Merge git://oss.sgi.com:8090/xfs/xfs-2.6

* git://oss.sgi.com:8090/xfs/xfs-2.6: (49 commits)
  [XFS] Remove v1 dir trace macro - missed in a past commit.
  [XFS] 955947: Infinite loop in xfs_bulkstat() on formatter() error
  [XFS] pv 956241, author: nathans, rv: vapo - make ino validation checks
  [XFS] pv 956240, author: nathans, rv: vapo - Minor fixes in
  [XFS] Really fix use after free in xfs_iunpin.
  [XFS] Collapse sv_init and init_sv into just the one interface.
  [XFS] standardize on one sema init macro
  [XFS] Reduce endian flipping in alloc_btree, same as was done for
  [XFS] Minor cleanup from dio locking fix, remove an extra conditional.
  [XFS] Fix kmem_zalloc_greedy warnings on 64 bit platforms.
  [XFS] pv 955157, rv bnaujok - break the loop on EFAULT formatter() error
  [XFS] pv 955157, rv bnaujok - break the loop on formatter() error
  [XFS] Fixes the leak in reservation space because we weren't ungranting
  [XFS] Add lock annotations to xfs_trans_update_ail and
  [XFS] Fix a porting botch on the realtime subvol growfs code path.
  [XFS] Minor code rearranging and cleanup to prevent some coverity false
  [XFS] Remove a no-longer-correct debug assert from dio completion
  [XFS] Add a greedy allocation interface, allocating within a min/max size
  [XFS] Improve error handling for the zero-fsblock extent detection code.
  [XFS] Be more defensive with page flags (error/private) for metadata
  ...

fs/xfs/Makefile-linux-2.6

@@ -30,7 +30,6 @@ ifeq ($(CONFIG_XFS_TRACE),y)
 	EXTRA_CFLAGS += -DXFS_BLI_TRACE
 	EXTRA_CFLAGS += -DXFS_BMAP_TRACE
 	EXTRA_CFLAGS += -DXFS_BMBT_TRACE
-	EXTRA_CFLAGS += -DXFS_DIR_TRACE
 	EXTRA_CFLAGS += -DXFS_DIR2_TRACE
 	EXTRA_CFLAGS += -DXFS_DQUOT_TRACE
 	EXTRA_CFLAGS += -DXFS_ILOCK_TRACE

fs/xfs/linux-2.6/kmem.c

@@ -34,6 +34,14 @@ kmem_alloc(size_t size, unsigned int __nocast flags)
 	gfp_t	lflags = kmem_flags_convert(flags);
 	void	*ptr;
 
+#ifdef DEBUG
+	if (unlikely(!(flags & KM_LARGE) && (size > PAGE_SIZE))) {
+		printk(KERN_WARNING "Large %s attempt, size=%ld\n",
+			__FUNCTION__, (long)size);
+		dump_stack();
+	}
+#endif
+
 	do {
 		if (size < MAX_SLAB_SIZE || retries > MAX_VMALLOCS)
 			ptr = kmalloc(size, lflags);
@@ -60,6 +68,27 @@ kmem_zalloc(size_t size, unsigned int __nocast flags)
 	return ptr;
 }
 
+void *
+kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize,
+		   unsigned int __nocast flags)
+{
+	void		*ptr;
+	size_t		kmsize = maxsize;
+	unsigned int	kmflags = (flags & ~KM_SLEEP) | KM_NOSLEEP;
+
+	while (!(ptr = kmem_zalloc(kmsize, kmflags))) {
+		if ((kmsize <= minsize) && (flags & KM_NOSLEEP))
+			break;
+		if ((kmsize >>= 1) <= minsize) {
+			kmsize = minsize;
+			kmflags = flags;
+		}
+	}
+	if (ptr)
+		*size = kmsize;
+	return ptr;
+}
+
 void
 kmem_free(void *ptr, size_t size)
 {

fs/xfs/linux-2.6/kmem.h

@@ -30,6 +30,7 @@
 #define KM_NOSLEEP	0x0002u
 #define KM_NOFS		0x0004u
 #define KM_MAYFAIL	0x0008u
+#define KM_LARGE	0x0010u
 
 /*
  * We use a special process flag to avoid recursive callbacks into
@@ -41,7 +42,7 @@ kmem_flags_convert(unsigned int __nocast flags)
 {
 	gfp_t	lflags;
 
-	BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL));
+	BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL|KM_LARGE));
 
 	if (flags & KM_NOSLEEP) {
 		lflags = GFP_ATOMIC | __GFP_NOWARN;
@@ -54,8 +55,9 @@ kmem_flags_convert(unsigned int __nocast flags)
 }
 
 extern void *kmem_alloc(size_t, unsigned int __nocast);
-extern void *kmem_realloc(void *, size_t, size_t, unsigned int __nocast);
 extern void *kmem_zalloc(size_t, unsigned int __nocast);
+extern void *kmem_zalloc_greedy(size_t *, size_t, size_t, unsigned int __nocast);
+extern void *kmem_realloc(void *, size_t, size_t, unsigned int __nocast);
 extern void  kmem_free(void *, size_t);
 
 /*

fs/xfs/linux-2.6/sema.h

@@ -29,8 +29,6 @@
 
 typedef struct semaphore sema_t;
 
-#define init_sema(sp, val, c, d)	sema_init(sp, val)
-#define initsema(sp, val)		sema_init(sp, val)
 #define initnsema(sp, val, name)	sema_init(sp, val)
 #define psema(sp, b)			down(sp)
 #define vsema(sp)			up(sp)

fs/xfs/linux-2.6/sv.h

@@ -53,8 +53,6 @@ static inline void _sv_wait(sv_t *sv, spinlock_t *lock, int state,
 	remove_wait_queue(&sv->waiters, &wait);
 }
 
-#define init_sv(sv,type,name,flag) \
-	init_waitqueue_head(&(sv)->waiters)
 #define sv_init(sv,flag,name) \
 	init_waitqueue_head(&(sv)->waiters)
 #define sv_destroy(sv) \

fs/xfs/linux-2.6/xfs_aops.c

@@ -71,7 +71,7 @@ xfs_page_trace(
 	int		tag,
 	struct inode	*inode,
 	struct page	*page,
-	int		mask)
+	unsigned long	pgoff)
 {
 	xfs_inode_t	*ip;
 	bhv_vnode_t	*vp = vn_from_inode(inode);
@@ -91,7 +91,7 @@ xfs_page_trace(
 		(void *)ip,
 		(void *)inode,
 		(void *)page,
-		(void *)((unsigned long)mask),
+		(void *)pgoff,
 		(void *)((unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff)),
 		(void *)((unsigned long)(ip->i_d.di_size & 0xffffffff)),
 		(void *)((unsigned long)((isize >> 32) & 0xffffffff)),
@@ -105,7 +105,7 @@ xfs_page_trace(
 		(void *)NULL);
 }
 #else
-#define xfs_page_trace(tag, inode, page, mask)
+#define xfs_page_trace(tag, inode, page, pgoff)
 #endif
 
 /*
@@ -1197,7 +1197,7 @@ xfs_vm_releasepage(
 		.nr_to_write = 1,
 	};
 
-	xfs_page_trace(XFS_RELEASEPAGE_ENTER, inode, page, gfp_mask);
+	xfs_page_trace(XFS_RELEASEPAGE_ENTER, inode, page, 0);
 
 	if (!page_has_buffers(page))
 		return 0;
@@ -1356,7 +1356,6 @@ xfs_end_io_direct(
 		ioend->io_size = size;
 		xfs_finish_ioend(ioend);
 	} else {
-		ASSERT(size >= 0);
 		xfs_destroy_ioend(ioend);
 	}
 

fs/xfs/linux-2.6/xfs_buf.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * Copyright (c) 2000-2006 Silicon Graphics, Inc.
  * All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or
@@ -318,8 +318,12 @@ xfs_buf_free(
 		if ((bp->b_flags & XBF_MAPPED) && (bp->b_page_count > 1))
 			free_address(bp->b_addr - bp->b_offset);
 
-		for (i = 0; i < bp->b_page_count; i++)
-			page_cache_release(bp->b_pages[i]);
+		for (i = 0; i < bp->b_page_count; i++) {
+			struct page	*page = bp->b_pages[i];
+
+			ASSERT(!PagePrivate(page));
+			page_cache_release(page);
+		}
 		_xfs_buf_free_pages(bp);
 	} else if (bp->b_flags & _XBF_KMEM_ALLOC) {
 		 /*
@@ -400,6 +404,7 @@ _xfs_buf_lookup_pages(
 		nbytes = min_t(size_t, size, PAGE_CACHE_SIZE - offset);
 		size -= nbytes;
 
+		ASSERT(!PagePrivate(page));
 		if (!PageUptodate(page)) {
 			page_count--;
 			if (blocksize >= PAGE_CACHE_SIZE) {
@@ -768,7 +773,7 @@ xfs_buf_get_noaddr(
 	_xfs_buf_initialize(bp, target, 0, len, 0);
 
  try_again:
-	data = kmem_alloc(malloc_len, KM_SLEEP | KM_MAYFAIL);
+	data = kmem_alloc(malloc_len, KM_SLEEP | KM_MAYFAIL | KM_LARGE);
 	if (unlikely(data == NULL))
 		goto fail_free_buf;
 
@@ -1117,10 +1122,10 @@ xfs_buf_bio_end_io(
 	do {
 		struct page	*page = bvec->bv_page;
 
+		ASSERT(!PagePrivate(page));
 		if (unlikely(bp->b_error)) {
 			if (bp->b_flags & XBF_READ)
 				ClearPageUptodate(page);
-			SetPageError(page);
 		} else if (blocksize >= PAGE_CACHE_SIZE) {
 			SetPageUptodate(page);
 		} else if (!PagePrivate(page) &&
@@ -1156,16 +1161,16 @@ _xfs_buf_ioapply(
 	total_nr_pages = bp->b_page_count;
 	map_i = 0;
 
-	if (bp->b_flags & _XBF_RUN_QUEUES) {
-		bp->b_flags &= ~_XBF_RUN_QUEUES;
-		rw = (bp->b_flags & XBF_READ) ? READ_SYNC : WRITE_SYNC;
-	} else {
-		rw = (bp->b_flags & XBF_READ) ? READ : WRITE;
-	}
-
 	if (bp->b_flags & XBF_ORDERED) {
 		ASSERT(!(bp->b_flags & XBF_READ));
 		rw = WRITE_BARRIER;
+	} else if (bp->b_flags & _XBF_RUN_QUEUES) {
+		ASSERT(!(bp->b_flags & XBF_READ_AHEAD));
+		bp->b_flags &= ~_XBF_RUN_QUEUES;
+		rw = (bp->b_flags & XBF_WRITE) ? WRITE_SYNC : READ_SYNC;
+	} else {
+		rw = (bp->b_flags & XBF_WRITE) ? WRITE :
+		     (bp->b_flags & XBF_READ_AHEAD) ? READA : READ;
 	}
 
 	/* Special code path for reading a sub page size buffer in --
@@ -1681,6 +1686,7 @@ xfsbufd(
 	xfs_buf_t		*bp, *n;
 	struct list_head	*dwq = &target->bt_delwrite_queue;
 	spinlock_t		*dwlk = &target->bt_delwrite_lock;
+	int			count;
 
 	current->flags |= PF_MEMALLOC;
 
@@ -1696,6 +1702,7 @@ xfsbufd(
 		schedule_timeout_interruptible(
 			xfs_buf_timer_centisecs * msecs_to_jiffies(10));
 
+		count = 0;
 		age = xfs_buf_age_centisecs * msecs_to_jiffies(10);
 		spin_lock(dwlk);
 		list_for_each_entry_safe(bp, n, dwq, b_list) {
@@ -1711,9 +1718,11 @@ xfsbufd(
 					break;
 				}
 
-				bp->b_flags &= ~(XBF_DELWRI|_XBF_DELWRI_Q);
+				bp->b_flags &= ~(XBF_DELWRI|_XBF_DELWRI_Q|
+						 _XBF_RUN_QUEUES);
 				bp->b_flags |= XBF_WRITE;
-				list_move(&bp->b_list, &tmp);
+				list_move_tail(&bp->b_list, &tmp);
+				count++;
 			}
 		}
 		spin_unlock(dwlk);
@@ -1724,12 +1733,12 @@ xfsbufd(
 
 			list_del_init(&bp->b_list);
 			xfs_buf_iostrategy(bp);
-
-			blk_run_address_space(target->bt_mapping);
 		}
 
 		if (as_list_len > 0)
 			purge_addresses();
+		if (count)
+			blk_run_address_space(target->bt_mapping);
 
 		clear_bit(XBT_FORCE_FLUSH, &target->bt_flags);
 	} while (!kthread_should_stop());
@@ -1767,7 +1776,7 @@ xfs_flush_buftarg(
 			continue;
 		}
 
-		list_move(&bp->b_list, &tmp);
+		list_move_tail(&bp->b_list, &tmp);
 	}
 	spin_unlock(dwlk);
 
@@ -1776,7 +1785,7 @@ xfs_flush_buftarg(
 	 */
 	list_for_each_entry_safe(bp, n, &tmp, b_list) {
 		xfs_buf_lock(bp);
-		bp->b_flags &= ~(XBF_DELWRI|_XBF_DELWRI_Q);
+		bp->b_flags &= ~(XBF_DELWRI|_XBF_DELWRI_Q|_XBF_RUN_QUEUES);
 		bp->b_flags |= XBF_WRITE;
 		if (wait)
 			bp->b_flags &= ~XBF_ASYNC;
@@ -1786,6 +1795,9 @@ xfs_flush_buftarg(
 		xfs_buf_iostrategy(bp);
 	}
 
+	if (wait)
+		blk_run_address_space(target->bt_mapping);
+
 	/*
 	 * Remaining list items must be flushed before returning
 	 */
@@ -1797,9 +1809,6 @@ xfs_flush_buftarg(
 		xfs_buf_relse(bp);
 	}
 
-	if (wait)
-		blk_run_address_space(target->bt_mapping);
-
 	return pincount;
 }
 

fs/xfs/linux-2.6/xfs_buf.h

@@ -298,11 +298,6 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *);
 #define XFS_BUF_UNWRITE(bp)	((bp)->b_flags &= ~XBF_WRITE)
 #define XFS_BUF_ISWRITE(bp)	((bp)->b_flags & XBF_WRITE)
 
-#define XFS_BUF_ISUNINITIAL(bp)	(0)
-#define XFS_BUF_UNUNINITIAL(bp)	(0)
-
-#define XFS_BUF_BP_ISMAPPED(bp)	(1)
-
 #define XFS_BUF_IODONE_FUNC(bp)			((bp)->b_iodone)
 #define XFS_BUF_SET_IODONE_FUNC(bp, func)	((bp)->b_iodone = (func))
 #define XFS_BUF_CLR_IODONE_FUNC(bp)		((bp)->b_iodone = NULL)
@@ -393,8 +388,6 @@ static inline int XFS_bwrite(xfs_buf_t *bp)
 	return error;
 }
 
-#define XFS_bdwrite(bp)		xfs_buf_iostart(bp, XBF_DELWRI | XBF_ASYNC)
-
 static inline int xfs_bdwrite(void *mp, xfs_buf_t *bp)
 {
 	bp->b_strat = xfs_bdstrat_cb;

fs/xfs/linux-2.6/xfs_globals.c

@@ -34,7 +34,7 @@ xfs_param_t xfs_params = {
 	.restrict_chown	= {	0,		1,		1	},
 	.sgid_inherit	= {	0,		0,		1	},
 	.symlink_mode	= {	0,		0,		1	},
-	.panic_mask	= {	0,		0,		127	},
+	.panic_mask	= {	0,		0,		255	},
 	.error_level	= {	0,		3,		11	},
 	.syncd_timer	= {	1*100,		30*100,		7200*100},
 	.stats_clear	= {	0,		0,		1	},

fs/xfs/linux-2.6/xfs_ioctl.c

@@ -653,7 +653,7 @@ xfs_attrmulti_by_handle(
 STATIC int
 xfs_ioc_space(
 	bhv_desc_t		*bdp,
-	bhv_vnode_t		*vp,
+	struct inode		*inode,
 	struct file		*filp,
 	int			flags,
 	unsigned int		cmd,
@@ -735,7 +735,7 @@ xfs_ioctl(
 		    !capable(CAP_SYS_ADMIN))
 			return -EPERM;
 
-		return xfs_ioc_space(bdp, vp, filp, ioflags, cmd, arg);
+		return xfs_ioc_space(bdp, inode, filp, ioflags, cmd, arg);
 
 	case XFS_IOC_DIOINFO: {
 		struct dioattr	da;
@@ -763,6 +763,8 @@ xfs_ioctl(
 		return xfs_ioc_fsgeometry(mp, arg);
 
 	case XFS_IOC_GETVERSION:
+		return put_user(inode->i_generation, (int __user *)arg);
+
 	case XFS_IOC_GETXFLAGS:
 	case XFS_IOC_SETXFLAGS:
 	case XFS_IOC_FSGETXATTR:
@@ -957,7 +959,7 @@ xfs_ioctl(
 STATIC int
 xfs_ioc_space(
 	bhv_desc_t		*bdp,
-	bhv_vnode_t		*vp,
+	struct inode		*inode,
 	struct file		*filp,
 	int			ioflags,
 	unsigned int		cmd,
@@ -967,13 +969,13 @@ xfs_ioc_space(
 	int			attr_flags = 0;
 	int			error;
 
-	if (vp->v_inode.i_flags & (S_IMMUTABLE|S_APPEND))
+	if (inode->i_flags & (S_IMMUTABLE|S_APPEND))
 		return -XFS_ERROR(EPERM);
 
 	if (!(filp->f_mode & FMODE_WRITE))
 		return -XFS_ERROR(EBADF);
 
-	if (!VN_ISREG(vp))
+	if (!S_ISREG(inode->i_mode))
 		return -XFS_ERROR(EINVAL);
 
 	if (copy_from_user(&bf, arg, sizeof(bf)))
@@ -1264,13 +1266,6 @@ xfs_ioc_xattr(
 		break;
 	}
 
-	case XFS_IOC_GETVERSION: {
-		flags = vn_to_inode(vp)->i_generation;
-		if (copy_to_user(arg, &flags, sizeof(flags)))
-			error = -EFAULT;
-		break;
-	}
-
 	default:
 		error = -ENOTTY;
 		break;