diff options
author | Steve French <sfrench@us.ibm.com> | 2009-04-01 09:22:00 +0400 |
---|---|---|
committer | Steve French <sfrench@us.ibm.com> | 2009-04-17 05:26:48 +0400 |
commit | 85a6dac54a7e28112488b02523202985edc7e639 (patch) | |
tree | 141a35196516dc428fe2891b7c7da02d0133ecad /fs/cifs/readdir.c | |
parent | 74496d365ad171d11f21da4a8be71c945f6ec825 (diff) | |
download | linux-85a6dac54a7e28112488b02523202985edc7e639.tar.xz |
[CIFS] Endian convert UniqueId when reporting inode numbers from server files
Jeff made a good point that we should endian convert the UniqueId when we use
it to set i_ino Even though this value is opaque to the client, when comparing
the inode numbers of the same server file from two different clients (one
big endian, one little endian) or when we compare a big endian client's view
of i_ino with what the server thinks - we should get the same value
Signed-off-by: Steve French <sfrench@us.ibm.com>
Diffstat (limited to 'fs/cifs/readdir.c')
-rw-r--r-- | fs/cifs/readdir.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index c2c01ff4c32c..c3c3e6286af5 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -840,7 +840,7 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, len = strnlen(filename, PATH_MAX); } - *pinum = pFindData->UniqueId; + *pinum = le64_to_cpu(pFindData->UniqueId); } else if (level == SMB_FIND_FILE_DIRECTORY_INFO) { FILE_DIRECTORY_INFO *pFindData = (FILE_DIRECTORY_INFO *)current_entry; @@ -856,7 +856,7 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, (SEARCH_ID_FULL_DIR_INFO *)current_entry; filename = &pFindData->FileName[0]; len = le32_to_cpu(pFindData->FileNameLength); - *pinum = pFindData->UniqueId; + *pinum = le64_to_cpu(pFindData->UniqueId); } else if (level == SMB_FIND_FILE_BOTH_DIRECTORY_INFO) { FILE_BOTH_DIRECTORY_INFO *pFindData = (FILE_BOTH_DIRECTORY_INFO *)current_entry; |