On 02/12/2014 10:43 AM, Sage Weil wrote:
> On Wed, 12 Feb 2014, Yan, Zheng wrote:
>> On 02/12/2014 01:25 AM, Alex Elder wrote:
>>> On 02/11/2014 09:10 AM, Yan, Zheng wrote:
>>>> On Tue, Feb 11, 2014 at 10:47 PM, Alex Elder <elder@xxxxxxxx> wrote:
>>>>> On 02/10/2014 11:30 PM, Yan, Zheng wrote:
>>>>>> Signed-off-by: Yan, Zheng <zheng.z.yan@xxxxxxxxx>
>>>>>
>>>>> You really need to explain better under what circumstances
>>>>> a zero-size xattr is getting removed.
>>>>>
>>>>> But apparently it's only when you're updating an xattr
>>>>> (not building it up from a blob from storage).
>>>>>
>>>>> Why are you doing this? Why can't an xattr exist with
>>>>> an empty value?
>>>>
>>>> That is how other FS behave, at least for ext* and btrfs.
>>>
>>> I haven't tested this, I'm just glancing through code.
>>> But it looks to me like a zero-length value is OK, but
>>> a null value pointer means it should be deleted. Note
>>> in btrfs_setxattr(), for example, the same bit of code
>>> I referenced before:
>>>
>>> if (size == 0)
>>> value = ""; /* empty EA, do not remove */
>>>
>>> And ext4 seems to delete for a null value, but handle
>>> an xattr whose value *length* is zero. Same with XFS.
>>>
>>> So again, I haven't verified through testing, but my
>>> reading of the code (though rusty) still seems to show
>>> that an attribute can have an empty (zero-size) value.
>>
>> You are right, Thanks. how about below patch?
>>
>> ---
>> >From f11d5da84230e4993333a063019bad68c67b50d4 Mon Sep 17 00:00:00 2001
>> From: "Yan, Zheng" <zheng.z.yan@xxxxxxxxx>
>> Date: Tue, 11 Feb 2014 13:04:19 +0800
>> Subject: [PATCH 2/5] ceph: remove xattr when null value is given to setxattr()
>>
>> For the setxattr request, introduce a new flag CEPH_XATTR_REMOVE
>> to distinguish null value case from the zero-length value case.
>>
>> Signed-off-by: Yan, Zheng <zheng.z.yan@xxxxxxxxx>
>> ---
>> fs/ceph/xattr.c | 16 ++++++++++++++--
>> include/linux/ceph/ceph_fs.h | 5 +++--
>> 2 files changed, 17 insertions(+), 4 deletions(-)
>>
>> diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
>> index 28f9793..f6becf6 100644
>> --- a/fs/ceph/xattr.c
>> +++ b/fs/ceph/xattr.c
>> @@ -12,6 +12,9 @@
>> #define XATTR_CEPH_PREFIX "ceph."
>> #define XATTR_CEPH_PREFIX_LEN (sizeof (XATTR_CEPH_PREFIX) - 1)
>>
>> +static int __remove_xattr(struct ceph_inode_info *ci,
>> + struct ceph_inode_xattr *xattr);
>> +
>> /*
>> * List of handlers for synthetic system.* attributes. Other
>> * attributes are handled directly.
>> @@ -359,6 +362,12 @@ static int __set_xattr(struct ceph_inode_info *ci,
>> kfree(val);
>> return err;
>> }
>> + if (update_xattr < 0) {
>> + if (xattr)
>> + __remove_xattr(ci, xattr);
>> + kfree(name);
>> + return 0;
>> + }
>> }
>>
>> if (!xattr) {
>> @@ -862,6 +871,9 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
>>
>> dout("setxattr value=%.*s\n", (int)size, value);
>>
>> + if (!value)
>> + flags |= CEPH_XATTR_REMOVE;
>> +
>> /* do request */
>> req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETXATTR,
>
> How about just turning this into a REMOVEXATTR request in the !value case?
>
REMOVEXATTR request doesn't respect the CEPH_XATTR_REPLACE flag.
Regards
Yan, Zheng
> sage
>
>> USE_AUTH_MDS);
>> @@ -965,8 +977,8 @@ retry:
>> goto retry;
>> }
>>
>> - err = __set_xattr(ci, newname, name_len, newval,
>> - val_len, flags, 1, &xattr);
>> + err = __set_xattr(ci, newname, name_len, newval, val_len,
>> + flags, val ? 1 : -1, &xattr);
>>
>> if (!err) {
>> dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
>> diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h
>> index 2623cff..25bfb0e 100644
>> --- a/include/linux/ceph/ceph_fs.h
>> +++ b/include/linux/ceph/ceph_fs.h
>> @@ -373,8 +373,9 @@ extern const char *ceph_mds_op_name(int op);
>> /*
>> * Ceph setxattr request flags.
>> */
>> -#define CEPH_XATTR_CREATE 1
>> -#define CEPH_XATTR_REPLACE 2
>> +#define CEPH_XATTR_CREATE (1 << 0)
>> +#define CEPH_XATTR_REPLACE (1 << 1)
>> +#define CEPH_XATTR_REMOVE (1 << 31)
>>
>> union ceph_mds_request_args {
>> struct {
>> --
>> 1.8.5.3
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>>
>>> -Alex
>>>
>>>> Regards
>>>> Yan, Zheng
>>>>
>>>>>
>>>>> Looking at generic_setxattr() in "fs/xattr.c" we see:
>>>>> if (size == 0)
>>>>> value = ""; /* empty EA, do not remove */
>>>>>
>>>>> The code you have below looks OK, but it seems that you
>>>>> shouldn't be doing this.
>>>>>
>>>>> Am I missing something?
>>>>>
>>>>> -Alex
>>>>>
>>>>>> ---
>>>>>> fs/ceph/xattr.c | 9 +++++++++
>>>>>> 1 file changed, 9 insertions(+)
>>>>>>
>>>>>> diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
>>>>>> index 28f9793..6ed0e5a 100644
>>>>>> --- a/fs/ceph/xattr.c
>>>>>> +++ b/fs/ceph/xattr.c
>>>>>> @@ -12,6 +12,9 @@
>>>>>> #define XATTR_CEPH_PREFIX "ceph."
>>>>>> #define XATTR_CEPH_PREFIX_LEN (sizeof (XATTR_CEPH_PREFIX) - 1)
>>>>>>
>>>>>> +static int __remove_xattr(struct ceph_inode_info *ci,
>>>>>> + struct ceph_inode_xattr *xattr);
>>>>>> +
>>>>>> /*
>>>>>> * List of handlers for synthetic system.* attributes. Other
>>>>>> * attributes are handled directly.
>>>>>> @@ -359,6 +362,12 @@ static int __set_xattr(struct ceph_inode_info *ci,
>>>>>> kfree(val);
>>>>>> return err;
>>>>>> }
>>>>>> + if (!val_len) {
>>>>>> + if (xattr)
>>>>>> + __remove_xattr(ci, xattr);
>>>>>> + kfree(name);
>>>>>> + return 0;
>>>>>> + }
>>>>>> }
>>>>>>
>>>>>> if (!xattr) {
>>>>>>
>>>>>
>>>>> --
>>>>> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
>>>>> the body of a message to majordomo@xxxxxxxxxxxxxxx
>>>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
>> the body of a message to majordomo@xxxxxxxxxxxxxxx
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>
>>
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
