diff mbox series

[RESEND,v2,02/18] xfs: Increase XFS_DEFER_OPS_NR_INODES to 5

Message ID 20220804194013.99237-3-allison.henderson@oracle.com (mailing list archive)
State Superseded
Headers show
Series Parent Pointers | expand

Commit Message

Allison Henderson Aug. 4, 2022, 7:39 p.m. UTC 
Renames that generate parent pointer updates can join up to 5
inodes locked in sorted order.  So we need to increase the
number of defer ops inodes and relock them in the same way.

Signed-off-by: Allison Henderson <allison.henderson@oracle.com>
---
 fs/xfs/libxfs/xfs_defer.c | 28 ++++++++++++++++++++++++++--
 fs/xfs/libxfs/xfs_defer.h |  8 +++++++-
 fs/xfs/xfs_inode.c        |  2 +-
 fs/xfs/xfs_inode.h        |  1 +
 4 files changed, 35 insertions(+), 4 deletions(-)

Comments

Darrick J. Wong Aug. 9, 2022, 4:38 p.m. UTC | #1 
On Thu, Aug 04, 2022 at 12:39:57PM -0700, Allison Henderson wrote:
> Renames that generate parent pointer updates can join up to 5
> inodes locked in sorted order.  So we need to increase the
> number of defer ops inodes and relock them in the same way.
> 
> Signed-off-by: Allison Henderson <allison.henderson@oracle.com>
> ---
>  fs/xfs/libxfs/xfs_defer.c | 28 ++++++++++++++++++++++++++--
>  fs/xfs/libxfs/xfs_defer.h |  8 +++++++-
>  fs/xfs/xfs_inode.c        |  2 +-
>  fs/xfs/xfs_inode.h        |  1 +
>  4 files changed, 35 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c
> index 5a321b783398..c0279b57e51d 100644
> --- a/fs/xfs/libxfs/xfs_defer.c
> +++ b/fs/xfs/libxfs/xfs_defer.c
> @@ -820,13 +820,37 @@ xfs_defer_ops_continue(
>  	struct xfs_trans		*tp,
>  	struct xfs_defer_resources	*dres)
>  {
> -	unsigned int			i;
> +	unsigned int			i, j;
> +	struct xfs_inode		*sips[XFS_DEFER_OPS_NR_INODES];
> +	struct xfs_inode		*temp;
>  
>  	ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
>  	ASSERT(!(tp->t_flags & XFS_TRANS_DIRTY));
>  
>  	/* Lock the captured resources to the new transaction. */
> -	if (dfc->dfc_held.dr_inos == 2)
> +	if (dfc->dfc_held.dr_inos > 2) {
> +		/*
> +		 * Renames with parent pointer updates can lock up to 5 inodes,
> +		 * sorted by their inode number.  So we need to make sure they
> +		 * are relocked in the same way.
> +		 */
> +		memset(sips, 0, sizeof(sips));
> +		for (i = 0; i < dfc->dfc_held.dr_inos; i++)
> +			sips[i] = dfc->dfc_held.dr_ip[i];
> +
> +		/* Bubble sort of at most 5 inodes */
> +		for (i = 0; i < dfc->dfc_held.dr_inos; i++) {
> +			for (j = 1; j < dfc->dfc_held.dr_inos; j++) {
> +				if (sips[j]->i_ino < sips[j-1]->i_ino) {
> +					temp = sips[j];
> +					sips[j] = sips[j-1];
> +					sips[j-1] = temp;
> +				}
> +			}
> +		}

Why not reuse xfs_sort_for_rename?

I also wonder if it's worth the trouble to replace the open-coded
bubblesort with a call to sort_r(), but TBH I suspect the cost of a
retpoline for the compare function isn't worth the overhead.

> +
> +		xfs_lock_inodes(sips, dfc->dfc_held.dr_inos, XFS_ILOCK_EXCL);
> +	} else if (dfc->dfc_held.dr_inos == 2)
>  		xfs_lock_two_inodes(dfc->dfc_held.dr_ip[0], XFS_ILOCK_EXCL,
>  				    dfc->dfc_held.dr_ip[1], XFS_ILOCK_EXCL);
>  	else if (dfc->dfc_held.dr_inos == 1)
> diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h
> index 114a3a4930a3..3e4029d2ce41 100644
> --- a/fs/xfs/libxfs/xfs_defer.h
> +++ b/fs/xfs/libxfs/xfs_defer.h
> @@ -70,7 +70,13 @@ extern const struct xfs_defer_op_type xfs_attr_defer_type;
>  /*
>   * Deferred operation item relogging limits.
>   */
> -#define XFS_DEFER_OPS_NR_INODES	2	/* join up to two inodes */
> +
> +/*
> + * Rename w/ parent pointers can require up to 5 inodes with defered ops to
> + * be joined to the transaction: src_dp, target_dp, src_ip, target_ip, and wip.
> + * These inodes are locked in sorted order by their inode numbers

Much inode.  Thanks for recording this.

--D

> + */
> +#define XFS_DEFER_OPS_NR_INODES	5
>  #define XFS_DEFER_OPS_NR_BUFS	2	/* join up to two buffers */
>  
>  /* Resources that must be held across a transaction roll. */
> diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
> index 3022918bf96a..cfdcca95594f 100644
> --- a/fs/xfs/xfs_inode.c
> +++ b/fs/xfs/xfs_inode.c
> @@ -447,7 +447,7 @@ xfs_lock_inumorder(
>   * lock more than one at a time, lockdep will report false positives saying we
>   * have violated locking orders.
>   */
> -static void
> +void
>  xfs_lock_inodes(
>  	struct xfs_inode	**ips,
>  	int			inodes,
> diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
> index 4d626f4321bc..bc06d6e4164a 100644
> --- a/fs/xfs/xfs_inode.h
> +++ b/fs/xfs/xfs_inode.h
> @@ -573,5 +573,6 @@ void xfs_end_io(struct work_struct *work);
>  
>  int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2);
>  void xfs_iunlock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2);
> +void xfs_lock_inodes(struct xfs_inode **ips, int inodes, uint lock_mode);
>  
>  #endif	/* __XFS_INODE_H__ */
> -- 
> 2.25.1
>
Allison Henderson Aug. 10, 2022, 3:07 a.m. UTC | #2 
On Tue, 2022-08-09 at 09:38 -0700, Darrick J. Wong wrote:
> On Thu, Aug 04, 2022 at 12:39:57PM -0700, Allison Henderson wrote:
> > Renames that generate parent pointer updates can join up to 5
> > inodes locked in sorted order.  So we need to increase the
> > number of defer ops inodes and relock them in the same way.
> > 
> > Signed-off-by: Allison Henderson <allison.henderson@oracle.com>
> > ---
> >  fs/xfs/libxfs/xfs_defer.c | 28 ++++++++++++++++++++++++++--
> >  fs/xfs/libxfs/xfs_defer.h |  8 +++++++-
> >  fs/xfs/xfs_inode.c        |  2 +-
> >  fs/xfs/xfs_inode.h        |  1 +
> >  4 files changed, 35 insertions(+), 4 deletions(-)
> > 
> > diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c
> > index 5a321b783398..c0279b57e51d 100644
> > --- a/fs/xfs/libxfs/xfs_defer.c
> > +++ b/fs/xfs/libxfs/xfs_defer.c
> > @@ -820,13 +820,37 @@ xfs_defer_ops_continue(
> >  	struct xfs_trans		*tp,
> >  	struct xfs_defer_resources	*dres)
> >  {
> > -	unsigned int			i;
> > +	unsigned int			i, j;
> > +	struct xfs_inode		*sips[XFS_DEFER_OPS_NR_INODES];
> > +	struct xfs_inode		*temp;
> >  
> >  	ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
> >  	ASSERT(!(tp->t_flags & XFS_TRANS_DIRTY));
> >  
> >  	/* Lock the captured resources to the new transaction. */
> > -	if (dfc->dfc_held.dr_inos == 2)
> > +	if (dfc->dfc_held.dr_inos > 2) {
> > +		/*
> > +		 * Renames with parent pointer updates can lock up to 5
> > inodes,
> > +		 * sorted by their inode number.  So we need to make
> > sure they
> > +		 * are relocked in the same way.
> > +		 */
> > +		memset(sips, 0, sizeof(sips));
> > +		for (i = 0; i < dfc->dfc_held.dr_inos; i++)
> > +			sips[i] = dfc->dfc_held.dr_ip[i];
> > +
> > +		/* Bubble sort of at most 5 inodes */
> > +		for (i = 0; i < dfc->dfc_held.dr_inos; i++) {
> > +			for (j = 1; j < dfc->dfc_held.dr_inos; j++) {
> > +				if (sips[j]->i_ino < sips[j-1]->i_ino)
> > {
> > +					temp = sips[j];
> > +					sips[j] = sips[j-1];
> > +					sips[j-1] = temp;
> > +				}
> > +			}
> > +		}
> 
> Why not reuse xfs_sort_for_rename?
Initially I had looked at doing that, but it would need some
refactoring as it is not meant for an arbitrary number of inodes.
Either some logic specific to rename would get pulled up, or we'd need
another helper to repackage the parameters, but it's such a small bit
of code, I'm not sure it saves much LOC either way.  

> 
> I also wonder if it's worth the trouble to replace the open-coded
> bubblesort with a call to sort_r(), but TBH I suspect the cost of a
> retpoline for the compare function isn't worth the overhead.
Yeah, it would make sense if there was lot of other places we sorted
inodes, but with only two callers it does seem like a bit much.

I am fine with what ever method folks prefer tho.

> 
> > +
> > +		xfs_lock_inodes(sips, dfc->dfc_held.dr_inos,
> > XFS_ILOCK_EXCL);
> > +	} else if (dfc->dfc_held.dr_inos == 2)
> >  		xfs_lock_two_inodes(dfc->dfc_held.dr_ip[0],
> > XFS_ILOCK_EXCL,
> >  				    dfc->dfc_held.dr_ip[1],
> > XFS_ILOCK_EXCL);
> >  	else if (dfc->dfc_held.dr_inos == 1)
> > diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h
> > index 114a3a4930a3..3e4029d2ce41 100644
> > --- a/fs/xfs/libxfs/xfs_defer.h
> > +++ b/fs/xfs/libxfs/xfs_defer.h
> > @@ -70,7 +70,13 @@ extern const struct xfs_defer_op_type
> > xfs_attr_defer_type;
> >  /*
> >   * Deferred operation item relogging limits.
> >   */
> > -#define XFS_DEFER_OPS_NR_INODES	2	/* join up to two inodes */
> > +
> > +/*
> > + * Rename w/ parent pointers can require up to 5 inodes with
> > defered ops to
> > + * be joined to the transaction: src_dp, target_dp, src_ip,
> > target_ip, and wip.
> > + * These inodes are locked in sorted order by their inode numbers
> 
> Much inode.  Thanks for recording this.
Sure, thx for the reviews!

Allison
> 
> --D
> 
> > + */
> > +#define XFS_DEFER_OPS_NR_INODES	5
> >  #define XFS_DEFER_OPS_NR_BUFS	2	/* join up to two buffers
> > */
> >  
> >  /* Resources that must be held across a transaction roll. */
> > diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
> > index 3022918bf96a..cfdcca95594f 100644
> > --- a/fs/xfs/xfs_inode.c
> > +++ b/fs/xfs/xfs_inode.c
> > @@ -447,7 +447,7 @@ xfs_lock_inumorder(
> >   * lock more than one at a time, lockdep will report false
> > positives saying we
> >   * have violated locking orders.
> >   */
> > -static void
> > +void
> >  xfs_lock_inodes(
> >  	struct xfs_inode	**ips,
> >  	int			inodes,
> > diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
> > index 4d626f4321bc..bc06d6e4164a 100644
> > --- a/fs/xfs/xfs_inode.h
> > +++ b/fs/xfs/xfs_inode.h
> > @@ -573,5 +573,6 @@ void xfs_end_io(struct work_struct *work);
> >  
> >  int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode
> > *ip2);
> >  void xfs_iunlock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode
> > *ip2);
> > +void xfs_lock_inodes(struct xfs_inode **ips, int inodes, uint
> > lock_mode);
> >  
> >  #endif	/* __XFS_INODE_H__ */
> > -- 
> > 2.25.1
> >
diff mbox series

Patch

diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c
index 5a321b783398..c0279b57e51d 100644
--- a/fs/xfs/libxfs/xfs_defer.c
+++ b/fs/xfs/libxfs/xfs_defer.c
@@ -820,13 +820,37 @@  xfs_defer_ops_continue(
 	struct xfs_trans		*tp,
 	struct xfs_defer_resources	*dres)
 {
-	unsigned int			i;
+	unsigned int			i, j;
+	struct xfs_inode		*sips[XFS_DEFER_OPS_NR_INODES];
+	struct xfs_inode		*temp;
 
 	ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
 	ASSERT(!(tp->t_flags & XFS_TRANS_DIRTY));
 
 	/* Lock the captured resources to the new transaction. */
-	if (dfc->dfc_held.dr_inos == 2)
+	if (dfc->dfc_held.dr_inos > 2) {
+		/*
+		 * Renames with parent pointer updates can lock up to 5 inodes,
+		 * sorted by their inode number.  So we need to make sure they
+		 * are relocked in the same way.
+		 */
+		memset(sips, 0, sizeof(sips));
+		for (i = 0; i < dfc->dfc_held.dr_inos; i++)
+			sips[i] = dfc->dfc_held.dr_ip[i];
+
+		/* Bubble sort of at most 5 inodes */
+		for (i = 0; i < dfc->dfc_held.dr_inos; i++) {
+			for (j = 1; j < dfc->dfc_held.dr_inos; j++) {
+				if (sips[j]->i_ino < sips[j-1]->i_ino) {
+					temp = sips[j];
+					sips[j] = sips[j-1];
+					sips[j-1] = temp;
+				}
+			}
+		}
+
+		xfs_lock_inodes(sips, dfc->dfc_held.dr_inos, XFS_ILOCK_EXCL);
+	} else if (dfc->dfc_held.dr_inos == 2)
 		xfs_lock_two_inodes(dfc->dfc_held.dr_ip[0], XFS_ILOCK_EXCL,
 				    dfc->dfc_held.dr_ip[1], XFS_ILOCK_EXCL);
 	else if (dfc->dfc_held.dr_inos == 1)
diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h
index 114a3a4930a3..3e4029d2ce41 100644
--- a/fs/xfs/libxfs/xfs_defer.h
+++ b/fs/xfs/libxfs/xfs_defer.h
@@ -70,7 +70,13 @@  extern const struct xfs_defer_op_type xfs_attr_defer_type;
 /*
  * Deferred operation item relogging limits.
  */
-#define XFS_DEFER_OPS_NR_INODES	2	/* join up to two inodes */
+
+/*
+ * Rename w/ parent pointers can require up to 5 inodes with defered ops to
+ * be joined to the transaction: src_dp, target_dp, src_ip, target_ip, and wip.
+ * These inodes are locked in sorted order by their inode numbers
+ */
+#define XFS_DEFER_OPS_NR_INODES	5
 #define XFS_DEFER_OPS_NR_BUFS	2	/* join up to two buffers */
 
 /* Resources that must be held across a transaction roll. */
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 3022918bf96a..cfdcca95594f 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -447,7 +447,7 @@  xfs_lock_inumorder(
  * lock more than one at a time, lockdep will report false positives saying we
  * have violated locking orders.
  */
-static void
+void
 xfs_lock_inodes(
 	struct xfs_inode	**ips,
 	int			inodes,
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 4d626f4321bc..bc06d6e4164a 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -573,5 +573,6 @@  void xfs_end_io(struct work_struct *work);
 
 int xfs_ilock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2);
 void xfs_iunlock2_io_mmap(struct xfs_inode *ip1, struct xfs_inode *ip2);
+void xfs_lock_inodes(struct xfs_inode **ips, int inodes, uint lock_mode);
 
 #endif	/* __XFS_INODE_H__ */