On Sat, Oct 9, 2010 at 3:42 PM, Jeff Layton <jlayton@xxxxxxxxx> wrote:
> On Thu, 7 Oct 2010 09:03:27 -0500
> shirishpargaonkar@xxxxxxxxx wrote:
>
>> From: Shirish Pargaonkar <shirishpargaonkar@xxxxxxxxx>
>>
>>
>> Add extended attribute name system.cifsacl
>>
>> Get/generate cifs/ntfs acl blob and hand over to the invoker however
>> it wants to parse/process it.
>>
>> Stop generating cifs/ntfs acl blob for xattr name system.posix_acl_access
>>
>>
>> Signed-off-by: Shirish Pargaonkar <shirishpargaonkar@xxxxxxxxx>
>> ---
>> fs/cifs/xattr.c | 64 +++++++++++++++++++++++++++++++++++++------------------
>> 1 files changed, 43 insertions(+), 21 deletions(-)
>>
>> diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c
>> --- a/fs/cifs/xattr.c
>> +++ b/fs/cifs/xattr.c
<snip>
>> +static int
>> +get_cifs_acl(int xid, struct cifsTconInfo *pTcon, char *full_path,
>> + void *ea_value, size_t size, struct nls_table *nls_cp)
>> +{
>> + __u16 fid;
>> + int rc;
>> + int oplock = 0;
>> + unsigned int acllen = 0;
>> + struct cifs_ntsd *pacl = NULL;
>> +
>> + rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, GENERIC_READ,
>> + 0, &fid, &oplock, NULL, nls_cp, 0);
>> + if (rc) {
>> + cERROR(1, "%s: file open error: %d", __func__, rc);
>> + return rc;
>> + }
>> +
> ^^^^
> Hmm possible oplock break. Wouldn't it be better to see if you already
> have a readable filehandle before you do this?
Yes - absolutely need to check for matching file handle
>> + rc = CIFSSMBGetCIFSACL(xid, pTcon, fid, &pacl, &acllen);
>> + CIFSSMBClose(xid, pTcon, fid);
>> + if (rc) {
>> + cERROR(1, "%s: get acl error: %d", __func__, rc);
>> + return rc;
>> + }
>> +
>> + if (ea_value) {
>> + if (acllen > size) {
>> + kfree(pacl);
>> + return -ERANGE;
>> + }
>> + memcpy(ea_value, pacl, acllen);
>> + kfree(pacl);
>> + }
>> +
>> + return acllen;
>> +}
>> +
>> int cifs_setxattr(struct dentry *direntry, const char *ea_name,
>> const void *ea_value, size_t value_size, int flags)
>> {
>> @@ -277,27 +314,6 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name,
>> cifs_sb->local_nls,
>> cifs_sb->mnt_cifs_flags &
>> CIFS_MOUNT_MAP_SPECIAL_CHR);
>> -#ifdef CONFIG_CIFS_EXPERIMENTAL
>> - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) {
>> - __u16 fid;
>> - int oplock = 0;
>> - struct cifs_ntsd *pacl = NULL;
>> - __u32 buflen = 0;
>> - if (experimEnabled)
>> - rc = CIFSSMBOpen(xid, pTcon, full_path,
>> - FILE_OPEN, GENERIC_READ, 0, &fid,
>> - &oplock, NULL, cifs_sb->local_nls,
>> - cifs_sb->mnt_cifs_flags &
>> - CIFS_MOUNT_MAP_SPECIAL_CHR);
>> - /* else rc is EOPNOTSUPP from above */
>> -
>> - if (rc == 0) {
>> - rc = CIFSSMBGetCIFSACL(xid, pTcon, fid, &pacl,
>> - &buflen);
>> - CIFSSMBClose(xid, pTcon, fid);
>> - }
>> - }
>> -#endif /* EXPERIMENTAL */
>> #else
>> cFYI(1, "query POSIX ACL not supported yet");
>> #endif /* CONFIG_CIFS_POSIX */
>> @@ -313,6 +329,12 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name,
>> #else
>> cFYI(1, "query POSIX default ACL not supported yet");
>> #endif
>> +#ifdef CONFIG_CIFS_EXPERIMENTAL
>> + } else if (strncmp(ea_name, CIFS_ACL_XATTR,
>> + strlen(CIFS_ACL_XATTR)) == 0) {
>> + rc = get_cifs_acl(xid, pTcon, full_path, ea_value,
>> + buf_size, cifs_sb->local_nls);
>> +#endif /* CONFIG_CIFS_EXPERIMENTAL */
>> } else if (strncmp(ea_name,
>> CIFS_XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) == 0) {
>> cFYI(1, "Trusted xattr namespace not supported yet");
>
> I'd probably prefer not to see this new code wrapped in experimental
> tags. I don't think that adds any real value, especially not with
> something as simple as this.
I agree - seems low risk as well.
--
Thanks,
Steve
--
Thanks,
Steve
--
To unsubscribe from this list: send the line "unsubscribe linux-cifs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
