Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 5619a693 authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge branch 'for-linus' of git://oss.sgi.com/xfs/xfs 

* 'for-linus' of git://oss.sgi.com/xfs/xfs: (69 commits)
  xfs: add AIL pushing tracepoints
  xfs: put in missed fix for merge problem
  xfs: do not flush data workqueues in xfs_flush_buftarg
  xfs: remove XFS_bflush
  xfs: remove xfs_buf_target_name
  xfs: use xfs_ioerror_alert in xfs_buf_iodone_callbacks
  xfs: clean up xfs_ioerror_alert
  xfs: clean up buffer allocation
  xfs: remove buffers from the delwri list in xfs_buf_stale
  xfs: remove XFS_BUF_STALE and XFS_BUF_SUPER_STALE
  xfs: remove XFS_BUF_SET_VTYPE and XFS_BUF_SET_VTYPE_REF
  xfs: remove XFS_BUF_FINISH_IOWAIT
  xfs: remove xfs_get_buftarg_list
  xfs: fix buffer flushing during unmount
  xfs: optimize fsync on directories
  xfs: reduce the number of log forces from tail pushing
  xfs: Don't allocate new buffers on every call to _xfs_buf_find
  xfs: simplify xfs_trans_ijoin* again
  xfs: unlock the inode before log force in xfs_change_file_space
  xfs: unlock the inode before log force in xfs_fs_nfs_commit_metadata
  ...
parents 37be944a 9e4c109a

fs/xfs/xfs_alloc.c

+2 −2
Original line number Diff line number Diff line
@@ -452,7 +452,7 @@ xfs_alloc_read_agfl( 
	if (error)

		return error;

	ASSERT(!xfs_buf_geterror(bp));

	XFS_BUF_SET_VTYPE_REF(bp, B_FS_AGFL, XFS_AGFL_REF);

	xfs_buf_set_ref(bp, XFS_AGFL_REF);

	*bpp = bp;

	return 0;

}
@@ -2139,7 +2139,7 @@ xfs_read_agf( 
		xfs_trans_brelse(tp, *bpp);

		return XFS_ERROR(EFSCORRUPTED);

	}

	XFS_BUF_SET_VTYPE_REF(*bpp, B_FS_AGF, XFS_AGF_REF);

	xfs_buf_set_ref(*bpp, XFS_AGF_REF);

	return 0;

}

fs/xfs/xfs_aops.c

+42 −77
Original line number Diff line number Diff line
@@ -38,40 +38,6 @@ 
#include <linux/pagevec.h>

#include <linux/writeback.h>





/*

 * Prime number of hash buckets since address is used as the key.

 */

#define NVSYNC		37

#define to_ioend_wq(v)	(&xfs_ioend_wq[((unsigned long)v) % NVSYNC])

static wait_queue_head_t xfs_ioend_wq[NVSYNC];



void __init

xfs_ioend_init(void)

{

	int i;



	for (i = 0; i < NVSYNC; i++)

		init_waitqueue_head(&xfs_ioend_wq[i]);

}



void

xfs_ioend_wait(

	xfs_inode_t	*ip)

{

	wait_queue_head_t *wq = to_ioend_wq(ip);



	wait_event(*wq, (atomic_read(&ip->i_iocount) == 0));

}



STATIC void

xfs_ioend_wake(

	xfs_inode_t	*ip)

{

	if (atomic_dec_and_test(&ip->i_iocount))

		wake_up(to_ioend_wq(ip));

}



void

xfs_count_page_state(

	struct page		*page,
@@ -115,25 +81,20 @@ xfs_destroy_ioend( 
	xfs_ioend_t		*ioend)

{

	struct buffer_head	*bh, *next;

	struct xfs_inode	*ip = XFS_I(ioend->io_inode);



	for (bh = ioend->io_buffer_head; bh; bh = next) {

		next = bh->b_private;

		bh->b_end_io(bh, !ioend->io_error);

	}



	/*

	 * Volume managers supporting multiple paths can send back ENODEV

	 * when the final path disappears.  In this case continuing to fill

	 * the page cache with dirty data which cannot be written out is

	 * evil, so prevent that.

	 */

	if (unlikely(ioend->io_error == -ENODEV)) {

		xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ,

				      __FILE__, __LINE__);

	if (ioend->io_iocb) {

		if (ioend->io_isasync) {

			aio_complete(ioend->io_iocb, ioend->io_error ?

					ioend->io_error : ioend->io_result, 0);

		}

		inode_dio_done(ioend->io_inode);

	}



	xfs_ioend_wake(ip);

	mempool_free(ioend, xfs_ioend_pool);

}
@@ -155,6 +116,15 @@ xfs_ioend_new_eof( 
	return isize > ip->i_d.di_size ? isize : 0;

}



/*

 * Fast and loose check if this write could update the on-disk inode size.

 */

static inline bool xfs_ioend_is_append(struct xfs_ioend *ioend)

{

	return ioend->io_offset + ioend->io_size >

		XFS_I(ioend->io_inode)->i_d.di_size;

}



/*

 * Update on-disk file size now that data has been written to disk.  The

 * current in-memory file size is i_size.  If a write is beyond eof i_new_size
@@ -173,9 +143,6 @@ xfs_setfilesize( 
	xfs_inode_t		*ip = XFS_I(ioend->io_inode);

	xfs_fsize_t		isize;



	if (unlikely(ioend->io_error))

		return 0;



	if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL))

		return EAGAIN;
@@ -192,6 +159,9 @@ xfs_setfilesize( 


/*

 * Schedule IO completion handling on the final put of an ioend.

 *

 * If there is no work to do we might as well call it a day and free the

 * ioend right now.

 */

STATIC void

xfs_finish_ioend(
@@ -200,8 +170,10 @@ xfs_finish_ioend( 
	if (atomic_dec_and_test(&ioend->io_remaining)) {

		if (ioend->io_type == IO_UNWRITTEN)

			queue_work(xfsconvertd_workqueue, &ioend->io_work);

		else

		else if (xfs_ioend_is_append(ioend))

			queue_work(xfsdatad_workqueue, &ioend->io_work);

		else

			xfs_destroy_ioend(ioend);

	}

}
@@ -216,17 +188,24 @@ xfs_end_io( 
	struct xfs_inode *ip = XFS_I(ioend->io_inode);

	int		error = 0;



	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {

		error = -EIO;

		goto done;

	}

	if (ioend->io_error)

		goto done;



	/*

	 * For unwritten extents we need to issue transactions to convert a

	 * range to normal written extens after the data I/O has finished.

	 */

	if (ioend->io_type == IO_UNWRITTEN &&

	    likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) {



	if (ioend->io_type == IO_UNWRITTEN) {

		error = xfs_iomap_write_unwritten(ip, ioend->io_offset,

						 ioend->io_size);

		if (error)

			ioend->io_error = error;

		if (error) {

			ioend->io_error = -error;

			goto done;

		}

	}



	/*
@@ -236,6 +215,7 @@ xfs_end_io( 
	error = xfs_setfilesize(ioend);

	ASSERT(!error || error == EAGAIN);



done:

	/*

	 * If we didn't complete processing of the ioend, requeue it to the

	 * tail of the workqueue for another attempt later. Otherwise destroy
@@ -247,8 +227,6 @@ xfs_end_io( 
		/* ensure we don't spin on blocked ioends */

		delay(1);

	} else {

		if (ioend->io_iocb)

			aio_complete(ioend->io_iocb, ioend->io_result, 0);

		xfs_destroy_ioend(ioend);

	}

}
@@ -285,13 +263,13 @@ xfs_alloc_ioend( 
	 * all the I/O from calling the completion routine too early.

	 */

	atomic_set(&ioend->io_remaining, 1);

	ioend->io_isasync = 0;

	ioend->io_error = 0;

	ioend->io_list = NULL;

	ioend->io_type = type;

	ioend->io_inode = inode;

	ioend->io_buffer_head = NULL;

	ioend->io_buffer_tail = NULL;

	atomic_inc(&XFS_I(ioend->io_inode)->i_iocount);

	ioend->io_offset = 0;

	ioend->io_size = 0;

	ioend->io_iocb = NULL;
@@ -337,8 +315,8 @@ xfs_map_blocks( 
		count = mp->m_maxioffset - offset;

	end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);

	offset_fsb = XFS_B_TO_FSBT(mp, offset);

	error = xfs_bmapi(NULL, ip, offset_fsb, end_fsb - offset_fsb,

			  bmapi_flags,  NULL, 0, imap, &nimaps, NULL);

	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb,

				imap, &nimaps, bmapi_flags);

	xfs_iunlock(ip, XFS_ILOCK_SHARED);



	if (error)
@@ -551,7 +529,6 @@ xfs_cancel_ioend( 
			unlock_buffer(bh);

		} while ((bh = next_bh) != NULL);



		xfs_ioend_wake(XFS_I(ioend->io_inode));

		mempool_free(ioend, xfs_ioend_pool);

	} while ((ioend = next) != NULL);

}
@@ -1161,8 +1138,8 @@ __xfs_get_blocks( 
	end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + size);

	offset_fsb = XFS_B_TO_FSBT(mp, offset);



	error = xfs_bmapi(NULL, ip, offset_fsb, end_fsb - offset_fsb,

			  XFS_BMAPI_ENTIRE,  NULL, 0, &imap, &nimaps, NULL);

	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb,

				&imap, &nimaps, XFS_BMAPI_ENTIRE);

	if (error)

		goto out_unlock;
@@ -1300,7 +1277,6 @@ xfs_end_io_direct_write( 
	bool			is_async)

{

	struct xfs_ioend	*ioend = iocb->private;

	struct inode		*inode = ioend->io_inode;



	/*

	 * blockdev_direct_IO can return an error even after the I/O
@@ -1311,28 +1287,17 @@ xfs_end_io_direct_write( 


	ioend->io_offset = offset;

	ioend->io_size = size;

	ioend->io_iocb = iocb;

	ioend->io_result = ret;

	if (private && size > 0)

		ioend->io_type = IO_UNWRITTEN;



	if (is_async) {

		/*

		 * If we are converting an unwritten extent we need to delay

		 * the AIO completion until after the unwrittent extent

		 * conversion has completed, otherwise do it ASAP.

		 */

		if (ioend->io_type == IO_UNWRITTEN) {

			ioend->io_iocb = iocb;

			ioend->io_result = ret;

		} else {

			aio_complete(iocb, ret, 0);

		}

		ioend->io_isasync = 1;

		xfs_finish_ioend(ioend);

	} else {

		xfs_finish_ioend_sync(ioend);

	}



	/* XXX: probably should move into the real I/O completion handler */

	inode_dio_done(inode);

}



STATIC ssize_t

fs/xfs/xfs_aops.h

+1 −3
Original line number Diff line number Diff line
@@ -47,6 +47,7 @@ typedef struct xfs_ioend { 
	unsigned int		io_type;	/* delalloc / unwritten */

	int			io_error;	/* I/O error code */

	atomic_t		io_remaining;	/* hold count */

	unsigned int		io_isasync : 1;	/* needs aio_complete */

	struct inode		*io_inode;	/* file being written to */

	struct buffer_head	*io_buffer_head;/* buffer linked list head */

	struct buffer_head	*io_buffer_tail;/* buffer linked list tail */
@@ -60,9 +61,6 @@ typedef struct xfs_ioend { 
extern const struct address_space_operations xfs_address_space_operations;

extern int xfs_get_blocks(struct inode *, sector_t, struct buffer_head *, int);



extern void xfs_ioend_init(void);

extern void xfs_ioend_wait(struct xfs_inode *);



extern void xfs_count_page_state(struct page *, int *, int *);



#endif /* __XFS_AOPS_H__ */

fs/xfs/xfs_attr.c

+34 −55
Original line number Diff line number Diff line
@@ -319,7 +319,7 @@ xfs_attr_set_int( 
		return (error);

	}



	xfs_trans_ijoin(args.trans, dp);

	xfs_trans_ijoin(args.trans, dp, 0);



	/*

	 * If the attribute list is non-existent or a shortform list,
@@ -389,7 +389,7 @@ xfs_attr_set_int( 
		 * a new one.  We need the inode to be in all transactions.

		 */

		if (committed)

			xfs_trans_ijoin(args.trans, dp);

			xfs_trans_ijoin(args.trans, dp, 0);



		/*

		 * Commit the leaf transformation.  We'll need another (linked)
@@ -537,7 +537,7 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags) 
	 * No need to make quota reservations here. We expect to release some

	 * blocks not allocate in the common case.

	 */

	xfs_trans_ijoin(args.trans, dp);

	xfs_trans_ijoin(args.trans, dp, 0);



	/*

	 * Decide on what work routines to call based on the inode size.
@@ -809,7 +809,7 @@ xfs_attr_inactive(xfs_inode_t *dp) 
	 * No need to make quota reservations here. We expect to release some

	 * blocks, not allocate, in the common case.

	 */

	xfs_trans_ijoin(trans, dp);

	xfs_trans_ijoin(trans, dp, 0);



	/*

	 * Decide on what work routines to call based on the inode size.
@@ -823,18 +823,6 @@ xfs_attr_inactive(xfs_inode_t *dp) 
	if (error)

		goto out;



	/*

	 * Signal synchronous inactive transactions unless this is a

	 * synchronous mount filesystem in which case we know that we're here

	 * because we've been called out of xfs_inactive which means that the

	 * last reference is gone and the unlink transaction has already hit

	 * the disk so async inactive transactions are safe.

	 */

	if (!(mp->m_flags & XFS_MOUNT_WSYNC)) {

		if (dp->i_d.di_anextents > 0)

			xfs_trans_set_sync(trans);

	}



	error = xfs_itruncate_extents(&trans, dp, XFS_ATTR_FORK, 0);

	if (error)

		goto out;
@@ -973,7 +961,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) 
		 * a new one.  We need the inode to be in all transactions.

		 */

		if (committed)

			xfs_trans_ijoin(args->trans, dp);

			xfs_trans_ijoin(args->trans, dp, 0);



		/*

		 * Commit the current trans (including the inode) and start
@@ -1075,7 +1063,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) 
			 * in all transactions.

			 */

			if (committed)

				xfs_trans_ijoin(args->trans, dp);

				xfs_trans_ijoin(args->trans, dp, 0);

		} else

			xfs_da_buf_done(bp);
@@ -1149,7 +1137,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args) 
		 * a new one.  We need the inode to be in all transactions.

		 */

		if (committed)

			xfs_trans_ijoin(args->trans, dp);

			xfs_trans_ijoin(args->trans, dp, 0);

	} else

		xfs_da_buf_done(bp);

	return(0);
@@ -1303,7 +1291,7 @@ xfs_attr_node_addname(xfs_da_args_t *args) 
			 * in all transactions.

			 */

			if (committed)

				xfs_trans_ijoin(args->trans, dp);

				xfs_trans_ijoin(args->trans, dp, 0);



			/*

			 * Commit the node conversion and start the next
@@ -1340,7 +1328,7 @@ xfs_attr_node_addname(xfs_da_args_t *args) 
		 * a new one.  We need the inode to be in all transactions.

		 */

		if (committed)

			xfs_trans_ijoin(args->trans, dp);

			xfs_trans_ijoin(args->trans, dp, 0);

	} else {

		/*

		 * Addition succeeded, update Btree hashvals.
@@ -1452,7 +1440,7 @@ xfs_attr_node_addname(xfs_da_args_t *args) 
			 * in all transactions.

			 */

			if (committed)

				xfs_trans_ijoin(args->trans, dp);

				xfs_trans_ijoin(args->trans, dp, 0);

		}



		/*
@@ -1584,7 +1572,7 @@ xfs_attr_node_removename(xfs_da_args_t *args) 
		 * a new one.  We need the inode to be in all transactions.

		 */

		if (committed)

			xfs_trans_ijoin(args->trans, dp);

			xfs_trans_ijoin(args->trans, dp, 0);



		/*

		 * Commit the Btree join operation and start a new trans.
@@ -1635,7 +1623,7 @@ xfs_attr_node_removename(xfs_da_args_t *args) 
			 * in all transactions.

			 */

			if (committed)

				xfs_trans_ijoin(args->trans, dp);

				xfs_trans_ijoin(args->trans, dp, 0);

		} else

			xfs_da_brelse(args->trans, bp);

	}
@@ -1975,10 +1963,9 @@ xfs_attr_rmtval_get(xfs_da_args_t *args) 
	lblkno = args->rmtblkno;

	while (valuelen > 0) {

		nmap = ATTR_RMTVALUE_MAPSIZE;

		error = xfs_bmapi(args->trans, args->dp, (xfs_fileoff_t)lblkno,

				  args->rmtblkcnt,

				  XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,

				  NULL, 0, map, &nmap, NULL);

		error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,

				       args->rmtblkcnt, map, &nmap,

				       XFS_BMAPI_ATTRFORK);

		if (error)

			return(error);

		ASSERT(nmap >= 1);
@@ -2052,10 +2039,9 @@ xfs_attr_rmtval_set(xfs_da_args_t *args) 
		 */

		xfs_bmap_init(args->flist, args->firstblock);

		nmap = 1;

		error = xfs_bmapi(args->trans, dp, (xfs_fileoff_t)lblkno,

		error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno,

				  blkcnt,

				  XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA |

							XFS_BMAPI_WRITE,

				  XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,

				  args->firstblock, args->total, &map, &nmap,

				  args->flist);

		if (!error) {
@@ -2074,7 +2060,7 @@ xfs_attr_rmtval_set(xfs_da_args_t *args) 
		 * a new one.  We need the inode to be in all transactions.

		 */

		if (committed)

			xfs_trans_ijoin(args->trans, dp);

			xfs_trans_ijoin(args->trans, dp, 0);



		ASSERT(nmap == 1);

		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
@@ -2104,14 +2090,11 @@ xfs_attr_rmtval_set(xfs_da_args_t *args) 
		 */

		xfs_bmap_init(args->flist, args->firstblock);

		nmap = 1;

		error = xfs_bmapi(NULL, dp, (xfs_fileoff_t)lblkno,

				  args->rmtblkcnt,

				  XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,

				  args->firstblock, 0, &map, &nmap,

				  NULL);

		if (error) {

		error = xfs_bmapi_read(dp, (xfs_fileoff_t)lblkno,

				       args->rmtblkcnt, &map, &nmap,

				       XFS_BMAPI_ATTRFORK);

		if (error)

			return(error);

		}

		ASSERT(nmap == 1);

		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&

		       (map.br_startblock != HOLESTARTBLOCK));
@@ -2121,16 +2104,17 @@ xfs_attr_rmtval_set(xfs_da_args_t *args) 


		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt,

				 XBF_LOCK | XBF_DONT_BLOCK);

		ASSERT(!xfs_buf_geterror(bp));



		if (!bp)

			return ENOMEM;

		tmp = (valuelen < XFS_BUF_SIZE(bp)) ? valuelen :

							XFS_BUF_SIZE(bp);

		xfs_buf_iomove(bp, 0, tmp, src, XBRW_WRITE);

		if (tmp < XFS_BUF_SIZE(bp))

			xfs_buf_zero(bp, tmp, XFS_BUF_SIZE(bp) - tmp);

		if ((error = xfs_bwrite(mp, bp))) {/* GROT: NOTE: synchronous write */

			return (error);

		}

		error = xfs_bwrite(bp);	/* GROT: NOTE: synchronous write */

		xfs_buf_relse(bp);

		if (error)

			return error;

		src += tmp;

		valuelen -= tmp;
@@ -2166,16 +2150,12 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args) 
		/*

		 * Try to remember where we decided to put the value.

		 */

		xfs_bmap_init(args->flist, args->firstblock);

		nmap = 1;

		error = xfs_bmapi(NULL, args->dp, (xfs_fileoff_t)lblkno,

					args->rmtblkcnt,

					XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,

					args->firstblock, 0, &map, &nmap,

					args->flist);

		if (error) {

		error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,

				       args->rmtblkcnt, &map, &nmap,

				       XFS_BMAPI_ATTRFORK);

		if (error)

			return(error);

		}

		ASSERT(nmap == 1);

		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&

		       (map.br_startblock != HOLESTARTBLOCK));
@@ -2188,8 +2168,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args) 
		 */

		bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt, XBF_TRYLOCK);

		if (bp) {

			XFS_BUF_STALE(bp);

			XFS_BUF_UNDELAYWRITE(bp);

			xfs_buf_stale(bp);

			xfs_buf_relse(bp);

			bp = NULL;

		}
@@ -2227,7 +2206,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args) 
		 * a new one.  We need the inode to be in all transactions.

		 */

		if (committed)

			xfs_trans_ijoin(args->trans, args->dp);

			xfs_trans_ijoin(args->trans, args->dp, 0);



		/*

		 * Close out trans and start the next one in the chain.

fs/xfs/xfs_attr_leaf.c

+4 −3
Original line number Diff line number Diff line
@@ -2926,9 +2926,8 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp, 
		 * Try to remember where we decided to put the value.

		 */

		nmap = 1;

		error = xfs_bmapi(*trans, dp, (xfs_fileoff_t)tblkno, tblkcnt,

					XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,

					NULL, 0, &map, &nmap, NULL);

		error = xfs_bmapi_read(dp, (xfs_fileoff_t)tblkno, tblkcnt,

				       &map, &nmap, XFS_BMAPI_ATTRFORK);

		if (error) {

			return(error);

		}
@@ -2948,6 +2947,8 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp, 
			bp = xfs_trans_get_buf(*trans,

					dp->i_mount->m_ddev_targp,

					dblkno, dblkcnt, XBF_LOCK);

			if (!bp)

				return ENOMEM;

			xfs_trans_binval(*trans, bp);

			/*

			 * Roll to next transaction.