Merge branch 'master' into next

5 files changed, 115 insertions, 20 deletions

diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index d71dc44e21e..aa1016bb913 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -34,6 +34,12 @@
 #include <linux/backing-dev.h>
 #include <linux/freezer.h>
 
+#include "xfs_sb.h"
+#include "xfs_inum.h"
+#include "xfs_ag.h"
+#include "xfs_dmapi.h"
+#include "xfs_mount.h"
+
 static kmem_zone_t *xfs_buf_zone;
 STATIC int xfsbufd(void *);
 STATIC int xfsbufd_wakeup(int, gfp_t);
@@ -166,6 +172,75 @@ test_page_region(
 }
 
 /*
+ *	Mapping of multi-page buffers into contiguous virtual space
+ */
+
+typedef struct a_list {
+	void		*vm_addr;
+	struct a_list	*next;
+} a_list_t;
+
+static a_list_t		*as_free_head;
+static int		as_list_len;
+static DEFINE_SPINLOCK(as_lock);
+
+/*
+ *	Try to batch vunmaps because they are costly.
+ */
+STATIC void
+free_address(
+	void		*addr)
+{
+	a_list_t	*aentry;
+
+#ifdef CONFIG_XEN
+	/*
+	 * Xen needs to be able to make sure it can get an exclusive
+	 * RO mapping of pages it wants to turn into a pagetable.  If
+	 * a newly allocated page is also still being vmap()ed by xfs,
+	 * it will cause pagetable construction to fail.  This is a
+	 * quick workaround to always eagerly unmap pages so that Xen
+	 * is happy.
+	 */
+	vunmap(addr);
+	return;
+#endif
+
+	aentry = kmalloc(sizeof(a_list_t), GFP_NOWAIT);
+	if (likely(aentry)) {
+		spin_lock(&as_lock);
+		aentry->next = as_free_head;
+		aentry->vm_addr = addr;
+		as_free_head = aentry;
+		as_list_len++;
+		spin_unlock(&as_lock);
+	} else {
+		vunmap(addr);
+	}
+}
+
+STATIC void
+purge_addresses(void)
+{
+	a_list_t	*aentry, *old;
+
+	if (as_free_head == NULL)
+		return;
+
+	spin_lock(&as_lock);
+	aentry = as_free_head;
+	as_free_head = NULL;
+	as_list_len = 0;
+	spin_unlock(&as_lock);
+
+	while ((old = aentry) != NULL) {
+		vunmap(aentry->vm_addr);
+		aentry = aentry->next;
+		kfree(old);
+	}
+}
+
+/*
  *	Internal xfs_buf_t object manipulation
  */
 
@@ -264,7 +339,7 @@ xfs_buf_free(
 		uint		i;
 
 		if ((bp->b_flags & XBF_MAPPED) && (bp->b_page_count > 1))
-                       vm_unmap_ram(bp->b_addr - bp->b_offset, bp->b_page_count);
+			free_address(bp->b_addr - bp->b_offset);
 
 		for (i = 0; i < bp->b_page_count; i++) {
 			struct page	*page = bp->b_pages[i];
@@ -386,8 +461,10 @@ _xfs_buf_map_pages(
 		bp->b_addr = page_address(bp->b_pages[0]) + bp->b_offset;
 		bp->b_flags |= XBF_MAPPED;
 	} else if (flags & XBF_MAPPED) {
-               bp->b_addr = vm_map_ram(bp->b_pages, bp->b_page_count,
-                                       -1, PAGE_KERNEL);
+		if (as_list_len > 64)
+			purge_addresses();
+		bp->b_addr = vmap(bp->b_pages, bp->b_page_count,
+					VM_MAP, PAGE_KERNEL);
 		if (unlikely(bp->b_addr == NULL))
 			return -ENOMEM;
 		bp->b_addr += bp->b_offset;
@@ -1364,10 +1441,12 @@ xfs_unregister_buftarg(
 
 void
 xfs_free_buftarg(
-	xfs_buftarg_t		*btp)
+	struct xfs_mount	*mp,
+	struct xfs_buftarg	*btp)
 {
 	xfs_flush_buftarg(btp, 1);
-	xfs_blkdev_issue_flush(btp);
+	if (mp->m_flags & XFS_MOUNT_BARRIER)
+		xfs_blkdev_issue_flush(btp);
 	xfs_free_bufhash(btp);
 	iput(btp->bt_mapping->host);
 
@@ -1672,6 +1751,8 @@ xfsbufd(
 			count++;
 		}
 
+		if (as_list_len > 0)
+			purge_addresses();
 		if (count)
 			blk_run_address_space(target->bt_mapping);
 
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index 288ae7c4c80..9b4d666ad31 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -413,7 +413,7 @@ static inline int XFS_bwrite(xfs_buf_t *bp)
  *	Handling of buftargs.
  */
 extern xfs_buftarg_t *xfs_alloc_buftarg(struct block_device *, int);
-extern void xfs_free_buftarg(xfs_buftarg_t *);
+extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *);
 extern void xfs_wait_buftarg(xfs_buftarg_t *);
 extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int);
 extern int xfs_flush_buftarg(xfs_buftarg_t *, int);
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index c71e226da7f..32ae5028e96 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -734,15 +734,15 @@ xfs_close_devices(
 {
 	if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
 		struct block_device *logdev = mp->m_logdev_targp->bt_bdev;
-		xfs_free_buftarg(mp->m_logdev_targp);
+		xfs_free_buftarg(mp, mp->m_logdev_targp);
 		xfs_blkdev_put(logdev);
 	}
 	if (mp->m_rtdev_targp) {
 		struct block_device *rtdev = mp->m_rtdev_targp->bt_bdev;
-		xfs_free_buftarg(mp->m_rtdev_targp);
+		xfs_free_buftarg(mp, mp->m_rtdev_targp);
 		xfs_blkdev_put(rtdev);
 	}
-	xfs_free_buftarg(mp->m_ddev_targp);
+	xfs_free_buftarg(mp, mp->m_ddev_targp);
 }
 
 /*
@@ -811,9 +811,9 @@ xfs_open_devices(
 
  out_free_rtdev_targ:
 	if (mp->m_rtdev_targp)
-		xfs_free_buftarg(mp->m_rtdev_targp);
+		xfs_free_buftarg(mp, mp->m_rtdev_targp);
  out_free_ddev_targ:
-	xfs_free_buftarg(mp->m_ddev_targp);
+	xfs_free_buftarg(mp, mp->m_ddev_targp);
  out_close_rtdev:
 	if (rtdev)
 		xfs_blkdev_put(rtdev);
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index e2fb6210d4c..478e587087f 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -246,9 +246,6 @@ xfs_iget_cache_miss(
 		goto out_destroy;
 	}
 
-	if (lock_flags)
-		xfs_ilock(ip, lock_flags);
-
 	/*
 	 * Preload the radix tree so we can insert safely under the
 	 * write spinlock. Note that we cannot sleep inside the preload
@@ -256,7 +253,16 @@ xfs_iget_cache_miss(
 	 */
 	if (radix_tree_preload(GFP_KERNEL)) {
 		error = EAGAIN;
-		goto out_unlock;
+		goto out_destroy;
+	}
+
+	/*
+	 * Because the inode hasn't been added to the radix-tree yet it can't
+	 * be found by another thread, so we can do the non-sleeping lock here.
+	 */
+	if (lock_flags) {
+		if (!xfs_ilock_nowait(ip, lock_flags))
+			BUG();
 	}
 
 	mask = ~(((XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog)) - 1);
@@ -284,7 +290,6 @@ xfs_iget_cache_miss(
 out_preload_end:
 	write_unlock(&pag->pag_ici_lock);
 	radix_tree_preload_end();
-out_unlock:
 	if (lock_flags)
 		xfs_iunlock(ip, lock_flags);
 out_destroy:
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index b1047de2fff..61af610d79b 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -1455,10 +1455,19 @@ xlog_recover_add_to_trans(
 	item = item->ri_prev;
 
 	if (item->ri_total == 0) {		/* first region to be added */
-		item->ri_total	= in_f->ilf_size;
-		ASSERT(item->ri_total <= XLOG_MAX_REGIONS_IN_ITEM);
-		item->ri_buf = kmem_zalloc((item->ri_total *
-					    sizeof(xfs_log_iovec_t)), KM_SLEEP);
+		if (in_f->ilf_size == 0 ||
+		    in_f->ilf_size > XLOG_MAX_REGIONS_IN_ITEM) {
+			xlog_warn(
+	"XFS: bad number of regions (%d) in inode log format",
+				  in_f->ilf_size);
+			ASSERT(0);
+			return XFS_ERROR(EIO);
+		}
+
+		item->ri_total = in_f->ilf_size;
+		item->ri_buf =
+			kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t),
+				    KM_SLEEP);
 	}
 	ASSERT(item->ri_total > item->ri_cnt);
 	/* Description region is ri_buf[0] */