cifs: fix cifsConvertToUCS() for the mapchars case (11379b5e) · Commits · 戴 / test

As Metze pointed out, commit broke mapchars option: Commit "cifs: fix unaligned accesses in cifsConvertToUCS" () does multiple steps in just one commit (moving the function and changing it without testing). put_unaligned_le16(temp, &target[j]); is never called for any codepoint the goes via the 'default' switch statement. As a result we put just zero (or maybe uninitialized) bytes into the target buffer. His proposed patch looks correct, but doesn't apply to the current head of the tree. This patch should also fix it. Cc: <stable@kernel.org> # .38.x: : cifs: clean up various nits in unicode routines (try #2) Reported-by:

Stefan Metzmacher <metze@samba.org> Signed-off-by:

Jeff Layton <jlayton@redhat.com> Signed-off-by:

Steve French <sfrench@us.ibm.com>

fs/cifs/cifs_unicode.c

+6 −8

Original line number	Original line	Diff line number	Diff line
	@@ -277,6 +277,7 @@ cifsConvertToUCS(__le16 target, const char source, int srclen,

	for (i = 0, j = 0; i < srclen; j++) {		for (i = 0, j = 0; i < srclen; j++) {
	src_char = source[i];		src_char = source[i];
			charlen = 1;
	switch (src_char) {		switch (src_char) {
	case 0:		case 0:
	put_unaligned(0, &target[j]);		put_unaligned(0, &target[j]);
	@@ -316,16 +317,13 @@ cifsConvertToUCS(__le16 target, const char source, int srclen,
	dst_char = cpu_to_le16(0x003f);		dst_char = cpu_to_le16(0x003f);
	charlen = 1;		charlen = 1;
	}		}
			}
	/*		/*
	* character may take more than one byte in the source		* character may take more than one byte in the source string,
	* string, but will take exactly two bytes in the		* but will take exactly two bytes in the target string
	* target string
	*/		*/
	i += charlen;		i += charlen;
	continue;
	}
	put_unaligned(dst_char, &target[j]);		put_unaligned(dst_char, &target[j]);
	i++; /* move to next char in source string */
	}		}

	ctoUCS_out:		ctoUCS_out:

Admin message