_RSVD)
	480	dst_char = convert_to_sfu_char(src_char);
45e8a258 SF	481	else if (map_chars == SFM_MAP_UNI_RSVD) {
	482	bool end_of_string;
	483
57c17607 BP	484	/**
	485	* Remap spaces and periods found at the end of every
	486	* component of the path. The special cases of '.' and
	487	* '..' do not need to be dealt with explicitly because
	488	* they are addressed in namei.c:link_path_walk().
	489	**/
	490	if ((i == srclen - 1) \|\| (source[i+1] == '\\'))
45e8a258 SF	491	end_of_string = true;
	492	else
	493	end_of_string = false;
	494
	495	dst_char = convert_to_sfm_char(src_char, end_of_string);
	496	} else
a4153cb1	497	dst_char = 0;
84cdf74e JL	498	/*
	499	* FIXME: We can not handle remapping backslash (UNI_SLASH)
	500	* until all the calls to build_path_from_dentry are modified,
	501	* as they use backslash as separator.
	502	*/
a4153cb1	503	if (dst_char == 0) {
581ade4d JL	504	charlen = cp->char2uni(source + i, srclen - i, &tmp);
	505	dst_char = cpu_to_le16(tmp);
	506
84cdf74e JL	507	/*
	508	* if no match, use question mark, which at least in
	509	* some cases serves as wild card
	510	*/
b2910307 NA	511	if (charlen > 0)
	512	goto ctoUTF16;
	513
	514	/* convert SURROGATE_PAIR */
	515	if (strcmp(cp->charset, "utf8") \|\| !wchar_to)
	516	goto unknown;
	517	if (*(source + i) & 0x80) {
	518	charlen = utf8_to_utf32(source + i, 6, &u);
	519	if (charlen < 0)
	520	goto unknown;
	521	} else
	522	goto unknown;
	523	ret = utf8s_to_utf16s(source + i, charlen,
	524	UTF16_LITTLE_ENDIAN,
	525	wchar_to, 6);
	526	if (ret < 0)
	527	goto unknown;
	528
	529	i += charlen;
	530	dst_char = cpu_to_le16(*wchar_to);
	531	if (charlen <= 3)
	532	/* 1-3bytes UTF-8 to 2bytes UTF-16 */
	533	put_unaligned(dst_char, &target[j]);
	534	else if (charlen == 4) {
	535	/* 4bytes UTF-8(surrogate pair) to 4bytes UTF-16
	536	* 7-8bytes UTF-8(IVS) divided to 2 UTF-16
	537	* (charlen=3+4 or 4+4) */
	538	put_unaligned(dst_char, &target[j]);
	539	dst_char = cpu_to_le16(*(wchar_to + 1));
	540	j++;
	541	put_unaligned(dst_char, &target[j]);
	542	} else if (charlen >= 5) {
	543	/* 5-6bytes UTF-8 to 6bytes UTF-16 */
	544	put_unaligned(dst_char, &target[j]);
	545	dst_char = cpu_to_le16(*(wchar_to + 1));
	546	j++;
	547	put_unaligned(dst_char, &target[j]);
	548	dst_char = cpu_to_le16(*(wchar_to + 2));
	549	j++;
	550	put_unaligned(dst_char, &target[j]);
84cdf74e	551	}
b2910307 NA	552	continue;
	553
	554	unknown:
	555	dst_char = cpu_to_le16(0x003f);
	556	charlen = 1;
84cdf74e	557	}
b2910307 NA	558
b2910307 NA	559	ctoUTF16:
11379b5e JL	560	/*
	561	* character may take more than one byte in the source string,
	562	* but will take exactly two bytes in the target string
	563	*/
	564	i += charlen;
581ade4d	565	put_unaligned(dst_char, &target[j]);
84cdf74e JL	566	}
84cdf74e JL	567
acbbb76a	568	ctoUTF16_out:
ce36d9ab	569	put_unaligned(0, &target[j]); /* Null terminate target unicode string */
b2910307	570	kfree(wchar_to);
c73f6939	571	return j;
84cdf74e	572	}
2503a0db	573
2503a0db PS	574	/*
	575	* cifs_local_to_utf16_bytes - how long will a string be after conversion?
	576	* @from - pointer to input string
	577	* @maxbytes - don't go past this many bytes of input string
	578	* @codepage - source codepage
	579	*
	580	* Walk a string and return the number of bytes that the string will
	581	* be after being converted to the given charset, not including any null
	582	* termination required. Don't walk past maxbytes in the source buffer.
	583	*/
	584
	585	static int
	586	cifs_local_to_utf16_bytes(const char *from, int len,
	587	const struct nls_table *codepage)
	588	{
	589	int charlen;
	590	int i;
	591	wchar_t wchar_to;
	592
	593	for (i = 0; len && *from; i++, from += charlen, len -= charlen) {
	594	charlen = codepage->char2uni(from, len, &wchar_to);
	595	/* Failed conversion defaults to a question mark */
	596	if (charlen < 1)
	597	charlen = 1;
	598	}
	599	return 2 * i; /* UTF16 characters are two bytes */
	600	}
	601
	602	/*
	603	* cifs_strndup_to_utf16 - copy a string to wire format from the local codepage
	604	* @src - source string
	605	* @maxlen - don't walk past this many bytes in the source string
	606	* @utf16_len - the length of the allocated string in bytes (including null)
	607	* @cp - source codepage
	608	* @remap - map special chars
	609	*
	610	* Take a string convert it from the local codepage to UTF16 and
	611	* put it in a new buffer. Returns a pointer to the new string or NULL on
	612	* error.
	613	*/
	614	__le16 *
	615	cifs_strndup_to_utf16(const char src, const int maxlen, int utf16_len,
	616	const struct nls_table *cp, int remap)
	617	{
	618	int len;
	619	__le16 *dst;
	620
	621	len = cifs_local_to_utf16_bytes(src, maxlen, cp);
	622	len += 2; /* NULL */
	623	dst = kmalloc(len, GFP_KERNEL);
	624	if (!dst) {
	625	*utf16_len = 0;
	626	return NULL;
	627	}
	628	cifsConvertToUTF16(dst, src, strlen(src), cp, remap);
	629	*utf16_len = len;
	630	return dst;
	631	}