[PATCH v2 0/7] xfs: random fixes and code cleanup

[PATCH v2 0/7] xfs: random fixes and code cleanup

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

Hi all,

This patchset include random fixes and code cleanups, and there are no
connections among these patches. In order to make it easier to track,
I bundle them up and put all the scattered patches into a single patchset.

Changes for v2: 
 -detect intent-done items by their item ops.
 -update the commit messages.
 -code cleanup for xfs_attr_leaf_entsize_{remote,local}.

Kaixu Xia (7):
  xfs: remove the unused SYNCHRONIZE macro
  xfs: use the existing type definition for di_projid
  xfs: remove the unnecessary xfs_dqid_t type cast
  xfs: do the assert for all the log done items in xfs_trans_cancel
  xfs: remove the redundant crc feature check in xfs_attr3_rmt_verify
  xfs: code cleanup in xfs_attr_leaf_entsize_{remote,local}
  xfs: fix some comments

 fs/xfs/libxfs/xfs_attr_remote.c |  2 --
 fs/xfs/libxfs/xfs_da_format.h   | 12 ++++++------
 fs/xfs/libxfs/xfs_inode_buf.h   |  2 +-
 fs/xfs/xfs_dquot.c              |  4 ++--
 fs/xfs/xfs_linux.h              |  1 -
 fs/xfs/xfs_qm.c                 |  2 +-
 fs/xfs/xfs_trans.c              |  9 ++++++++-
 7 files changed, 18 insertions(+), 14 deletions(-)

-- 
2.20.0

[PATCH v2 1/7] xfs: remove the unused SYNCHRONIZE macro

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

There are no callers of the SYNCHRONIZE() macro, so remove it.

Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 fs/xfs/xfs_linux.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h
index ab737fed7b12..ad1009778d33 100644
--- a/fs/xfs/xfs_linux.h
+++ b/fs/xfs/xfs_linux.h
@@ -123,7 +123,6 @@ typedef __u32			xfs_nlink_t;
 #define EFSCORRUPTED	EUCLEAN		/* Filesystem is corrupted */
 #define EFSBADCRC	EBADMSG		/* Bad CRC detected */
 
-#define SYNCHRONIZE()	barrier()
 #define __return_address __builtin_return_address(0)
 
 /*
-- 
2.20.0

[PATCH v2 2/7] xfs: use the existing type definition for di_projid

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

We have already defined the project ID type prid_t, so maybe should
use it here.

Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 fs/xfs/libxfs/xfs_inode_buf.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/xfs/libxfs/xfs_inode_buf.h b/fs/xfs/libxfs/xfs_inode_buf.h
index 536666143fe7..ef5eaf33d146 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.h
+++ b/fs/xfs/libxfs/xfs_inode_buf.h
@@ -17,7 +17,7 @@ struct xfs_dinode;
  */
 struct xfs_icdinode {
 	uint16_t	di_flushiter;	/* incremented on flush */
-	uint32_t	di_projid;	/* owner's project id */
+	prid_t		di_projid;	/* owner's project id */
 	xfs_fsize_t	di_size;	/* number of bytes in file */
 	xfs_rfsblock_t	di_nblocks;	/* # of direct & btree blocks used */
 	xfs_extlen_t	di_extsize;	/* basic/minimum extent size for file */
-- 
2.20.0

[PATCH v2 3/7] xfs: remove the unnecessary xfs_dqid_t type cast

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

Since the type prid_t and xfs_dqid_t both are uint32_t, seems the
type cast is unnecessary, so remove it.

Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 fs/xfs/xfs_qm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 3f82e0c92c2d..41a459ffd1f2 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -1715,7 +1715,7 @@ xfs_qm_vop_dqalloc(
 	if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) {
 		if (ip->i_d.di_projid != prid) {
 			xfs_iunlock(ip, lockflags);
-			error = xfs_qm_dqget(mp, (xfs_dqid_t)prid,
+			error = xfs_qm_dqget(mp, prid,
 					XFS_DQTYPE_PROJ, true, &pq);
 			if (error) {
 				ASSERT(error != -ENOENT);
-- 
2.20.0

[PATCH v2 4/7] xfs: do the assert for all the log done items in xfs_trans_cancel

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

We should do the assert for all the log intent-done items if they appear
here. This patch detect intent-done items by the fact that their item ops
don't have iop_unpin and iop_push methods.

Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
---
 fs/xfs/xfs_trans.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index ca18a040336a..0d5d5a53fa5a 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -925,6 +925,13 @@ xfs_trans_commit(
 	return __xfs_trans_commit(tp, false);
 }
 
+/* Is this a log intent-done item? */
+static inline bool xlog_item_is_intent_done(struct xfs_log_item *lip)
+{
+	return lip->li_ops->iop_unpin == NULL &&
+	       lip->li_ops->iop_push == NULL;
+}
+
 /*
  * Unlock all of the transaction's items and free the transaction.
  * The transaction must not have modified any of its items, because
@@ -959,7 +966,7 @@ xfs_trans_cancel(
 		struct xfs_log_item *lip;
 
 		list_for_each_entry(lip, &tp->t_items, li_trans)
-			ASSERT(!(lip->li_type == XFS_LI_EFD));
+			ASSERT(!xlog_item_is_intent_done(lip));
 	}
 #endif
 	xfs_trans_unreserve_and_mod_sb(tp);
-- 
2.20.0

Re: [PATCH v2 4/7] xfs: do the assert for all the log done items in xfs_trans_cancel

[Darrick J. Wong]

On Thu, Sep 17, 2020 at 07:38:45PM +0800, xiakaixu1987@gmail.com wrote:
> From: Kaixu Xia <kaixuxia@tencent.com>
> 
> We should do the assert for all the log intent-done items if they appear
> here. This patch detect intent-done items by the fact that their item ops
> don't have iop_unpin and iop_push methods.
> 
> Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>

Seems ok to me,
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>

--D

> ---
>  fs/xfs/xfs_trans.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
> index ca18a040336a..0d5d5a53fa5a 100644
> --- a/fs/xfs/xfs_trans.c
> +++ b/fs/xfs/xfs_trans.c
> @@ -925,6 +925,13 @@ xfs_trans_commit(
>  	return __xfs_trans_commit(tp, false);
>  }
>  
> +/* Is this a log intent-done item? */
> +static inline bool xlog_item_is_intent_done(struct xfs_log_item *lip)
> +{
> +	return lip->li_ops->iop_unpin == NULL &&
> +	       lip->li_ops->iop_push == NULL;
> +}
> +
>  /*
>   * Unlock all of the transaction's items and free the transaction.
>   * The transaction must not have modified any of its items, because
> @@ -959,7 +966,7 @@ xfs_trans_cancel(
>  		struct xfs_log_item *lip;
>  
>  		list_for_each_entry(lip, &tp->t_items, li_trans)
> -			ASSERT(!(lip->li_type == XFS_LI_EFD));
> +			ASSERT(!xlog_item_is_intent_done(lip));
>  	}
>  #endif
>  	xfs_trans_unreserve_and_mod_sb(tp);
> -- 
> 2.20.0
> 

Re: [PATCH v2 4/7] xfs: do the assert for all the log done items in xfs_trans_cancel

[Christoph Hellwig]

On Thu, Sep 17, 2020 at 07:38:45PM +0800, xiakaixu1987@gmail.com wrote:
> From: Kaixu Xia <kaixuxia@tencent.com>
> 
> We should do the assert for all the log intent-done items if they appear
> here. This patch detect intent-done items by the fact that their item ops
> don't have iop_unpin and iop_push methods.
> 
> Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
> ---
>  fs/xfs/xfs_trans.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
> index ca18a040336a..0d5d5a53fa5a 100644
> --- a/fs/xfs/xfs_trans.c
> +++ b/fs/xfs/xfs_trans.c
> @@ -925,6 +925,13 @@ xfs_trans_commit(
>  	return __xfs_trans_commit(tp, false);
>  }
>  
> +/* Is this a log intent-done item? */
> +static inline bool xlog_item_is_intent_done(struct xfs_log_item *lip)
> +{
> +	return lip->li_ops->iop_unpin == NULL &&
> +	       lip->li_ops->iop_push == NULL;
> +}

I think this helper should go into xfs_trans.h, next to the
xfs_log_item log item definition.  And xlog_item_is_intent should
be moved there as well.

Otherwise looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 4/7] xfs: do the assert for all the log done items in xfs_trans_cancel

[kaixuxia]

On 2020/9/19 14:27, Christoph Hellwig wrote:
> On Thu, Sep 17, 2020 at 07:38:45PM +0800, xiakaixu1987@gmail.com wrote:
>> From: Kaixu Xia <kaixuxia@tencent.com>
>>
>> We should do the assert for all the log intent-done items if they appear
>> here. This patch detect intent-done items by the fact that their item ops
>> don't have iop_unpin and iop_push methods.
>>
>> Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
>> ---
>>  fs/xfs/xfs_trans.c | 9 ++++++++-
>>  1 file changed, 8 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
>> index ca18a040336a..0d5d5a53fa5a 100644
>> --- a/fs/xfs/xfs_trans.c
>> +++ b/fs/xfs/xfs_trans.c
>> @@ -925,6 +925,13 @@ xfs_trans_commit(
>>  	return __xfs_trans_commit(tp, false);
>>  }
>>  
>> +/* Is this a log intent-done item? */
>> +static inline bool xlog_item_is_intent_done(struct xfs_log_item *lip)
>> +{
>> +	return lip->li_ops->iop_unpin == NULL &&
>> +	       lip->li_ops->iop_push == NULL;
>> +}
> 
> I think this helper should go into xfs_trans.h, next to the
> xfs_log_item log item definition.  And xlog_item_is_intent should
> be moved there as well.
> 
Yeah，make more sense, will move them to xfs_trans.h in the next version.

Thanks,
Kaixu

> Otherwise looks good:
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> 

-- 
kaixuxia

[PATCH v2 5/7] xfs: remove the redundant crc feature check in xfs_attr3_rmt_verify

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

We already check whether the crc feature is enabled before calling
xfs_attr3_rmt_verify(), so remove the redundant feature check in that
function.

Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
---
 fs/xfs/libxfs/xfs_attr_remote.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c
index 3f80cede7406..48d8e9caf86f 100644
--- a/fs/xfs/libxfs/xfs_attr_remote.c
+++ b/fs/xfs/libxfs/xfs_attr_remote.c
@@ -96,8 +96,6 @@ xfs_attr3_rmt_verify(
 {
 	struct xfs_attr3_rmt_hdr *rmt = ptr;
 
-	if (!xfs_sb_version_hascrc(&mp->m_sb))
-		return __this_address;
 	if (!xfs_verify_magic(bp, rmt->rm_magic))
 		return __this_address;
 	if (!uuid_equal(&rmt->rm_uuid, &mp->m_sb.sb_meta_uuid))
-- 
2.20.0

Re: [PATCH v2 5/7] xfs: remove the redundant crc feature check in xfs_attr3_rmt_verify

[Darrick J. Wong]

On Thu, Sep 17, 2020 at 07:38:46PM +0800, xiakaixu1987@gmail.com wrote:
> From: Kaixu Xia <kaixuxia@tencent.com>
> 
> We already check whether the crc feature is enabled before calling
> xfs_attr3_rmt_verify(), so remove the redundant feature check in that
> function.
> 
> Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>

Looks good,
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>

--D

> ---
>  fs/xfs/libxfs/xfs_attr_remote.c | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c
> index 3f80cede7406..48d8e9caf86f 100644
> --- a/fs/xfs/libxfs/xfs_attr_remote.c
> +++ b/fs/xfs/libxfs/xfs_attr_remote.c
> @@ -96,8 +96,6 @@ xfs_attr3_rmt_verify(
>  {
>  	struct xfs_attr3_rmt_hdr *rmt = ptr;
>  
> -	if (!xfs_sb_version_hascrc(&mp->m_sb))
> -		return __this_address;
>  	if (!xfs_verify_magic(bp, rmt->rm_magic))
>  		return __this_address;
>  	if (!uuid_equal(&rmt->rm_uuid, &mp->m_sb.sb_meta_uuid))
> -- 
> 2.20.0
> 

Re: [PATCH v2 5/7] xfs: remove the redundant crc feature check in xfs_attr3_rmt_verify

[Christoph Hellwig]

On Thu, Sep 17, 2020 at 07:38:46PM +0800, xiakaixu1987@gmail.com wrote:
> From: Kaixu Xia <kaixuxia@tencent.com>
> 
> We already check whether the crc feature is enabled before calling
> xfs_attr3_rmt_verify(), so remove the redundant feature check in that
> function.
> 
> Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 5/7] xfs: remove the redundant crc feature check in xfs_attr3_rmt_verify

[Christoph Hellwig]

On Thu, Sep 17, 2020 at 07:38:46PM +0800, xiakaixu1987@gmail.com wrote:
> From: Kaixu Xia <kaixuxia@tencent.com>
> 
> We already check whether the crc feature is enabled before calling
> xfs_attr3_rmt_verify(), so remove the redundant feature check in that
> function.
> 
> Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

[PATCH v2 6/7] xfs: code cleanup in xfs_attr_leaf_entsize_{remote,local}

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

Cleanup the typedef usage, the unnecessary parentheses, the unnecessary
backslash and use the open-coded round_up call in
xfs_attr_leaf_entsize_{remote,local}.

Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
---
 fs/xfs/libxfs/xfs_da_format.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
index b40a4e80f5ee..09f0f5d42728 100644
--- a/fs/xfs/libxfs/xfs_da_format.h
+++ b/fs/xfs/libxfs/xfs_da_format.h
@@ -746,14 +746,14 @@ xfs_attr3_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx)
  */
 static inline int xfs_attr_leaf_entsize_remote(int nlen)
 {
-	return ((uint)sizeof(xfs_attr_leaf_name_remote_t) - 1 + (nlen) + \
-		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
+	return round_up(sizeof(struct xfs_attr_leaf_name_remote) - 1 + nlen,
+			XFS_ATTR_LEAF_NAME_ALIGN);
 }
 
 static inline int xfs_attr_leaf_entsize_local(int nlen, int vlen)
 {
-	return ((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) +
-		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
+	return round_up(sizeof(struct xfs_attr_leaf_name_local) - 1 + nlen + vlen,
+			XFS_ATTR_LEAF_NAME_ALIGN);
 }
 
 static inline int xfs_attr_leaf_entsize_local_max(int bsize)
-- 
2.20.0

Re: [PATCH v2 6/7] xfs: code cleanup in xfs_attr_leaf_entsize_{remote,local}

[Darrick J. Wong]

On Thu, Sep 17, 2020 at 07:38:47PM +0800, xiakaixu1987@gmail.com wrote:
> From: Kaixu Xia <kaixuxia@tencent.com>
> 
> Cleanup the typedef usage, the unnecessary parentheses, the unnecessary
> backslash and use the open-coded round_up call in
> xfs_attr_leaf_entsize_{remote,local}.
> 
> Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>

Seems alright,
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>

--D

> ---
>  fs/xfs/libxfs/xfs_da_format.h | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
> index b40a4e80f5ee..09f0f5d42728 100644
> --- a/fs/xfs/libxfs/xfs_da_format.h
> +++ b/fs/xfs/libxfs/xfs_da_format.h
> @@ -746,14 +746,14 @@ xfs_attr3_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx)
>   */
>  static inline int xfs_attr_leaf_entsize_remote(int nlen)
>  {
> -	return ((uint)sizeof(xfs_attr_leaf_name_remote_t) - 1 + (nlen) + \
> -		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
> +	return round_up(sizeof(struct xfs_attr_leaf_name_remote) - 1 + nlen,
> +			XFS_ATTR_LEAF_NAME_ALIGN);
>  }
>  
>  static inline int xfs_attr_leaf_entsize_local(int nlen, int vlen)
>  {
> -	return ((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) +
> -		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
> +	return round_up(sizeof(struct xfs_attr_leaf_name_local) - 1 + nlen + vlen,
> +			XFS_ATTR_LEAF_NAME_ALIGN);
>  }
>  
>  static inline int xfs_attr_leaf_entsize_local_max(int bsize)
> -- 
> 2.20.0
> 

Re: [PATCH v2 6/7] xfs: code cleanup in xfs_attr_leaf_entsize_{remote,local}

[Christoph Hellwig]

On Thu, Sep 17, 2020 at 07:38:47PM +0800, xiakaixu1987@gmail.com wrote:
>  static inline int xfs_attr_leaf_entsize_local(int nlen, int vlen)
>  {
> -	return ((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) +
> -		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
> +	return round_up(sizeof(struct xfs_attr_leaf_name_local) - 1 + nlen + vlen,

Please avoid the overly long line here.

Re: [PATCH v2 6/7] xfs: code cleanup in xfs_attr_leaf_entsize_{remote,local}

[kaixuxia]

On 2020/9/19 14:29, Christoph Hellwig wrote:
> On Thu, Sep 17, 2020 at 07:38:47PM +0800, xiakaixu1987@gmail.com wrote:
>>  static inline int xfs_attr_leaf_entsize_local(int nlen, int vlen)
>>  {
>> -	return ((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) +
>> -		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
>> +	return round_up(sizeof(struct xfs_attr_leaf_name_local) - 1 + nlen + vlen,
> 
> Please avoid the overly long line here.

Okay, will fix it in the next version.

Thanks,
Kaixu
> 

-- 
kaixuxia

[PATCH v2 7/7] xfs: fix some comments

[xiakaixu1987]

From: Kaixu Xia <kaixuxia@tencent.com>

Fix the comments to help people understand the code.

Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/libxfs/xfs_da_format.h | 4 ++--
 fs/xfs/xfs_dquot.c            | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
index 09f0f5d42728..2a72f1760169 100644
--- a/fs/xfs/libxfs/xfs_da_format.h
+++ b/fs/xfs/libxfs/xfs_da_format.h
@@ -35,8 +35,8 @@ typedef struct xfs_da_blkinfo {
  */
 #define XFS_DA3_NODE_MAGIC	0x3ebe	/* magic number: non-leaf blocks */
 #define XFS_ATTR3_LEAF_MAGIC	0x3bee	/* magic number: attribute leaf blks */
-#define	XFS_DIR3_LEAF1_MAGIC	0x3df1	/* magic number: v2 dirlf single blks */
-#define	XFS_DIR3_LEAFN_MAGIC	0x3dff	/* magic number: v2 dirlf multi blks */
+#define	XFS_DIR3_LEAF1_MAGIC	0x3df1	/* magic number: v3 dirlf single blks */
+#define	XFS_DIR3_LEAFN_MAGIC	0x3dff	/* magic number: v3 dirlf multi blks */
 
 struct xfs_da3_blkinfo {
 	/*
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index 3072814e407d..1d95ed387d66 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -831,8 +831,8 @@ xfs_qm_dqget_checks(
 }
 
 /*
- * Given the file system, id, and type (UDQUOT/GDQUOT), return a locked
- * dquot, doing an allocation (if requested) as needed.
+ * Given the file system, id, and type (UDQUOT/GDQUOT/PDQUOT), return a
+ * locked dquot, doing an allocation (if requested) as needed.
  */
 int
 xfs_qm_dqget(
-- 
2.20.0

Re: [PATCH v2 7/7] xfs: fix some comments

[Christoph Hellwig]

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>