Merge with /pub/scm/linux/kernel/git/torvalds/linux-2.6.git

102 files changed, 5722 insertions, 2172 deletions

diff --git a/fs/Kconfig b/fs/Kconfig
index 48f5422cb19a..01a295232f75 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -810,7 +810,7 @@ config TMPFS
 
 config HUGETLBFS
         bool "HugeTLB file system support"
-        depends X86 || IA64 || PPC64 || SPARC64 || SUPERH || X86_64 || BROKEN
+        depends X86 || IA64 || PPC64 || SPARC64 || SUPERH || BROKEN
 
 config HUGETLB_PAGE
         def_bool HUGETLBFS
diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt
index 434c19d076ac..175b2e8177c1 100644
--- a/fs/Kconfig.binfmt
+++ b/fs/Kconfig.binfmt
@@ -57,7 +57,7 @@ config BINFMT_SHARED_FLAT
 
 config BINFMT_AOUT
         tristate "Kernel support for a.out and ECOFF binaries"
-        depends on (X86 && !X86_64) || ALPHA || ARM || M68K || SPARC32
+        depends on X86_32 || ALPHA || ARM || M68K || SPARC32
         ---help---
           A.out (Assembler.OUTput) is a set of formats for libraries and
           executables used in the earliest versions of UNIX.  Linux used
diff --git a/fs/afs/file.c b/fs/afs/file.c
index 23c125128024..4975c9c193dd 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -29,7 +29,7 @@ static int afs_file_release(struct inode *inode, struct file *file);
 
 static int afs_file_readpage(struct file *file, struct page *page);
 static int afs_file_invalidatepage(struct page *page, unsigned long offset);
-static int afs_file_releasepage(struct page *page, int gfp_flags);
+static int afs_file_releasepage(struct page *page, gfp_t gfp_flags);
 
 static ssize_t afs_file_write(struct file *file, const char __user *buf,
                               size_t size, loff_t *off);
@@ -279,7 +279,7 @@ static int afs_file_invalidatepage(struct page *page, unsigned long offset)
 /*
  * release a page and cleanup its private data
  */
-static int afs_file_releasepage(struct page *page, int gfp_flags)
+static int afs_file_releasepage(struct page *page, gfp_t gfp_flags)
 {
         struct cachefs_page *pageio;
 
@@ -291,8 +291,8 @@ static int afs_file_releasepage(struct page *page, int gfp_flags)
                 cachefs_uncache_page(vnode->cache, page);
 #endif
 
-                pageio = (struct cachefs_page *) page->private;
-                page->private = 0;
+                pageio = (struct cachefs_page *) page_private(page);
+                set_page_private(page, 0);
                 ClearPagePrivate(page);
 
                 if (pageio)
diff --git a/fs/attr.c b/fs/attr.c
index b1796fb9e524..67bcd9b14ea5 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -117,9 +117,6 @@ int notify_change(struct dentry * dentry, struct iattr * attr)
         struct timespec now;
         unsigned int ia_valid = attr->ia_valid;
 
-        if (!inode)
-                BUG();
-
         mode = inode->i_mode;
         now = current_fs_time(inode->i_sb);
 
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index dd9baabaf016..72011826f0cb 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -318,7 +318,6 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
         current->mm->free_area_cache = current->mm->mmap_base;
         current->mm->cached_hole_size = 0;
 
-        set_mm_counter(current->mm, rss, 0);
         current->mm->mmap = NULL;
         compute_creds(bprm);
         current->flags &= ~PF_FORKNOEXEC;
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index d4b15576e584..6fa6adc40972 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -773,7 +773,6 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 
         /* Do this so that we can load the interpreter, if need be.  We will
            change some of these later */
-        set_mm_counter(current->mm, rss, 0);
         current->mm->free_area_cache = current->mm->mmap_base;
         current->mm->cached_hole_size = 0;
         retval = setup_arg_pages(bprm, randomize_stack_top(STACK_TOP),
@@ -1503,9 +1502,7 @@ static int elf_core_dump(long signr, struct pt_regs * regs, struct file * file)
         fill_psinfo(psinfo, current->group_leader, current->mm);
         fill_note(notes +1, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo);
         
-        fill_note(notes +2, "CORE", NT_TASKSTRUCT, sizeof(*current), current);
-  
-        numnote = 3;
+        numnote = 2;
 
         auxv = (elf_addr_t *) current->mm->saved_auxv;
 
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index 134c9c0d1f54..dda87c4c82a3 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -294,14 +294,7 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs
                                   &interp_params,
                                   &current->mm->start_stack,
                                   &current->mm->start_brk);
-#endif
-
-        /* do this so that we can load the interpreter, if need be
-         * - we will change some of these later
-         */
-        set_mm_counter(current->mm, rss, 0);
 
-#ifdef CONFIG_MMU
         retval = setup_arg_pages(bprm, current->mm->start_stack, executable_stack);
         if (retval < 0) {
                 send_sig(SIGKILL, current, 0);
diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c
index 7974efa107bc..9d6625829b99 100644
--- a/fs/binfmt_flat.c
+++ b/fs/binfmt_flat.c
@@ -650,7 +650,6 @@ static int load_flat_file(struct linux_binprm * bprm,
                 current->mm->start_brk = datapos + data_len + bss_len;
                 current->mm->brk = (current->mm->start_brk + 3) & ~3;
                 current->mm->context.end_brk = memp + ksize((void *) memp) - stack_len;
-                set_mm_counter(current->mm, rss, 0);
         }
 
         if (flags & FLAT_FLAG_KTRACE)
diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c
index 227a2682d2bf..00a91dc25d16 100644
--- a/fs/binfmt_som.c
+++ b/fs/binfmt_som.c
@@ -259,7 +259,6 @@ load_som_binary(struct linux_binprm * bprm, struct pt_regs * regs)
         create_som_tables(bprm);
 
         current->mm->start_stack = bprm->p;
-        set_mm_counter(current->mm, rss, 0);
 
 #if 0
         printk("(start_brk) %08lx\n" , (unsigned long) current->mm->start_brk);
diff --git a/fs/bio.c b/fs/bio.c
index 7d81a93afd48..460554b07ff9 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -778,7 +778,7 @@ static int bio_map_kern_endio(struct bio *bio, unsigned int bytes_done, int err)
 
 
 static struct bio *__bio_map_kern(request_queue_t *q, void *data,
-                                  unsigned int len, unsigned int gfp_mask)
+                                  unsigned int len, gfp_t gfp_mask)
 {
         unsigned long kaddr = (unsigned long)data;
         unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
@@ -825,7 +825,7 @@ static struct bio *__bio_map_kern(request_queue_t *q, void *data,
  *      device. Returns an error pointer in case of error.
  */
 struct bio *bio_map_kern(request_queue_t *q, void *data, unsigned int len,
-                         unsigned int gfp_mask)
+                         gfp_t gfp_mask)
 {
         struct bio *bio;
 
diff --git a/fs/buffer.c b/fs/buffer.c
index 1216c0d3c8ce..35fa34977e81 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -96,7 +96,7 @@ static void
 __clear_page_buffers(struct page *page)
 {
         ClearPagePrivate(page);
-        page->private = 0;
+        set_page_private(page, 0);
         page_cache_release(page);
 }
 
@@ -502,7 +502,7 @@ static void free_more_memory(void)
         yield();
 
         for_each_pgdat(pgdat) {
-                zones = pgdat->node_zonelists[GFP_NOFS&GFP_ZONEMASK].zones;
+                zones = pgdat->node_zonelists[gfp_zone(GFP_NOFS)].zones;
                 if (*zones)
                         try_to_free_pages(zones, GFP_NOFS);
         }
@@ -1478,8 +1478,10 @@ EXPORT_SYMBOL(__getblk);
 void __breadahead(struct block_device *bdev, sector_t block, int size)
 {
         struct buffer_head *bh = __getblk(bdev, block, size);
-        ll_rw_block(READA, 1, &bh);
-        brelse(bh);
+        if (likely(bh)) {
+                ll_rw_block(READA, 1, &bh);
+                brelse(bh);
+        }
 }
 EXPORT_SYMBOL(__breadahead);
 
@@ -1497,7 +1499,7 @@ __bread(struct block_device *bdev, sector_t block, int size)
 {
         struct buffer_head *bh = __getblk(bdev, block, size);
 
-        if (!buffer_uptodate(bh))
+        if (likely(bh) && !buffer_uptodate(bh))
                 bh = __bread_slow(bh);
         return bh;
 }
@@ -1571,7 +1573,7 @@ static inline void discard_buffer(struct buffer_head * bh)
  *
  * NOTE: @gfp_mask may go away, and this function may become non-blocking.
  */
-int try_to_release_page(struct page *page, int gfp_mask)
+int try_to_release_page(struct page *page, gfp_t gfp_mask)
 {
         struct address_space * const mapping = page->mapping;
 
@@ -1637,6 +1639,15 @@ out:
 }
 EXPORT_SYMBOL(block_invalidatepage);
 
+int do_invalidatepage(struct page *page, unsigned long offset)
+{
+        int (*invalidatepage)(struct page *, unsigned long);
+        invalidatepage = page->mapping->a_ops->invalidatepage;
+        if (invalidatepage == NULL)
+                invalidatepage = block_invalidatepage;
+        return (*invalidatepage)(page, offset);
+}
+
 /*
  * We attach and possibly dirty the buffers atomically wrt
  * __set_page_dirty_buffers() via private_lock.  try_to_free_buffers
@@ -2696,7 +2707,7 @@ int block_write_full_page(struct page *page, get_block_t *get_block,
                  * they may have been added in ext3_writepage().  Make them
                  * freeable here, so the page does not leak.
                  */
-                block_invalidatepage(page, 0);
+                do_invalidatepage(page, 0);
                 unlock_page(page);
                 return 0; /* don't care */
         }
diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c
index 3d1cce3653b8..6a3df88accfe 100644
--- a/fs/coda/psdev.c
+++ b/fs/coda/psdev.c
@@ -370,8 +370,8 @@ static int init_coda_psdev(void)
         }               
         devfs_mk_dir ("coda");
         for (i = 0; i < MAX_CODADEVS; i++) {
-                class_device_create(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR,i),
-                                NULL, "cfs%d", i);
+                class_device_create(coda_psdev_class, NULL,
+                                MKDEV(CODA_PSDEV_MAJOR,i), NULL, "cfs%d", i);
                 err = devfs_mk_cdev(MKDEV(CODA_PSDEV_MAJOR, i),
                                 S_IFCHR|S_IRUSR|S_IWUSR, "coda/%d", i);
                 if (err)
diff --git a/fs/compat.c b/fs/compat.c
index a719e158e002..8e71cdbecc7c 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -1490,7 +1490,6 @@ int compat_do_execve(char * filename,
                 /* execve success */
                 security_bprm_free(bprm);
                 acct_update_integrals(current);
-                update_mem_hiwater(current);
                 kfree(bprm);
                 return retval;
         }
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index e28a74203f3b..43dbcb0b21eb 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -3046,10 +3046,15 @@ HANDLE_IOCTL(RAW_GETBIND, raw_ioctl)
 /* Serial */
 HANDLE_IOCTL(TIOCGSERIAL, serial_struct_ioctl)
 HANDLE_IOCTL(TIOCSSERIAL, serial_struct_ioctl)
+#ifdef TIOCGLTC
+COMPATIBLE_IOCTL(TIOCGLTC)
+COMPATIBLE_IOCTL(TIOCSLTC)
+#endif
 /* Usbdevfs */
 HANDLE_IOCTL(USBDEVFS_CONTROL32, do_usbdevfs_control)
 HANDLE_IOCTL(USBDEVFS_BULK32, do_usbdevfs_bulk)
 HANDLE_IOCTL(USBDEVFS_DISCSIGNAL32, do_usbdevfs_discsignal)
+COMPATIBLE_IOCTL(USBDEVFS_IOCTL32)
 /* i2c */
 HANDLE_IOCTL(I2C_FUNCS, w_long)
 HANDLE_IOCTL(I2C_RDWR, do_i2c_rdwr_ioctl)
diff --git a/fs/dcache.c b/fs/dcache.c
index fb10386c59be..e90512ed35a4 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -689,7 +689,7 @@ void shrink_dcache_anon(struct hlist_head *head)
  *
  * In this case we return -1 to tell the caller that we baled.
  */
-static int shrink_dcache_memory(int nr, unsigned int gfp_mask)
+static int shrink_dcache_memory(int nr, gfp_t gfp_mask)
 {
         if (nr) {
                 if (!(gfp_mask & __GFP_FS))
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 0d06097bc995..3931e7f1e6bf 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -162,6 +162,7 @@ static int dio_refill_pages(struct dio *dio)
         up_read(&current->mm->mmap_sem);
 
         if (ret < 0 && dio->blocks_available && (dio->rw == WRITE)) {
+                struct page *page = ZERO_PAGE(dio->curr_user_address);
                 /*
                  * A memory fault, but the filesystem has some outstanding
                  * mapped blocks.  We need to use those blocks up to avoid
@@ -169,7 +170,8 @@ static int dio_refill_pages(struct dio *dio)
                  */
                 if (dio->page_errors == 0)
                         dio->page_errors = ret;
-                dio->pages[0] = ZERO_PAGE(dio->curr_user_address);
+                page_cache_get(page);
+                dio->pages[0] = page;
                 dio->head = 0;
                 dio->tail = 1;
                 ret = 0;
diff --git a/fs/dquot.c b/fs/dquot.c
index b9732335bcdc..ea7644227a65 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -500,7 +500,7 @@ static void prune_dqcache(int count)
  * more memory
  */
 
-static int shrink_dqcache_memory(int nr, unsigned int gfp_mask)
+static int shrink_dqcache_memory(int nr, gfp_t gfp_mask)
 {
         if (nr) {
                 spin_lock(&dq_list_lock);
@@ -662,7 +662,7 @@ static void add_dquot_ref(struct super_block *sb, int type)
 restart:
         file_list_lock();
         list_for_each(p, &sb->s_files) {
-                struct file *filp = list_entry(p, struct file, f_list);
+                struct file *filp = list_entry(p, struct file, f_u.fu_list);
                 struct inode *inode = filp->f_dentry->d_inode;
                 if (filp->f_mode & FMODE_WRITE && dqinit_needed(inode, type)) {
                         struct dentry *dentry = dget(filp->f_dentry);
diff --git a/fs/exec.c b/fs/exec.c
index a04a575ad433..10d493fea7ce 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -126,8 +126,7 @@ asmlinkage long sys_uselib(const char __user * library)
         struct nameidata nd;
         int error;
 
-        nd.intent.open.flags = FMODE_READ;
-        error = __user_walk(library, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd);
+        error = __user_path_lookup_open(library, LOOKUP_FOLLOW, &nd, FMODE_READ);
         if (error)
                 goto out;
 
@@ -139,7 +138,7 @@ asmlinkage long sys_uselib(const char __user * library)
         if (error)
                 goto exit;
 
-        file = dentry_open(nd.dentry, nd.mnt, O_RDONLY);
+        file = nameidata_to_filp(&nd, O_RDONLY);
         error = PTR_ERR(file);
         if (IS_ERR(file))
                 goto out;
@@ -167,6 +166,7 @@ asmlinkage long sys_uselib(const char __user * library)
 out:
         return error;
 exit:
+        release_open_intent(&nd);
         path_release(&nd);
         goto out;
 }
@@ -309,40 +309,36 @@ void install_arg_page(struct vm_area_struct *vma,
         pud_t * pud;
         pmd_t * pmd;
         pte_t * pte;
+        spinlock_t *ptl;
 
         if (unlikely(anon_vma_prepare(vma)))
-                goto out_sig;
+                goto out;
 
         flush_dcache_page(page);
         pgd = pgd_offset(mm, address);
-
-        spin_lock(&mm->page_table_lock);
         pud = pud_alloc(mm, pgd, address);
         if (!pud)
                 goto out;
         pmd = pmd_alloc(mm, pud, address);
         if (!pmd)
                 goto out;
-        pte = pte_alloc_map(mm, pmd, address);
+        pte = pte_alloc_map_lock(mm, pmd, address, &ptl);
         if (!pte)
                 goto out;
         if (!pte_none(*pte)) {
-                pte_unmap(pte);
+                pte_unmap_unlock(pte, ptl);
                 goto out;
         }
-        inc_mm_counter(mm, rss);
+        inc_mm_counter(mm, anon_rss);
         lru_cache_add_active(page);
         set_pte_at(mm, address, pte, pte_mkdirty(pte_mkwrite(mk_pte(
                                         page, vma->vm_page_prot))));
         page_add_anon_rmap(page, vma, address);
-        pte_unmap(pte);
-        spin_unlock(&mm->page_table_lock);
+        pte_unmap_unlock(pte, ptl);
 
         /* no need for flush_tlb */
         return;
 out:
-        spin_unlock(&mm->page_table_lock);
-out_sig:
         __free_page(page);
         force_sig(SIGKILL, current);
 }
@@ -490,8 +486,7 @@ struct file *open_exec(const char *name)
         int err;
         struct file *file;
 
-        nd.intent.open.flags = FMODE_READ;
-        err = path_lookup(name, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd);
+        err = path_lookup_open(name, LOOKUP_FOLLOW, &nd, FMODE_READ);
         file = ERR_PTR(err);
 
         if (!err) {
@@ -504,7 +499,7 @@ struct file *open_exec(const char *name)
                                 err = -EACCES;
                         file = ERR_PTR(err);
                         if (!err) {
-                                file = dentry_open(nd.dentry, nd.mnt, O_RDONLY);
+                                file = nameidata_to_filp(&nd, O_RDONLY);
                                 if (!IS_ERR(file)) {
                                         err = deny_write_access(file);
                                         if (err) {
@@ -516,6 +511,7 @@ out:
                                 return file;
                         }
                 }
+                release_open_intent(&nd);
                 path_release(&nd);
         }
         goto out;
@@ -634,10 +630,9 @@ static inline int de_thread(struct task_struct *tsk)
         /*
          * Account for the thread group leader hanging around:
          */
-        count = 2;
-        if (thread_group_leader(current))
-                count = 1;
-        else {
+        count = 1;
+        if (!thread_group_leader(current)) {
+                count = 2;
                 /*
                  * The SIGALRM timer survives the exec, but needs to point
                  * at us as the new group leader now.  We have a race with
@@ -646,8 +641,10 @@ static inline int de_thread(struct task_struct *tsk)
                  * before we can safely let the old group leader die.
                  */
                 sig->real_timer.data = (unsigned long)current;
+                spin_unlock_irq(lock);
                 if (del_timer_sync(&sig->real_timer))
                         add_timer(&sig->real_timer);
+                spin_lock_irq(lock);
         }
         while (atomic_read(&sig->count) > count) {
                 sig->group_exit_task = current;
@@ -659,7 +656,6 @@ static inline int de_thread(struct task_struct *tsk)
         }
         sig->group_exit_task = NULL;
         sig->notify_count = 0;
-        sig->real_timer.data = (unsigned long)current;
         spin_unlock_irq(lock);
 
         /*
@@ -1207,7 +1203,6 @@ int do_execve(char * filename,
                 /* execve success */
                 security_bprm_free(bprm);
                 acct_update_integrals(current);
-                update_mem_hiwater(current);
                 kfree(bprm);
                 return retval;
         }
@@ -1422,19 +1417,16 @@ static void zap_threads (struct mm_struct *mm)
 static void coredump_wait(struct mm_struct *mm)
 {
         DECLARE_COMPLETION(startup_done);
+        int core_waiters;
 
-        mm->core_waiters++; /* let other threads block */
         mm->core_startup_done = &startup_done;
 
-        /* give other threads a chance to run: */
-        yield();
-
         zap_threads(mm);
-        if (--mm->core_waiters) {
-                up_write(&mm->mmap_sem);
+        core_waiters = mm->core_waiters;
+        up_write(&mm->mmap_sem);
+
+        if (core_waiters)
                 wait_for_completion(&startup_done);
-        } else
-                up_write(&mm->mmap_sem);
         BUG_ON(mm->core_waiters);
 }
 
@@ -1468,11 +1460,21 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
                 current->fsuid = 0;     /* Dump root private */
         }
         mm->dumpable = 0;
-        init_completion(&mm->core_done);
+
+        retval = -EAGAIN;
         spin_lock_irq(&current->sighand->siglock);
-        current->signal->flags = SIGNAL_GROUP_EXIT;
-        current->signal->group_exit_code = exit_code;
+        if (!(current->signal->flags & SIGNAL_GROUP_EXIT)) {
+                current->signal->flags = SIGNAL_GROUP_EXIT;
+                current->signal->group_exit_code = exit_code;
+                retval = 0;
+        }
         spin_unlock_irq(&current->sighand->siglock);
+        if (retval) {
+                up_write(&mm->mmap_sem);
+                goto fail;
+        }
+
+        init_completion(&mm->core_done);
         coredump_wait(mm);
 
         /*
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index fdba4d1d3c60..e7d3f0522d01 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -440,6 +440,10 @@ static int ext2_alloc_branch(struct inode *inode,
                  * the pointer to new one, then send parent to disk.
                  */
                 bh = sb_getblk(inode->i_sb, parent);
+                if (!bh) {
+                        err = -EIO;
+                        break;
+                }
                 lock_buffer(bh);
                 memset(bh->b_data, 0, blocksize);
                 branch[n].bh = bh;
diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index 0213db4911a2..7992d21e0e09 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -20,6 +20,8 @@
 #include <linux/quotaops.h>
 #include <linux/buffer_head.h>
 
+#include "bitmap.h"
+
 /*
  * balloc.c contains the blocks allocation and deallocation routines
  */
@@ -1010,7 +1012,7 @@ retry:
  * allocation within the reservation window.
  *
  * This will avoid keeping on searching the reservation list again and
- * again when someboday is looking for a free block (without
+ * again when somebody is looking for a free block (without
  * reservation), and there are lots of free blocks, but they are all
  * being reserved.
  *
@@ -1416,12 +1418,12 @@ unsigned long ext3_count_free_blocks(struct super_block *sb)
         unsigned long bitmap_count, x;
         struct buffer_head *bitmap_bh = NULL;
 
-        lock_super(sb);
         es = EXT3_SB(sb)->s_es;
         desc_count = 0;
         bitmap_count = 0;
         gdp = NULL;
 
+        smp_rmb();
         for (i = 0; i < ngroups; i++) {
                 gdp = ext3_get_group_desc(sb, i, NULL);
                 if (!gdp)
@@ -1440,7 +1442,6 @@ unsigned long ext3_count_free_blocks(struct super_block *sb)
         brelse(bitmap_bh);
         printk("ext3_count_free_blocks: stored = %u, computed = %lu, %lu\n",
                le32_to_cpu(es->s_free_blocks_count), desc_count, bitmap_count);
-        unlock_super(sb);
         return bitmap_count;
 #else
         desc_count = 0;
diff --git a/fs/ext3/bitmap.c b/fs/ext3/bitmap.c
index 6c419b9ab0e8..5b4ba3e246e6 100644
--- a/fs/ext3/bitmap.c
+++ b/fs/ext3/bitmap.c
@@ -8,7 +8,7 @@
  */
 
 #include <linux/buffer_head.h>
-
+#include "bitmap.h"
 
 static int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0};
 
diff --git a/fs/ext3/bitmap.h b/fs/ext3/bitmap.h
new file mode 100644
index 000000000000..6ee503a6bb4e
--- /dev/null
+++ b/fs/ext3/bitmap.h
@@ -0,0 +1,8 @@
+/*  linux/fs/ext3/bitmap.c
+ *
+ * Copyright (C) 2005 Simtec Electronics
+ *      Ben Dooks <ben@simtec.co.uk>
+ *
+*/
+
+extern unsigned long ext3_count_free (struct buffer_head *, unsigned int );
diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 6549945f9ac1..df3f517c54ac 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -26,6 +26,7 @@
 
 #include <asm/byteorder.h>
 
+#include "bitmap.h"
 #include "xattr.h"
 #include "acl.h"
 
@@ -704,7 +705,6 @@ unsigned long ext3_count_free_inodes (struct super_block * sb)
         unsigned long bitmap_count, x;
         struct buffer_head *bitmap_bh = NULL;
 
-        lock_super (sb);
         es = EXT3_SB(sb)->s_es;
         desc_count = 0;
         bitmap_count = 0;
@@ -727,7 +727,6 @@ unsigned long ext3_count_free_inodes (struct super_block * sb)
         brelse(bitmap_bh);
         printk("ext3_count_free_inodes: stored = %u, computed = %lu, %lu\n",
                 le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
-        unlock_super(sb);
         return desc_count;
 #else
         desc_count = 0;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index b5177c90d6f1..5d9b00e28837 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -491,7 +491,7 @@ static unsigned long ext3_find_goal(struct inode *inode, long block,
  *      the same format as ext3_get_branch() would do. We are calling it after
  *      we had read the existing part of chain and partial points to the last
  *      triple of that (one with zero ->key). Upon the exit we have the same
- *      picture as after the successful ext3_get_block(), excpet that in one
+ *      picture as after the successful ext3_get_block(), except that in one
  *      place chain is disconnected - *branch->p is still zero (we did not
  *      set the last link), but branch->key contains the number that should
  *      be placed into *branch->p to fill that gap.
@@ -523,7 +523,6 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode,
                         if (!nr)
                                 break;
                         branch[n].key = cpu_to_le32(nr);
-                        keys = n+1;
 
                         /*
                          * Get buffer_head for parent block, zero it out
@@ -531,6 +530,9 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode,
                          * parent to disk.  
                          */
                         bh = sb_getblk(inode->i_sb, parent);
+                        if (!bh)
+                                break;
+                        keys = n+1;
                         branch[n].bh = bh;
                         lock_buffer(bh);
                         BUFFER_TRACE(bh, "call get_create_access");
@@ -864,6 +866,10 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode * inode,
         if (!*errp && buffer_mapped(&dummy)) {
                 struct buffer_head *bh;
                 bh = sb_getblk(inode->i_sb, dummy.b_blocknr);
+                if (!bh) {
+                        *errp = -EIO;
+                        goto err;
+                }
                 if (buffer_new(&dummy)) {
                         J_ASSERT(create != 0);
                         J_ASSERT(handle != 0);
@@ -896,6 +902,7 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode * inode,
                 }
                 return bh;
         }
+err:
         return NULL;
 }
 
@@ -1434,7 +1441,7 @@ static int ext3_invalidatepage(struct page *page, unsigned long offset)
         return journal_invalidatepage(journal, page, offset);
 }
 
-static int ext3_releasepage(struct page *page, int wait)
+static int ext3_releasepage(struct page *page, gfp_t wait)
 {
         journal_t *journal = EXT3_JOURNAL(page->mapping->host);
 
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 50378d8ff84b..b3c690a3b54a 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -36,6 +36,8 @@
 #include <linux/quotaops.h>
 #include <linux/buffer_head.h>
 #include <linux/smp_lock.h>
+
+#include "namei.h"
 #include "xattr.h"
 #include "acl.h"
 
diff --git a/fs/ext3/namei.h b/fs/ext3/namei.h
new file mode 100644
index 000000000000..f2ce2b0065c9
--- /dev/null
+++ b/fs/ext3/namei.h
@@ -0,0 +1,8 @@
+/*  linux/fs/ext3/namei.h
+ *
+ * Copyright (C) 2005 Simtec Electronics
+ *      Ben Dooks <ben@simtec.co.uk>
+ *
+*/
+
+extern struct dentry *ext3_get_parent(struct dentry *child);
diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c
index 57f79106267d..1be78b4b4de9 100644
--- a/fs/ext3/resize.c
+++ b/fs/ext3/resize.c
@@ -118,6 +118,8 @@ static struct buffer_head *bclean(handle_t *handle, struct super_block *sb,
         int err;
 
         bh = sb_getblk(sb, blk);
+        if (!bh)
+                return ERR_PTR(-EIO);
         if ((err = ext3_journal_get_write_access(handle, bh))) {
                 brelse(bh);
                 bh = ERR_PTR(err);
@@ -202,6 +204,10 @@ static int setup_new_group_blocks(struct super_block *sb,
                 ext3_debug("update backup group %#04lx (+%d)\n", block, bit);
 
                 gdb = sb_getblk(sb, block);
+                if (!gdb) {
+                        err = -EIO;
+                        goto exit_bh;
+                }
                 if ((err = ext3_journal_get_write_access(handle, gdb))) {
                         brelse(gdb);
                         goto exit_bh;
@@ -643,6 +649,10 @@ static void update_backups(struct super_block *sb,
                         break;
 
                 bh = sb_getblk(sb, group * bpg + blk_off);
+                if (!bh) {
+                        err = -EIO;
+                        break;
+                }
                 ext3_debug("update metadata backup %#04lx\n",
                           (unsigned long)bh->b_blocknr);
                 if ((err = ext3_journal_get_write_access(handle, bh)))
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 9e24ceb019fe..f594989ccb7a 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -36,9 +36,12 @@
 #include <linux/namei.h>
 #include <linux/quotaops.h>
 #include <linux/seq_file.h>
+
 #include <asm/uaccess.h>
+
 #include "xattr.h"
 #include "acl.h"
+#include "namei.h"
 
 static int ext3_load_journal(struct super_block *, struct ext3_super_block *);
 static int ext3_create_journal(struct super_block *, struct ext3_super_block *,
@@ -510,19 +513,11 @@ static void ext3_clear_inode(struct inode *inode)
         kfree(rsv);
 }
 
-static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
+static inline void ext3_show_quota_options(struct seq_file *seq, struct super_block *sb)
 {
-        struct super_block *sb = vfs->mnt_sb;
+#if defined(CONFIG_QUOTA)
         struct ext3_sb_info *sbi = EXT3_SB(sb);
 
-        if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA)
-                seq_puts(seq, ",data=journal");
-        else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA)
-                seq_puts(seq, ",data=ordered");
-        else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)
-                seq_puts(seq, ",data=writeback");
-
-#if defined(CONFIG_QUOTA)
         if (sbi->s_jquota_fmt)
                 seq_printf(seq, ",jqfmt=%s",
                 (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0");
@@ -539,6 +534,20 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
         if (sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA)
                 seq_puts(seq, ",grpquota");
 #endif
+}
+
+static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
+{
+        struct super_block *sb = vfs->mnt_sb;
+
+        if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA)
+                seq_puts(seq, ",data=journal");
+        else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA)
+                seq_puts(seq, ",data=ordered");
+        else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)
+                seq_puts(seq, ",data=writeback");
+
+        ext3_show_quota_options(seq, sb);
 
         return 0;
 }
@@ -609,7 +618,6 @@ static struct super_operations ext3_sops = {
 #endif
 };
 
-struct dentry *ext3_get_parent(struct dentry *child);
 static struct export_operations ext3_export_ops = {
         .get_parent = ext3_get_parent,
 };
diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c
index 269c7b92db9a..430de9f63be3 100644
--- a/fs/ext3/xattr.c
+++ b/fs/ext3/xattr.c
@@ -210,7 +210,7 @@ ext3_xattr_find_entry(struct ext3_xattr_entry **pentry, int name_index,
         return cmp ? -ENODATA : 0;
 }
 
-int
+static int
 ext3_xattr_block_get(struct inode *inode, int name_index, const char *name,
                      void *buffer, size_t buffer_size)
 {
@@ -354,7 +354,7 @@ ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry,
         return buffer_size - rest;
 }
 
-int
+static int
 ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size)
 {
         struct buffer_head *bh = NULL;
@@ -626,7 +626,7 @@ struct ext3_xattr_block_find {
         struct buffer_head *bh;
 };
 
-int
+static int
 ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i,
                       struct ext3_xattr_block_find *bs)
 {
@@ -859,7 +859,7 @@ struct ext3_xattr_ibody_find {
         struct ext3_iloc iloc;
 };
 
-int
+static int
 ext3_xattr_ibody_find(struct inode *inode, struct ext3_xattr_info *i,
                       struct ext3_xattr_ibody_find *is)
 {
diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index 895049b2ac9c..ba824964b9bb 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -222,6 +222,80 @@ fat_shortname2uni(struct nls_table *nls, unsigned char *buf, int buf_size,
         return len;
 }
 
+enum { PARSE_INVALID = 1, PARSE_NOT_LONGNAME, PARSE_EOF, };
+
+/**
+ * fat_parse_long - Parse extended directory entry.
+ *
+ * This function returns zero on success, negative value on error, or one of
+ * the following:
+ *
+ * %PARSE_INVALID - Directory entry is invalid.
+ * %PARSE_NOT_LONGNAME - Directory entry does not contain longname.
+ * %PARSE_EOF - Directory has no more entries.
+ */
+static int fat_parse_long(struct inode *dir, loff_t *pos,
+                          struct buffer_head **bh, struct msdos_dir_entry **de,
+                          wchar_t **unicode, unsigned char *nr_slots)
+{
+        struct msdos_dir_slot *ds;
+        unsigned char id, slot, slots, alias_checksum;
+
+        if (!*unicode) {
+                *unicode = (wchar_t *)__get_free_page(GFP_KERNEL);
+                if (!*unicode) {
+                        brelse(*bh);
+                        return -ENOMEM;
+                }
+        }
+parse_long:
+        slots = 0;
+        ds = (struct msdos_dir_slot *)*de;
+        id = ds->id;
+        if (!(id & 0x40))
+                return PARSE_INVALID;
+        slots = id & ~0x40;
+        if (slots > 20 || !slots)       /* ceil(256 * 2 / 26) */
+                return PARSE_INVALID;
+        *nr_slots = slots;
+        alias_checksum = ds->alias_checksum;
+
+        slot = slots;
+        while (1) {
+                int offset;
+
+                slot--;
+                offset = slot * 13;
+                fat16_towchar(*unicode + offset, ds->name0_4, 5);
+                fat16_towchar(*unicode + offset + 5, ds->name5_10, 6);
+                fat16_towchar(*unicode + offset + 11, ds->name11_12, 2);
+
+                if (ds->id & 0x40)
+                        (*unicode)[offset + 13] = 0;
+                if (fat_get_entry(dir, pos, bh, de) < 0)
+                        return PARSE_EOF;
+                if (slot == 0)
+                        break;
+                ds = (struct msdos_dir_slot *)*de;
+                if (ds->attr != ATTR_EXT)
+                        return PARSE_NOT_LONGNAME;
+                if ((ds->id & ~0x40) != slot)
+                        goto parse_long;
+                if (ds->alias_checksum != alias_checksum)
+                        goto parse_long;
+        }
+        if ((*de)->name[0] == DELETED_FLAG)
+                return PARSE_INVALID;
+        if ((*de)->attr == ATTR_EXT)
+                goto parse_long;
+        if (IS_FREE((*de)->name) || ((*de)->attr & ATTR_VOLUME))
+                return PARSE_INVALID;
+        if (fat_checksum((*de)->name) != alias_checksum)
+                *nr_slots = 0;
+
+        return 0;
+}
+
 /*
  * Return values: negative -> error, 0 -> not found, positive -> found,
  * value is the total amount of slots, including the shortname entry.
@@ -259,68 +333,16 @@ parse_record:
                 if (de->attr != ATTR_EXT && IS_FREE(de->name))
                         continue;
                 if (de->attr == ATTR_EXT) {
-                        struct msdos_dir_slot *ds;
-                        unsigned char id;
-                        unsigned char slot;
-                        unsigned char slots;
-                        unsigned char sum;
-                        unsigned char alias_checksum;
-
-                        if (!unicode) {
-                                unicode = (wchar_t *)
-                                        __get_free_page(GFP_KERNEL);
-                                if (!unicode) {
-                                        brelse(bh);
-                                        return -ENOMEM;
-                                }
-                        }
-parse_long:
-                        slots = 0;
-                        ds = (struct msdos_dir_slot *) de;
-                        id = ds->id;
-                        if (!(id & 0x40))
-                                continue;
-                        slots = id & ~0x40;
-                        if (slots > 20 || !slots)       /* ceil(256 * 2 / 26) */
-                                continue;
-                        nr_slots = slots;
-                        alias_checksum = ds->alias_checksum;
-
-                        slot = slots;
-                        while (1) {
-                                int offset;
-
-                                slot--;
-                                offset = slot * 13;
-                                fat16_towchar(unicode + offset, ds->name0_4, 5);
-                                fat16_towchar(unicode + offset + 5, ds->name5_10, 6);
-                                fat16_towchar(unicode + offset + 11, ds->name11_12, 2);
-
-                                if (ds->id & 0x40) {
-                                        unicode[offset + 13] = 0;
-                                }
-                                if (fat_get_entry(inode, &cpos, &bh, &de) < 0)
-                                        goto EODir;
-                                if (slot == 0)
-                                        break;
-                                ds = (struct msdos_dir_slot *) de;
-                                if (ds->attr !=  ATTR_EXT)
-                                        goto parse_record;
-                                if ((ds->id & ~0x40) != slot)
-                                        goto parse_long;
-                                if (ds->alias_checksum != alias_checksum)
-                                        goto parse_long;
-                        }
-                        if (de->name[0] == DELETED_FLAG)
-                                continue;
-                        if (de->attr ==  ATTR_EXT)
-                                goto parse_long;
-                        if (IS_FREE(de->name) || (de->attr & ATTR_VOLUME))
+                        int status = fat_parse_long(inode, &cpos, &bh, &de,
+                                                    &unicode, &nr_slots);
+                        if (status < 0)
+                                return status;
+                        else if (status == PARSE_INVALID)
                                 continue;
-                        for (sum = 0, i = 0; i < 11; i++)
-                                sum = (((sum&1)<<7)|((sum&0xfe)>>1)) + de->name[i];
-                        if (sum != alias_checksum)
-                                nr_slots = 0;
+                        else if (status == PARSE_NOT_LONGNAME)
+                                goto parse_record;
+                        else if (status == PARSE_EOF)
+                                goto EODir;
                 }
 
                 memcpy(work, de->name, sizeof(de->name));
@@ -408,8 +430,8 @@ struct fat_ioctl_filldir_callback {
         int short_len;
 };
 
-static int fat_readdirx(struct inode *inode, struct file *filp, void *dirent,
-                        filldir_t filldir, int short_only, int both)
+static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent,
+                         filldir_t filldir, int short_only, int both)
 {
         struct super_block *sb = inode->i_sb;
         struct msdos_sb_info *sbi = MSDOS_SB(sb);
@@ -458,9 +480,10 @@ static int fat_readdirx(struct inode *inode, struct file *filp, void *dirent,
 
         bh = NULL;
 GetNew:
-        long_slots = 0;
         if (fat_get_entry(inode, &cpos, &bh, &de) == -1)
                 goto EODir;
+parse_record:
+        long_slots = 0;
         /* Check for long filename entry */
         if (isvfat) {
                 if (de->name[0] == DELETED_FLAG)
@@ -475,69 +498,18 @@ GetNew:
         }
 
         if (isvfat && de->attr == ATTR_EXT) {
-                struct msdos_dir_slot *ds;
-                unsigned char id;
-                unsigned char slot;
-                unsigned char slots;
-                unsigned char sum;
-                unsigned char alias_checksum;
-
-                if (!unicode) {
-                        unicode = (wchar_t *)__get_free_page(GFP_KERNEL);
-                        if (!unicode) {
-                                filp->f_pos = cpos;
-                                brelse(bh);
-                                ret = -ENOMEM;
-                                goto out;
-                        }
-                }
-ParseLong:
-                slots = 0;
-                ds = (struct msdos_dir_slot *) de;
-                id = ds->id;
-                if (!(id & 0x40))
-                        goto RecEnd;
-                slots = id & ~0x40;
-                if (slots > 20 || !slots)       /* ceil(256 * 2 / 26) */
+                int status = fat_parse_long(inode, &cpos, &bh, &de,
+                                            &unicode, &long_slots);
+                if (status < 0) {
+                        filp->f_pos = cpos;
+                        ret = status;
+                        goto out;
+                } else if (status == PARSE_INVALID)
                         goto RecEnd;
-                long_slots = slots;
-                alias_checksum = ds->alias_checksum;
-
-                slot = slots;
-                while (1) {
-                        int offset;
-
-                        slot--;
-                        offset = slot * 13;
-                        fat16_towchar(unicode + offset, ds->name0_4, 5);
-                        fat16_towchar(unicode + offset + 5, ds->name5_10, 6);
-                        fat16_towchar(unicode + offset + 11, ds->name11_12, 2);
-
-                        if (ds->id & 0x40) {
-                                unicode[offset + 13] = 0;
-                        }
-                        if (fat_get_entry(inode, &cpos, &bh, &de) == -1)
-                                goto EODir;
-                        if (slot == 0)
-                                break;
-                        ds = (struct msdos_dir_slot *) de;
-                        if (ds->attr !=  ATTR_EXT)
-                                goto RecEnd;    /* XXX */
-                        if ((ds->id & ~0x40) != slot)
-                                goto ParseLong;
-                        if (ds->alias_checksum != alias_checksum)
-                                goto ParseLong;
-                }
-                if (de->name[0] == DELETED_FLAG)
-                        goto RecEnd;
-                if (de->attr ==  ATTR_EXT)
-                        goto ParseLong;
-                if (IS_FREE(de->name) || (de->attr & ATTR_VOLUME))
-                        goto RecEnd;
-                for (sum = 0, i = 0; i < 11; i++)
-                        sum = (((sum&1)<<7)|((sum&0xfe)>>1)) + de->name[i];
-                if (sum != alias_checksum)
-                        long_slots = 0;
+                else if (status == PARSE_NOT_LONGNAME)
+                        goto parse_record;
+                else if (status == PARSE_EOF)
+                        goto EODir;
         }
 
         if (sbi->options.dotsOK) {
@@ -671,7 +643,7 @@ out:
 static int fat_readdir(struct file *filp, void *dirent, filldir_t filldir)
 {
         struct inode *inode = filp->f_dentry->d_inode;
-        return fat_readdirx(inode, filp, dirent, filldir, 0, 0);
+        return __fat_readdir(inode, filp, dirent, filldir, 0, 0);
 }
 
 static int fat_ioctl_filldir(void *__buf, const char *name, int name_len,
@@ -760,8 +732,8 @@ static int fat_dir_ioctl(struct inode * inode, struct file * filp,
         down(&inode->i_sem);
         ret = -ENOENT;
         if (!IS_DEADDIR(inode)) {
-                ret = fat_readdirx(inode, filp, &buf, fat_ioctl_filldir,
-                                   short_only, both);
+                ret = __fat_readdir(inode, filp, &buf, fat_ioctl_filldir,
+                                    short_only, both);
         }
         up(&inode->i_sem);
         if (ret >= 0)
diff --git a/fs/file_table.c b/fs/file_table.c
index 86ec8ae985b4..4dc205546547 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -56,13 +56,13 @@ void filp_dtor(void * objp, struct kmem_cache_s *cachep, unsigned long dflags)
 
 static inline void file_free_rcu(struct rcu_head *head)
 {
-        struct file *f =  container_of(head, struct file, f_rcuhead);
+        struct file *f =  container_of(head, struct file, f_u.fu_rcuhead);
         kmem_cache_free(filp_cachep, f);
 }
 
 static inline void file_free(struct file *f)
 {
-        call_rcu(&f->f_rcuhead, file_free_rcu);
+        call_rcu(&f->f_u.fu_rcuhead, file_free_rcu);
 }
 
 /* Find an unused file structure and return a pointer to it.
@@ -95,7 +95,7 @@ struct file *get_empty_filp(void)
         f->f_gid = current->fsgid;
         rwlock_init(&f->f_owner.lock);
         /* f->f_version: 0 */
-        INIT_LIST_HEAD(&f->f_list);
+        INIT_LIST_HEAD(&f->f_u.fu_list);
         return f;
 
 over:
@@ -225,15 +225,15 @@ void file_move(struct file *file, struct list_head *list)
         if (!list)
                 return;
         file_list_lock();
-        list_move(&file->f_list, list);
+        list_move(&file->f_u.fu_list, list);
         file_list_unlock();
 }
 
 void file_kill(struct file *file)
 {
-        if (!list_empty(&file->f_list)) {
+        if (!list_empty(&file->f_u.fu_list)) {
                 file_list_lock();
-                list_del_init(&file->f_list);
+                list_del_init(&file->f_u.fu_list);
                 file_list_unlock();
         }
 }
@@ -245,7 +245,7 @@ int fs_may_remount_ro(struct super_block *sb)
         /* Check that no files are currently opened for writing. */
         file_list_lock();
         list_for_each(p, &sb->s_files) {
-                struct file *file = list_entry(p, struct file, f_list);
+                struct file *file = list_entry(p, struct file, f_u.fu_list);
                 struct inode *inode = file->f_dentry->d_inode;
 
                 /* File with pending delete? */
diff --git a/fs/filesystems.c b/fs/filesystems.c
index 44082bfdfec9..9f1072836c8e 100644
--- a/fs/filesystems.c
+++ b/fs/filesystems.c
@@ -12,6 +12,7 @@
 #include <linux/kmod.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/sched.h>        /* for 'current' */
 #include <asm/uaccess.h>
 
 /*
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index e94ab398b717..ffab4783ac64 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -230,7 +230,6 @@ __sync_single_inode(struct inode *inode, struct writeback_control *wbc)
                          * The inode is clean, unused
                          */
                         list_move(&inode->i_list, &inode_unused);
-                        inodes_stat.nr_unused++;
                 }
         }
         wake_up_inode(inode);
@@ -238,14 +237,20 @@ __sync_single_inode(struct inode *inode, struct writeback_control *wbc)
 }
 
 /*
- * Write out an inode's dirty pages.  Called under inode_lock.
+ * Write out an inode's dirty pages.  Called under inode_lock.  Either the
+ * caller has ref on the inode (either via __iget or via syscall against an fd)
+ * or the inode has I_WILL_FREE set (via generic_forget_inode)
  */
 static int
-__writeback_single_inode(struct inode *inode,
-                        struct writeback_control *wbc)
+__writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
 {
         wait_queue_head_t *wqh;
 
+        if (!atomic_read(&inode->i_count))
+                WARN_ON(!(inode->i_state & I_WILL_FREE));
+        else
+                WARN_ON(inode->i_state & I_WILL_FREE);
+
         if ((wbc->sync_mode != WB_SYNC_ALL) && (inode->i_state & I_LOCK)) {
                 list_move(&inode->i_list, &inode->i_sb->s_dirty);
                 return 0;
@@ -259,11 +264,9 @@ __writeback_single_inode(struct inode *inode,
 
                 wqh = bit_waitqueue(&inode->i_state, __I_LOCK);
                 do {
-                        __iget(inode);
                         spin_unlock(&inode_lock);
                         __wait_on_bit(wqh, &wq, inode_wait,
                                                         TASK_UNINTERRUPTIBLE);
-                        iput(inode);
                         spin_lock(&inode_lock);
                 } while (inode->i_state & I_LOCK);
         }
@@ -541,14 +544,15 @@ void sync_inodes(int wait)
 }
 
 /**
- *      write_inode_now -       write an inode to disk
- *      @inode: inode to write to disk
- *      @sync: whether the write should be synchronous or not
+ * write_inode_now      -       write an inode to disk
+ * @inode: inode to write to disk
+ * @sync: whether the write should be synchronous or not
+ *
+ * This function commits an inode to disk immediately if it is dirty. This is
+ * primarily needed by knfsd.
  *
- *      This function commits an inode to disk immediately if it is
- *      dirty. This is primarily needed by knfsd.
+ * The caller must either have a ref on the inode or must have set I_WILL_FREE.
  */
- 
 int write_inode_now(struct inode *inode, int sync)
 {
         int ret;
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index d4c869c6d01b..a6f90a6c754a 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -151,9 +151,9 @@ void fuse_release_background(struct fuse_req *req)
 /*
  * This function is called when a request is finished.  Either a reply
  * has arrived or it was interrupted (and not yet sent) or some error
- * occured during communication with userspace, or the device file was
- * closed.  It decreases the referece count for the request.  In case
- * of a background request the referece to the stored objects are
+ * occurred during communication with userspace, or the device file was
+ * closed.  It decreases the reference count for the request.  In case
+ * of a background request the reference to the stored objects are
  * released.  The requester thread is woken up (if still waiting), and
  * finally the request is either freed or put on the unused_list
  *
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 29f1e9f6e85c..70dba721acab 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -741,13 +741,14 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
         if (inode && S_ISDIR(inode->i_mode)) {
                 /* Don't allow creating an alias to a directory  */
                 struct dentry *alias = d_find_alias(inode);
-                if (alias && !(alias->d_flags & DCACHE_DISCONNECTED)) {
+                if (alias) {
                         dput(alias);
                         iput(inode);
                         return ERR_PTR(-EIO);
                 }
         }
-        return d_splice_alias(inode, entry);
+        d_add(entry, inode);
+        return NULL;
 }
 
 static int fuse_setxattr(struct dentry *entry, const char *name,
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 24d761518d86..5cb456f572c1 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -349,22 +349,22 @@ int fuse_fsync_common(struct file *file, struct dentry *de, int datasync,
                       int isdir);
 
 /**
- * Initialise file operations on a regular file
+ * Initialize file operations on a regular file
  */
 void fuse_init_file_inode(struct inode *inode);
 
 /**
- * Initialise inode operations on regular files and special files
+ * Initialize inode operations on regular files and special files
  */
 void fuse_init_common(struct inode *inode);
 
 /**
- * Initialise inode and file operations on a directory
+ * Initialize inode and file operations on a directory
  */
 void fuse_init_dir(struct inode *inode);
 
 /**
- * Initialise inode operations on a symlink
+ * Initialize inode operations on a symlink
  */
 void fuse_init_symlink(struct inode *inode);
 
@@ -411,7 +411,7 @@ struct fuse_req *fuse_get_request(struct fuse_conn *fc);
 
 /**
  * Decrement reference count of a request.  If count goes to zero put
- * on unused list (preallocated) or free reqest (not preallocated).
+ * on unused list (preallocated) or free request (not preallocated).
  */
 void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req);
 
@@ -431,7 +431,7 @@ void request_send_noreply(struct fuse_conn *fc, struct fuse_req *req);
 void request_send_background(struct fuse_conn *fc, struct fuse_req *req);
 
 /**
- * Release inodes and file assiciated with background request
+ * Release inodes and file associated with background request
  */
 void fuse_release_background(struct fuse_req *req);
 
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index f1570b9f9de3..3f680c5675bf 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -46,7 +46,7 @@ static sector_t hfs_bmap(struct address_space *mapping, sector_t block)
         return generic_block_bmap(mapping, block, hfs_get_block);
 }
 
-static int hfs_releasepage(struct page *page, int mask)
+static int hfs_releasepage(struct page *page, gfp_t mask)
 {
         struct inode *inode = page->mapping->host;
         struct super_block *sb = inode->i_sb;
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index d5642705f633..f205773ddfbe 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -40,7 +40,7 @@ static sector_t hfsplus_bmap(struct address_space *mapping, sector_t block)
         return generic_block_bmap(mapping, block, hfsplus_get_block);
 }
 
-static int hfsplus_releasepage(struct page *page, int mask)
+static int hfsplus_releasepage(struct page *page, gfp_t mask)
 {
         struct inode *inode = page->mapping->host;
         struct super_block *sb = inode->i_sb;
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 3a9b6d179cbd..e026c807e6b3 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -45,10 +45,58 @@ static struct backing_dev_info hugetlbfs_backing_dev_info = {
 
 int sysctl_hugetlb_shm_group;
 
+static void huge_pagevec_release(struct pagevec *pvec)
+{
+        int i;
+
+        for (i = 0; i < pagevec_count(pvec); ++i)
+                put_page(pvec->pages[i]);
+
+        pagevec_reinit(pvec);
+}
+
+/*
+ * huge_pages_needed tries to determine the number of new huge pages that
+ * will be required to fully populate this VMA.  This will be equal to
+ * the size of the VMA in huge pages minus the number of huge pages
+ * (covered by this VMA) that are found in the page cache.
+ *
+ * Result is in bytes to be compatible with is_hugepage_mem_enough()
+ */
+unsigned long
+huge_pages_needed(struct address_space *mapping, struct vm_area_struct *vma)
+{
+        int i;
+        struct pagevec pvec;
+        unsigned long start = vma->vm_start;
+        unsigned long end = vma->vm_end;
+        unsigned long hugepages = (end - start) >> HPAGE_SHIFT;
+        pgoff_t next = vma->vm_pgoff;
+        pgoff_t endpg = next + ((end - start) >> PAGE_SHIFT);
+
+        pagevec_init(&pvec, 0);
+        while (next < endpg) {
+                if (!pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE))
+                        break;
+                for (i = 0; i < pagevec_count(&pvec); i++) {
+                        struct page *page = pvec.pages[i];
+                        if (page->index > next)
+                                next = page->index;
+                        if (page->index >= endpg)
+                                break;
+                        next++;
+                        hugepages--;
+                }
+                huge_pagevec_release(&pvec);
+        }
+        return hugepages << HPAGE_SHIFT;
+}
+
 static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
 {
         struct inode *inode = file->f_dentry->d_inode;
         struct address_space *mapping = inode->i_mapping;
+        unsigned long bytes;
         loff_t len, vma_len;
         int ret;
 
@@ -67,6 +115,10 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
         if (vma->vm_end - vma->vm_start < HPAGE_SIZE)
                 return -EINVAL;
 
+        bytes = huge_pages_needed(mapping, vma);
+        if (!is_hugepage_mem_enough(bytes))
+                return -ENOMEM;
+
         vma_len = (loff_t)(vma->vm_end - vma->vm_start);
 
         down(&inode->i_sem);
@@ -79,10 +131,8 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
         if (!(vma->vm_flags & VM_WRITE) && len > inode->i_size)
                 goto out;
 
-        ret = hugetlb_prefault(mapping, vma);
-        if (ret)
-                goto out;
-
+        ret = 0;
+        hugetlb_prefault_arch_hook(vma->vm_mm);
         if (inode->i_size < len)
                 inode->i_size = len;
 out:
@@ -92,7 +142,7 @@ out:
 }
 
 /*
- * Called under down_write(mmap_sem), page_table_lock is not held
+ * Called under down_write(mmap_sem).
  */
 
 #ifdef HAVE_ARCH_HUGETLB_UNMAPPED_AREA
@@ -171,16 +221,6 @@ static int hugetlbfs_commit_write(struct file *file,
         return -EINVAL;
 }
 
-static void huge_pagevec_release(struct pagevec *pvec)
-{
-        int i;
-
-        for (i = 0; i < pagevec_count(pvec); ++i)
-                put_page(pvec->pages[i]);
-
-        pagevec_reinit(pvec);
-}
-
 static void truncate_huge_page(struct page *page)
 {
         clear_page_dirty(page);
@@ -224,52 +264,35 @@ static void truncate_hugepages(struct address_space *mapping, loff_t lstart)
 
 static void hugetlbfs_delete_inode(struct inode *inode)
 {
-        struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(inode->i_sb);
-
-        hlist_del_init(&inode->i_hash);
-        list_del_init(&inode->i_list);
-        list_del_init(&inode->i_sb_list);
-        inode->i_state |= I_FREEING;
-        inodes_stat.nr_inodes--;
-        spin_unlock(&inode_lock);
-
         if (inode->i_data.nrpages)
                 truncate_hugepages(&inode->i_data, 0);
-
-        security_inode_delete(inode);
-
-        if (sbinfo->free_inodes >= 0) {
-                spin_lock(&sbinfo->stat_lock);
-                sbinfo->free_inodes++;
-                spin_unlock(&sbinfo->stat_lock);
-        }
-
         clear_inode(inode);
-        destroy_inode(inode);
 }
 
 static void hugetlbfs_forget_inode(struct inode *inode)
 {
-        struct super_block *super_block = inode->i_sb;
-        struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(super_block);
+        struct super_block *sb = inode->i_sb;
 
-        if (hlist_unhashed(&inode->i_hash))
-                goto out_truncate;
-
-        if (!(inode->i_state & (I_DIRTY|I_LOCK))) {
-                list_del(&inode->i_list);
-                list_add(&inode->i_list, &inode_unused);
-        }
-        inodes_stat.nr_unused++;
-        if (!super_block || (super_block->s_flags & MS_ACTIVE)) {
+        if (!hlist_unhashed(&inode->i_hash)) {
+                if (!(inode->i_state & (I_DIRTY|I_LOCK)))
+                        list_move(&inode->i_list, &inode_unused);
+                inodes_stat.nr_unused++;
+                if (!sb || (sb->s_flags & MS_ACTIVE)) {
+                        spin_unlock(&inode_lock);
+                        return;
+                }
+                inode->i_state |= I_WILL_FREE;
                 spin_unlock(&inode_lock);
-                return;
+                /*
+                 * write_inode_now is a noop as we set BDI_CAP_NO_WRITEBACK
+                 * in our backing_dev_info.
+                 */
+                write_inode_now(inode, 1);
+                spin_lock(&inode_lock);
+                inode->i_state &= ~I_WILL_FREE;
+                inodes_stat.nr_unused--;
+                hlist_del_init(&inode->i_hash);
         }
-
-        /* write_inode_now() ? */
-        inodes_stat.nr_unused--;
-        hlist_del_init(&inode->i_hash);
-out_truncate:
         list_del_init(&inode->i_list);
         list_del_init(&inode->i_sb_list);
         inode->i_state |= I_FREEING;
@@ -277,13 +300,6 @@ out_truncate:
         spin_unlock(&inode_lock);
         if (inode->i_data.nrpages)
                 truncate_hugepages(&inode->i_data, 0);
-
-        if (sbinfo->free_inodes >= 0) {
-                spin_lock(&sbinfo->stat_lock);
-                sbinfo->free_inodes++;
-                spin_unlock(&sbinfo->stat_lock);
-        }
-
         clear_inode(inode);
         destroy_inode(inode);
 }
@@ -291,7 +307,7 @@ out_truncate:
 static void hugetlbfs_drop_inode(struct inode *inode)
 {
         if (!inode->i_nlink)
-                hugetlbfs_delete_inode(inode);
+                generic_delete_inode(inode);
         else
                 hugetlbfs_forget_inode(inode);
 }
@@ -308,7 +324,6 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff)
 
         vma_prio_tree_foreach(vma, &iter, root, h_pgoff, ULONG_MAX) {
                 unsigned long h_vm_pgoff;
-                unsigned long v_length;
                 unsigned long v_offset;
 
                 h_vm_pgoff = vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT);
@@ -319,11 +334,8 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff)
                 if (h_vm_pgoff >= h_pgoff)
                         v_offset = 0;
 
-                v_length = vma->vm_end - vma->vm_start;
-
-                zap_hugepage_range(vma,
-                                vma->vm_start + v_offset,
-                                v_length - v_offset);
+                unmap_hugepage_range(vma,
+                                vma->vm_start + v_offset, vma->vm_end);
         }
 }
 
@@ -379,17 +391,6 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid,
                                         gid_t gid, int mode, dev_t dev)
 {
         struct inode *inode;
-        struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(sb);
-
-        if (sbinfo->free_inodes >= 0) {
-                spin_lock(&sbinfo->stat_lock);
-                if (!sbinfo->free_inodes) {
-                        spin_unlock(&sbinfo->stat_lock);
-                        return NULL;
-                }
-                sbinfo->free_inodes--;
-                spin_unlock(&sbinfo->stat_lock);
-        }
 
         inode = new_inode(sb);
         if (inode) {
@@ -531,29 +532,51 @@ static void hugetlbfs_put_super(struct super_block *sb)
         }
 }
 
+static inline int hugetlbfs_dec_free_inodes(struct hugetlbfs_sb_info *sbinfo)
+{
+        if (sbinfo->free_inodes >= 0) {
+                spin_lock(&sbinfo->stat_lock);
+                if (unlikely(!sbinfo->free_inodes)) {
+                        spin_unlock(&sbinfo->stat_lock);
+                        return 0;
+                }
+                sbinfo->free_inodes--;
+                spin_unlock(&sbinfo->stat_lock);
+        }
+
+        return 1;
+}
+
+static void hugetlbfs_inc_free_inodes(struct hugetlbfs_sb_info *sbinfo)
+{
+        if (sbinfo->free_inodes >= 0) {
+                spin_lock(&sbinfo->stat_lock);
+                sbinfo->free_inodes++;
+                spin_unlock(&sbinfo->stat_lock);
+        }
+}
+
+
 static kmem_cache_t *hugetlbfs_inode_cachep;
 
 static struct inode *hugetlbfs_alloc_inode(struct super_block *sb)
 {
+        struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(sb);
         struct hugetlbfs_inode_info *p;
 
+        if (unlikely(!hugetlbfs_dec_free_inodes(sbinfo)))
+                return NULL;
         p = kmem_cache_alloc(hugetlbfs_inode_cachep, SLAB_KERNEL);
-        if (!p)
+        if (unlikely(!p)) {
+                hugetlbfs_inc_free_inodes(sbinfo);
                 return NULL;
+        }
         return &p->vfs_inode;
 }
 
-static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags)
-{
-        struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo;
-
-        if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
-            SLAB_CTOR_CONSTRUCTOR)
-                inode_init_once(&ei->vfs_inode);
-}
-
 static void hugetlbfs_destroy_inode(struct inode *inode)
 {
+        hugetlbfs_inc_free_inodes(HUGETLBFS_SB(inode->i_sb));
         mpol_free_shared_policy(&HUGETLBFS_I(inode)->policy);
         kmem_cache_free(hugetlbfs_inode_cachep, HUGETLBFS_I(inode));
 }
@@ -565,6 +588,16 @@ static struct address_space_operations hugetlbfs_aops = {
         .set_page_dirty = hugetlbfs_set_page_dirty,
 };
 
+
+static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags)
+{
+        struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo;
+
+        if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
+            SLAB_CTOR_CONSTRUCTOR)
+                inode_init_once(&ei->vfs_inode);
+}
+
 struct file_operations hugetlbfs_file_operations = {
         .mmap                   = hugetlbfs_file_mmap,
         .fsync                  = simple_sync_file,
@@ -592,6 +625,7 @@ static struct super_operations hugetlbfs_ops = {
         .alloc_inode    = hugetlbfs_alloc_inode,
         .destroy_inode  = hugetlbfs_destroy_inode,
         .statfs         = hugetlbfs_statfs,
+        .delete_inode   = hugetlbfs_delete_inode,
         .drop_inode     = hugetlbfs_drop_inode,
         .put_super      = hugetlbfs_put_super,
 };
diff --git a/fs/inode.c b/fs/inode.c
index f80a79ff156b..d8d04bd72b59 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -475,7 +475,7 @@ static void prune_icache(int nr_to_scan)
  * This function is passed the number of inodes to scan, and it returns the
  * total number of remaining possibly-reclaimable inodes.
  */
-static int shrink_icache_memory(int nr, unsigned int gfp_mask)
+static int shrink_icache_memory(int nr, gfp_t gfp_mask)
 {
         if (nr) {
                 /*
@@ -1088,6 +1088,7 @@ static void generic_forget_inode(struct inode *inode)
         if (inode->i_data.nrpages)
                 truncate_inode_pages(&inode->i_data, 0);
         clear_inode(inode);
+        wake_up_inode(inode);
         destroy_inode(inode);
 }
 
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 7ae2c4fe506b..e4b516ac4989 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -1606,7 +1606,7 @@ int journal_blocks_per_page(struct inode *inode)
  * Simple support for retrying memory allocations.  Introduced to help to
  * debug different VM deadlock avoidance strategies. 
  */
-void * __jbd_kmalloc (const char *where, size_t size, int flags, int retry)
+void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry)
 {
         return kmalloc(size, flags | (retry ? __GFP_NOFAIL : 0));
 }
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index 49bbc2be3d72..13cb05bf6048 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -1621,7 +1621,7 @@ out:
  * while the data is part of a transaction.  Yes?
  */
 int journal_try_to_free_buffers(journal_t *journal, 
-                                struct page *page, int unused_gfp_mask)
+                                struct page *page, gfp_t unused_gfp_mask)
 {
         struct buffer_head *head;
         struct buffer_head *bh;
diff --git a/fs/jffs2/background.c b/fs/jffs2/background.c
index 0f224384f176..8210ac16a368 100644
--- a/fs/jffs2/background.c
+++ b/fs/jffs2/background.c
@@ -15,6 +15,7 @@
 #include <linux/jffs2.h>
 #include <linux/mtd/mtd.h>
 #include <linux/completion.h>
+#include <linux/sched.h>
 #include "nodelist.h"
 
 
diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c
index 996d922e503e..316133c626b7 100644
--- a/fs/jffs2/wbuf.c
+++ b/fs/jffs2/wbuf.c
@@ -18,6 +18,8 @@
 #include <linux/mtd/mtd.h>
 #include <linux/crc32.h>
 #include <linux/mtd/nand.h>
+#include <linux/jiffies.h>
+
 #include "nodelist.h"
 
 /* For testing write failures */
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index eadf319bee22..68000a50ceb6 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -74,7 +74,7 @@
 static void dbAllocBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
                         int nblocks);
 static void dbSplit(dmtree_t * tp, int leafno, int splitsz, int newval);
-static void dbBackSplit(dmtree_t * tp, int leafno);
+static int dbBackSplit(dmtree_t * tp, int leafno);
 static int dbJoin(dmtree_t * tp, int leafno, int newval);
 static void dbAdjTree(dmtree_t * tp, int leafno, int newval);
 static int dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc,
@@ -305,7 +305,6 @@ int dbSync(struct inode *ipbmap)
         filemap_fdatawrite(ipbmap->i_mapping);
         filemap_fdatawait(ipbmap->i_mapping);
 
-        ipbmap->i_state |= I_DIRTY;
         diWriteSpecial(ipbmap, 0);
 
         return (0);
@@ -2467,7 +2466,9 @@ dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc, int level)
                  * that it is at the front of a binary buddy system.
                  */
                 if (oldval == NOFREE) {
-                        dbBackSplit((dmtree_t *) dcp, leafno);
+                        rc = dbBackSplit((dmtree_t *) dcp, leafno);
+                        if (rc)
+                                return rc;
                         oldval = dcp->stree[ti];
                 }
                 dbSplit((dmtree_t *) dcp, leafno, dcp->budmin, newval);
@@ -2627,7 +2628,7 @@ static void dbSplit(dmtree_t * tp, int leafno, int splitsz, int newval)
  *
  * serialization: IREAD_LOCK(ipbmap) or IWRITE_LOCK(ipbmap) held on entry/exit;
  */
-static void dbBackSplit(dmtree_t * tp, int leafno)
+static int dbBackSplit(dmtree_t * tp, int leafno)
 {
         int budsz, bud, w, bsz, size;
         int cursz;
@@ -2662,7 +2663,10 @@ static void dbBackSplit(dmtree_t * tp, int leafno)
                  */
                 for (w = leafno, bsz = budsz;; bsz <<= 1,
                      w = (w < bud) ? w : bud) {
-                        assert(bsz < le32_to_cpu(tp->dmt_nleafs));
+                        if (bsz >= le32_to_cpu(tp->dmt_nleafs)) {
+                                jfs_err("JFS: block map error in dbBackSplit");
+                                return -EIO;
+                        }
 
                         /* determine the buddy.
                          */
@@ -2681,7 +2685,11 @@ static void dbBackSplit(dmtree_t * tp, int leafno)
                 }
         }
 
-        assert(leaf[leafno] == size);
+        if (leaf[leafno] != size) {
+                jfs_err("JFS: wrong leaf value in dbBackSplit");
+                return -EIO;
+        }
+        return 0;
 }
 
 
diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
index 4021d46da7e3..28201b194f53 100644
--- a/fs/jfs/jfs_imap.c
+++ b/fs/jfs/jfs_imap.c
@@ -57,6 +57,12 @@
 #include "jfs_debug.h"
 
 /*
+ * __mark_inode_dirty expects inodes to be hashed.  Since we don't want
+ * special inodes in the fileset inode space, we hash them to a dummy head
+ */
+static HLIST_HEAD(aggregate_hash);
+
+/*
  * imap locks
  */
 /* iag free list lock */
@@ -491,6 +497,8 @@ struct inode *diReadSpecial(struct super_block *sb, ino_t inum, int secondary)
         /* release the page */
         release_metapage(mp);
 
+        hlist_add_head(&ip->i_hash, &aggregate_hash);
+
         return (ip);
 }
 
@@ -514,8 +522,6 @@ void diWriteSpecial(struct inode *ip, int secondary)
         ino_t inum = ip->i_ino;
         struct metapage *mp;
 
-        ip->i_state &= ~I_DIRTY;
-
         if (secondary)
                 address = addressPXD(&sbi->ait2) >> sbi->l2nbperpage;
         else
diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c
index 13d7e3f1feb4..8a53981f9f27 100644
--- a/fs/jfs/jfs_metapage.c
+++ b/fs/jfs/jfs_metapage.c
@@ -86,7 +86,7 @@ struct meta_anchor {
         atomic_t io_count;
         struct metapage *mp[MPS_PER_PAGE];
 };
-#define mp_anchor(page) ((struct meta_anchor *)page->private)
+#define mp_anchor(page) ((struct meta_anchor *)page_private(page))
 
 static inline struct metapage *page_to_mp(struct page *page, uint offset)
 {
@@ -108,7 +108,7 @@ static inline int insert_metapage(struct page *page, struct metapage *mp)
                 if (!a)
                         return -ENOMEM;
                 memset(a, 0, sizeof(struct meta_anchor));
-                page->private = (unsigned long)a;
+                set_page_private(page, (unsigned long)a);
                 SetPagePrivate(page);
                 kmap(page);
         }
@@ -136,7 +136,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp)
         a->mp[index] = NULL;
         if (--a->mp_count == 0) {
                 kfree(a);
-                page->private = 0;
+                set_page_private(page, 0);
                 ClearPagePrivate(page);
                 kunmap(page);
         }
@@ -156,13 +156,13 @@ static inline void dec_io(struct page *page, void (*handler) (struct page *))
 #else
 static inline struct metapage *page_to_mp(struct page *page, uint offset)
 {
-        return PagePrivate(page) ? (struct metapage *)page->private : NULL;
+        return PagePrivate(page) ? (struct metapage *)page_private(page) : NULL;
 }
 
 static inline int insert_metapage(struct page *page, struct metapage *mp)
 {
         if (mp) {
-                page->private = (unsigned long)mp;
+                set_page_private(page, (unsigned long)mp);
                 SetPagePrivate(page);
                 kmap(page);
         }
@@ -171,7 +171,7 @@ static inline int insert_metapage(struct page *page, struct metapage *mp)
 
 static inline void remove_metapage(struct page *page, struct metapage *mp)
 {
-        page->private = 0;
+        set_page_private(page, 0);
         ClearPagePrivate(page);
         kunmap(page);
 }
@@ -198,7 +198,7 @@ static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags)
         }
 }
 
-static inline struct metapage *alloc_metapage(unsigned int gfp_mask)
+static inline struct metapage *alloc_metapage(gfp_t gfp_mask)
 {
         return mempool_alloc(metapage_mempool, gfp_mask);
 }
@@ -395,6 +395,12 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc)
 
                 if (mp->nohomeok && !test_bit(META_forcewrite, &mp->flag)) {
                         redirty = 1;
+                        /*
+                         * Make sure this page isn't blocked indefinitely.
+                         * If the journal isn't undergoing I/O, push it
+                         */
+                        if (mp->log && !(mp->log->cflag & logGC_PAGEOUT))
+                                jfs_flush_journal(mp->log, 0);
                         continue;
                 }
 
@@ -534,7 +540,7 @@ add_failed:
         return -EIO;
 }
 
-static int metapage_releasepage(struct page *page, int gfp_mask)
+static int metapage_releasepage(struct page *page, gfp_t gfp_mask)
 {
         struct metapage *mp;
         int busy = 0;
diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index 9b71ed2674fe..b660c93c92de 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -2396,7 +2396,6 @@ static void txUpdateMap(struct tblock * tblk)
          */
         if (tblk->xflag & COMMIT_CREATE) {
                 diUpdatePMap(ipimap, tblk->ino, FALSE, tblk);
-                ipimap->i_state |= I_DIRTY;
                 /* update persistent block allocation map
                  * for the allocation of inode extent;
                  */
@@ -2407,7 +2406,6 @@ static void txUpdateMap(struct tblock * tblk)
         } else if (tblk->xflag & COMMIT_DELETE) {
                 ip = tblk->u.ip;
                 diUpdatePMap(ipimap, ip->i_ino, TRUE, tblk);
-                ipimap->i_state |= I_DIRTY;
                 iput(ip);
         }
 }
diff --git a/fs/jfs/jfs_xtree.c b/fs/jfs/jfs_xtree.c
index a7fe2f2b969f..e72f4ebb6e9c 100644
--- a/fs/jfs/jfs_xtree.c
+++ b/fs/jfs/jfs_xtree.c
@@ -3516,16 +3516,10 @@ s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag)
         /* process entries backward from last index */
         index = le16_to_cpu(p->header.nextindex) - 1;
 
-        if (p->header.flag & BT_INTERNAL)
-                goto getChild;
-
-        /*
-         *      leaf page
-         */
 
-        /* Since this is the rightmost leaf, and we may have already freed
-         * a page that was formerly to the right, let's make sure that the
-         * next pointer is zero.
+        /* Since this is the rightmost page at this level, and we may have
+         * already freed a page that was formerly to the right, let's make
+         * sure that the next pointer is zero.
          */
         if (p->header.next) {
                 if (log)
@@ -3539,6 +3533,12 @@ s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag)
                 p->header.next = 0;
         }
 
+        if (p->header.flag & BT_INTERNAL)
+                goto getChild;
+
+        /*
+         *      leaf page
+         */
         freed = 0;
 
         /* does region covered by leaf page precede Teof ? */
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 71bc34b96b2b..4226af3ea91b 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -442,6 +442,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
         inode->i_nlink = 1;
         inode->i_size = sb->s_bdev->bd_inode->i_size;
         inode->i_mapping->a_ops = &jfs_metapage_aops;
+        insert_inode_hash(inode);
         mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);
 
         sbi->direct_inode = inode;
diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 82c77df81c5f..c4c8601096e0 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -173,11 +173,10 @@ nlm_bind_host(struct nlm_host *host)
 
         /* If we've already created an RPC client, check whether
          * RPC rebind is required
-         * Note: why keep rebinding if we're on a tcp connection?
          */
         if ((clnt = host->h_rpcclnt) != NULL) {
                 xprt = clnt->cl_xprt;
-                if (!xprt->stream && time_after_eq(jiffies, host->h_nextrebind)) {
+                if (time_after_eq(jiffies, host->h_nextrebind)) {
                         clnt->cl_port = 0;
                         host->h_nextrebind = jiffies + NLM_HOST_REBIND;
                         dprintk("lockd: next rebind in %ld jiffies\n",
@@ -189,7 +188,6 @@ nlm_bind_host(struct nlm_host *host)
                         goto forgetit;
 
                 xprt_set_timeout(&xprt->timeout, 5, nlmsvc_timeout);
-                xprt->nocong = 1;       /* No congestion control for NLM */
                 xprt->resvport = 1;     /* NLM requires a reserved port */
 
                 /* Existing NLM servers accept AUTH_UNIX only */
diff --git a/fs/locks.c b/fs/locks.c
index f7daa5f48949..a1e8b2248014 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -316,21 +316,22 @@ static int flock_to_posix_lock(struct file *filp, struct file_lock *fl,
         /* POSIX-1996 leaves the case l->l_len < 0 undefined;
            POSIX-2001 defines it. */
         start += l->l_start;
-        end = start + l->l_len - 1;
-        if (l->l_len < 0) {
+        if (start < 0)
+                return -EINVAL;
+        fl->fl_end = OFFSET_MAX;
+        if (l->l_len > 0) {
+                end = start + l->l_len - 1;
+                fl->fl_end = end;
+        } else if (l->l_len < 0) {
                 end = start - 1;
+                fl->fl_end = end;
                 start += l->l_len;
+                if (start < 0)
+                        return -EINVAL;
         }
-
-        if (start < 0)
-                return -EINVAL;
-        if (l->l_len > 0 && end < 0)
-                return -EOVERFLOW;
-
         fl->fl_start = start;   /* we record the absolute position */
-        fl->fl_end = end;
-        if (l->l_len == 0)
-                fl->fl_end = OFFSET_MAX;
+        if (fl->fl_end < fl->fl_start)
+                return -EOVERFLOW;
         
         fl->fl_owner = current->files;
         fl->fl_pid = current->tgid;
@@ -362,14 +363,21 @@ static int flock64_to_posix_lock(struct file *filp, struct file_lock *fl,
                 return -EINVAL;
         }
 
-        if (((start += l->l_start) < 0) || (l->l_len < 0))
+        start += l->l_start;
+        if (start < 0)
                 return -EINVAL;
-        fl->fl_end = start + l->l_len - 1;
-        if (l->l_len > 0 && fl->fl_end < 0)
-                return -EOVERFLOW;
+        fl->fl_end = OFFSET_MAX;
+        if (l->l_len > 0) {
+                fl->fl_end = start + l->l_len - 1;
+        } else if (l->l_len < 0) {
+                fl->fl_end = start - 1;
+                start += l->l_len;
+                if (start < 0)
+                        return -EINVAL;
+        }
         fl->fl_start = start;   /* we record the absolute position */
-        if (l->l_len == 0)
-                fl->fl_end = OFFSET_MAX;
+        if (fl->fl_end < fl->fl_start)
+                return -EOVERFLOW;
         
         fl->fl_owner = current->files;
         fl->fl_pid = current->tgid;
@@ -829,12 +837,16 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request)
                 /* Detect adjacent or overlapping regions (if same lock type)
                  */
                 if (request->fl_type == fl->fl_type) {
+                        /* In all comparisons of start vs end, use
+                         * "start - 1" rather than "end + 1". If end
+                         * is OFFSET_MAX, end + 1 will become negative.
+                         */
                         if (fl->fl_end < request->fl_start - 1)
                                 goto next_lock;
                         /* If the next lock in the list has entirely bigger
                          * addresses than the new one, insert the lock here.
                          */
-                        if (fl->fl_start > request->fl_end + 1)
+                        if (fl->fl_start - 1 > request->fl_end)
                                 break;
 
                         /* If we come here, the new and old lock are of the
diff --git a/fs/mbcache.c b/fs/mbcache.c
index b002a088857d..298997f17475 100644
--- a/fs/mbcache.c
+++ b/fs/mbcache.c
@@ -116,7 +116,7 @@ mb_cache_indexes(struct mb_cache *cache)
  * What the mbcache registers as to get shrunk dynamically.
  */
 
-static int mb_cache_shrink_fn(int nr_to_scan, unsigned int gfp_mask);
+static int mb_cache_shrink_fn(int nr_to_scan, gfp_t gfp_mask);
 
 
 static inline int
@@ -140,7 +140,7 @@ __mb_cache_entry_unhash(struct mb_cache_entry *ce)
 
 
 static inline void
-__mb_cache_entry_forget(struct mb_cache_entry *ce, int gfp_mask)
+__mb_cache_entry_forget(struct mb_cache_entry *ce, gfp_t gfp_mask)
 {
         struct mb_cache *cache = ce->e_cache;
 
@@ -193,7 +193,7 @@ forget:
  * Returns the number of objects which are present in the cache.
  */
 static int
-mb_cache_shrink_fn(int nr_to_scan, unsigned int gfp_mask)
+mb_cache_shrink_fn(int nr_to_scan, gfp_t gfp_mask)
 {
         LIST_HEAD(free_list);
         struct list_head *l, *ltmp;
diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c
index 154f511c7245..626a367bcd81 100644
--- a/fs/msdos/namei.c
+++ b/fs/msdos/namei.c
@@ -454,10 +454,10 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
 {
         struct buffer_head *dotdot_bh;
         struct msdos_dir_entry *dotdot_de;
-        loff_t dotdot_i_pos;
         struct inode *old_inode, *new_inode;
         struct fat_slot_info old_sinfo, sinfo;
         struct timespec ts;
+        loff_t dotdot_i_pos, new_i_pos;
         int err, old_attrs, is_dir, update_dotdot, corrupt = 0;
 
         old_sinfo.bh = sinfo.bh = dotdot_bh = NULL;
@@ -516,28 +516,24 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
         if (new_inode) {
                 if (err)
                         goto out;
-                if (MSDOS_I(new_inode)->i_pos != sinfo.i_pos) {
-                        /* WTF??? Cry and fail. */
-                        printk(KERN_WARNING "msdos_rename: fs corrupted\n");
-                        goto out;
-                }
-
                 if (is_dir) {
                         err = fat_dir_empty(new_inode);
                         if (err)
                                 goto out;
                 }
+                new_i_pos = MSDOS_I(new_inode)->i_pos;
                 fat_detach(new_inode);
         } else {
                 err = msdos_add_entry(new_dir, new_name, is_dir, is_hid, 0,
                                       &ts, &sinfo);
                 if (err)
                         goto out;
+                new_i_pos = sinfo.i_pos;
         }
         new_dir->i_version++;
 
         fat_detach(old_inode);
-        fat_attach(old_inode, sinfo.i_pos);
+        fat_attach(old_inode, new_i_pos);
         if (is_hid)
                 MSDOS_I(old_inode)->i_attrs |= ATTR_HIDDEN;
         else
@@ -604,7 +600,7 @@ error_inode:
         fat_attach(old_inode, old_sinfo.i_pos);
         MSDOS_I(old_inode)->i_attrs = old_attrs;
         if (new_inode) {
-                fat_attach(new_inode, sinfo.i_pos);
+                fat_attach(new_inode, new_i_pos);
                 if (corrupt)
                         corrupt |= fat_sync_inode(new_inode);
         } else {
diff --git a/fs/namei.c b/fs/namei.c
index aa62dbda93ac..c5769c4fcab1 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -28,6 +28,7 @@
 #include <linux/syscalls.h>
 #include <linux/mount.h>
 #include <linux/audit.h>
+#include <linux/file.h>
 #include <asm/namei.h>
 #include <asm/uaccess.h>
 
@@ -317,6 +318,18 @@ void path_release_on_umount(struct nameidata *nd)
         mntput_no_expire(nd->mnt);
 }
 
+/**
+ * release_open_intent - free up open intent resources
+ * @nd: pointer to nameidata
+ */
+void release_open_intent(struct nameidata *nd)
+{
+        if (nd->intent.open.file->f_dentry == NULL)
+                put_filp(nd->intent.open.file);
+        else
+                fput(nd->intent.open.file);
+}
+
 /*
  * Internal lookup() using the new generic dcache.
  * SMP-safe
@@ -750,6 +763,7 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd)
                 struct qstr this;
                 unsigned int c;
 
+                nd->flags |= LOOKUP_CONTINUE;
                 err = exec_permission_lite(inode, nd);
                 if (err == -EAGAIN) { 
                         err = permission(inode, MAY_EXEC, nd);
@@ -802,7 +816,6 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd)
                         if (err < 0)
                                 break;
                 }
-                nd->flags |= LOOKUP_CONTINUE;
                 /* This does the actual lookups.. */
                 err = do_lookup(nd, &this, &next);
                 if (err)
@@ -1052,6 +1065,70 @@ out:
         return retval;
 }
 
+static int __path_lookup_intent_open(const char *name, unsigned int lookup_flags,
+                struct nameidata *nd, int open_flags, int create_mode)
+{
+        struct file *filp = get_empty_filp();
+        int err;
+
+        if (filp == NULL)
+                return -ENFILE;
+        nd->intent.open.file = filp;
+        nd->intent.open.flags = open_flags;
+        nd->intent.open.create_mode = create_mode;
+        err = path_lookup(name, lookup_flags|LOOKUP_OPEN, nd);
+        if (IS_ERR(nd->intent.open.file)) {
+                if (err == 0) {
+                        err = PTR_ERR(nd->intent.open.file);
+                        path_release(nd);
+                }
+        } else if (err != 0)
+                release_open_intent(nd);
+        return err;
+}
+
+/**
+ * path_lookup_open - lookup a file path with open intent
+ * @name: pointer to file name
+ * @lookup_flags: lookup intent flags
+ * @nd: pointer to nameidata
+ * @open_flags: open intent flags
+ */
+int path_lookup_open(const char *name, unsigned int lookup_flags,
+                struct nameidata *nd, int open_flags)
+{
+        return __path_lookup_intent_open(name, lookup_flags, nd,
+                        open_flags, 0);
+}
+
+/**
+ * path_lookup_create - lookup a file path with open + create intent
+ * @name: pointer to file name
+ * @lookup_flags: lookup intent flags
+ * @nd: pointer to nameidata
+ * @open_flags: open intent flags
+ * @create_mode: create intent flags
+ */
+int path_lookup_create(const char *name, unsigned int lookup_flags,
+                struct nameidata *nd, int open_flags, int create_mode)
+{
+        return __path_lookup_intent_open(name, lookup_flags|LOOKUP_CREATE, nd,
+                        open_flags, create_mode);
+}
+
+int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags,
+                struct nameidata *nd, int open_flags)
+{
+        char *tmp = getname(name);
+        int err = PTR_ERR(tmp);
+
+        if (!IS_ERR(tmp)) {
+                err = __path_lookup_intent_open(tmp, lookup_flags, nd, open_flags, 0);
+                putname(tmp);
+        }
+        return err;
+}
+
 /*
  * Restricted form of lookup. Doesn't follow links, single-component only,
  * needs parent already locked. Doesn't follow mounts.
@@ -1234,9 +1311,6 @@ static inline int may_create(struct inode *dir, struct dentry *child,
 }
 
 /* 
- * Special case: O_CREAT|O_EXCL implies O_NOFOLLOW for security
- * reasons.
- *
  * O_DIRECTORY translates into forcing a directory lookup.
  */
 static inline int lookup_flags(unsigned int f)
@@ -1246,9 +1320,6 @@ static inline int lookup_flags(unsigned int f)
         if (f & O_NOFOLLOW)
                 retval &= ~LOOKUP_FOLLOW;
         
-        if ((f & (O_CREAT|O_EXCL)) == (O_CREAT|O_EXCL))
-                retval &= ~LOOKUP_FOLLOW;
-        
         if (f & O_DIRECTORY)
                 retval |= LOOKUP_DIRECTORY;
 
@@ -1416,27 +1487,27 @@ int may_open(struct nameidata *nd, int acc_mode, int flag)
  */
 int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd)
 {
-        int acc_mode, error = 0;
+        int acc_mode, error;
         struct path path;
         struct dentry *dir;
         int count = 0;
 
         acc_mode = ACC_MODE(flag);
 
+        /* O_TRUNC implies we need access checks for write permissions */
+        if (flag & O_TRUNC)
+                acc_mode |= MAY_WRITE;
+
         /* Allow the LSM permission hook to distinguish append 
            access from general write access. */
         if (flag & O_APPEND)
                 acc_mode |= MAY_APPEND;
 
-        /* Fill in the open() intent data */
-        nd->intent.open.flags = flag;
-        nd->intent.open.create_mode = mode;
-
         /*
          * The simplest case - just a plain lookup.
          */
         if (!(flag & O_CREAT)) {
-                error = path_lookup(pathname, lookup_flags(flag)|LOOKUP_OPEN, nd);
+                error = path_lookup_open(pathname, lookup_flags(flag), nd, flag);
                 if (error)
                         return error;
                 goto ok;
@@ -1445,7 +1516,7 @@ int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd)
         /*
          * Create - we need to know the parent.
          */
-        error = path_lookup(pathname, LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE, nd);
+        error = path_lookup_create(pathname, LOOKUP_PARENT, nd, flag, mode);
         if (error)
                 return error;
 
@@ -1520,6 +1591,8 @@ ok:
 exit_dput:
         dput_path(&path, nd);
 exit:
+        if (!IS_ERR(nd->intent.open.file))
+                release_open_intent(nd);
         path_release(nd);
         return error;
 
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 4a36839f0bbd..44135af9894c 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -142,7 +142,7 @@ static void nfs_msync_inode(struct inode *inode)
 /*
  * Basic procedure for returning a delegation to the server
  */
-int nfs_inode_return_delegation(struct inode *inode)
+int __nfs_inode_return_delegation(struct inode *inode)
 {
         struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
         struct nfs_inode *nfsi = NFS_I(inode);
diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h
index 3f6c45a29d6a..8017846b561f 100644
--- a/fs/nfs/delegation.h
+++ b/fs/nfs/delegation.h
@@ -25,7 +25,7 @@ struct nfs_delegation {
 
 int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
 void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
-int nfs_inode_return_delegation(struct inode *inode);
+int __nfs_inode_return_delegation(struct inode *inode);
 int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid);
 
 struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle);
@@ -47,11 +47,25 @@ static inline int nfs_have_delegation(struct inode *inode, int flags)
                 return 1;
         return 0;
 }
+
+static inline int nfs_inode_return_delegation(struct inode *inode)
+{
+        int err = 0;
+
+        if (NFS_I(inode)->delegation != NULL)
+                err = __nfs_inode_return_delegation(inode);
+        return err;
+}
 #else
 static inline int nfs_have_delegation(struct inode *inode, int flags)
 {
         return 0;
 }
+
+static inline int nfs_inode_return_delegation(struct inode *inode)
+{
+        return 0;
+}
 #endif
 
 #endif
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 2df639f143e8..8272ed3fc707 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -532,6 +532,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
         my_entry.eof = 0;
         my_entry.fh = &fh;
         my_entry.fattr = &fattr;
+        nfs_fattr_init(&fattr);
         desc->entry = &my_entry;
 
         while(!desc->entry->eof) {
@@ -565,8 +566,6 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
                 }
         }
         unlock_kernel();
-        if (desc->error < 0)
-                return desc->error;
         if (res < 0)
                 return res;
         return 0;
@@ -803,6 +802,7 @@ static int nfs_dentry_delete(struct dentry *dentry)
  */
 static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode)
 {
+        nfs_inode_return_delegation(inode);
         if (dentry->d_flags & DCACHE_NFSFS_RENAMED) {
                 lock_kernel();
                 inode->i_nlink--;
@@ -853,12 +853,6 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
         dentry->d_op = NFS_PROTO(dir)->dentry_ops;
 
         lock_kernel();
-        /* Revalidate parent directory attribute cache */
-        error = nfs_revalidate_inode(NFS_SERVER(dir), dir);
-        if (error < 0) {
-                res = ERR_PTR(error);
-                goto out_unlock;
-        }
 
         /* If we're doing an exclusive create, optimize away the lookup */
         if (nfs_is_exclusive_create(dir, nd))
@@ -916,7 +910,6 @@ static int is_atomic_open(struct inode *dir, struct nameidata *nd)
 static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
         struct dentry *res = NULL;
-        struct inode *inode = NULL;
         int error;
 
         /* Check that we are indeed trying to open this file */
@@ -930,8 +923,10 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry
         dentry->d_op = NFS_PROTO(dir)->dentry_ops;
 
         /* Let vfs_create() deal with O_EXCL */
-        if (nd->intent.open.flags & O_EXCL)
-                goto no_entry;
+        if (nd->intent.open.flags & O_EXCL) {
+                d_add(dentry, NULL);
+                goto out;
+        }
 
         /* Open the file on the server */
         lock_kernel();
@@ -945,32 +940,30 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry
 
         if (nd->intent.open.flags & O_CREAT) {
                 nfs_begin_data_update(dir);
-                inode = nfs4_atomic_open(dir, dentry, nd);
+                res = nfs4_atomic_open(dir, dentry, nd);
                 nfs_end_data_update(dir);
         } else
-                inode = nfs4_atomic_open(dir, dentry, nd);
+                res = nfs4_atomic_open(dir, dentry, nd);
         unlock_kernel();
-        if (IS_ERR(inode)) {
-                error = PTR_ERR(inode);
+        if (IS_ERR(res)) {
+                error = PTR_ERR(res);
                 switch (error) {
                         /* Make a negative dentry */
                         case -ENOENT:
-                                inode = NULL;
-                                break;
+                                res = NULL;
+                                goto out;
                         /* This turned out not to be a regular file */
+                        case -EISDIR:
+                        case -ENOTDIR:
+                                goto no_open;
                         case -ELOOP:
                                 if (!(nd->intent.open.flags & O_NOFOLLOW))
                                         goto no_open;
-                        /* case -EISDIR: */
                         /* case -EINVAL: */
                         default:
-                                res = ERR_PTR(error);
                                 goto out;
                 }
-        }
-no_entry:
-        res = d_add_unique(dentry, inode);
-        if (res != NULL)
+        } else if (res != NULL)
                 dentry = res;
         nfs_renew_times(dentry);
         nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
@@ -1014,7 +1007,7 @@ static int nfs_open_revalidate(struct dentry *dentry, struct nameidata *nd)
          */
         lock_kernel();
         verifier = nfs_save_change_attribute(dir);
-        ret = nfs4_open_revalidate(dir, dentry, openflags);
+        ret = nfs4_open_revalidate(dir, dentry, openflags, nd);
         if (!ret)
                 nfs_set_verifier(dentry, verifier);
         unlock_kernel();
@@ -1137,7 +1130,7 @@ static int nfs_create(struct inode *dir, struct dentry *dentry, int mode,
 
         lock_kernel();
         nfs_begin_data_update(dir);
-        error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags);
+        error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags, nd);
         nfs_end_data_update(dir);
         if (error != 0)
                 goto out_err;
@@ -1332,6 +1325,7 @@ static int nfs_safe_remove(struct dentry *dentry)
 
         nfs_begin_data_update(dir);
         if (inode != NULL) {
+                nfs_inode_return_delegation(inode);
                 nfs_begin_data_update(inode);
                 error = NFS_PROTO(dir)->remove(dir, &dentry->d_name);
                 /* The VFS may want to delete this inode */
@@ -1438,17 +1432,14 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
                 old_dentry->d_parent->d_name.name, old_dentry->d_name.name,
                 dentry->d_parent->d_name.name, dentry->d_name.name);
 
-        /*
-         * Drop the dentry in advance to force a new lookup.
-         * Since nfs_proc_link doesn't return a file handle,
-         * we can't use the existing dentry.
-         */
         lock_kernel();
-        d_drop(dentry);
-
         nfs_begin_data_update(dir);
         nfs_begin_data_update(inode);
         error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name);
+        if (error == 0) {
+                atomic_inc(&inode->i_count);
+                d_instantiate(dentry, inode);
+        }
         nfs_end_data_update(inode);
         nfs_end_data_update(dir);
         unlock_kernel();
@@ -1512,9 +1503,11 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
          */
         if (!new_inode)
                 goto go_ahead;
-        if (S_ISDIR(new_inode->i_mode))
-                goto out;
-        else if (atomic_read(&new_dentry->d_count) > 2) {
+        if (S_ISDIR(new_inode->i_mode)) {
+                error = -EISDIR;
+                if (!S_ISDIR(old_inode->i_mode))
+                        goto out;
+        } else if (atomic_read(&new_dentry->d_count) > 2) {
                 int err;
                 /* copy the target dentry's name */
                 dentry = d_alloc(new_dentry->d_parent,
@@ -1539,7 +1532,8 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 #endif
                         goto out;
                 }
-        }
+        } else
+                new_inode->i_nlink--;
 
 go_ahead:
         /*
@@ -1549,6 +1543,7 @@ go_ahead:
                 nfs_wb_all(old_inode);
                 shrink_dcache_parent(old_dentry);
         }
+        nfs_inode_return_delegation(old_inode);
 
         if (new_inode)
                 d_delete(new_dentry);
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 6bdcfa95de94..57d3e77d97ee 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -205,8 +205,8 @@ nfs_file_flush(struct file *file)
         if (!status) {
                 status = ctx->error;
                 ctx->error = 0;
-                if (!status && !nfs_have_delegation(inode, FMODE_READ))
-                        __nfs_revalidate_inode(NFS_SERVER(inode), inode);
+                if (!status)
+                        nfs_revalidate_inode(NFS_SERVER(inode), inode);
         }
         unlock_kernel();
         return status;
@@ -376,22 +376,31 @@ out_swapfile:
 
 static int do_getlk(struct file *filp, int cmd, struct file_lock *fl)
 {
+        struct file_lock *cfl;
         struct inode *inode = filp->f_mapping->host;
         int status = 0;
 
         lock_kernel();
-        /* Use local locking if mounted with "-onolock" */
-        if (!(NFS_SERVER(inode)->flags & NFS_MOUNT_NONLM))
-                status = NFS_PROTO(inode)->lock(filp, cmd, fl);
-        else {
-                struct file_lock *cfl = posix_test_lock(filp, fl);
-
-                fl->fl_type = F_UNLCK;
-                if (cfl != NULL)
-                        memcpy(fl, cfl, sizeof(*fl));
+        /* Try local locking first */
+        cfl = posix_test_lock(filp, fl);
+        if (cfl != NULL) {
+                locks_copy_lock(fl, cfl);
+                goto out;
         }
+
+        if (nfs_have_delegation(inode, FMODE_READ))
+                goto out_noconflict;
+
+        if (NFS_SERVER(inode)->flags & NFS_MOUNT_NONLM)
+                goto out_noconflict;
+
+        status = NFS_PROTO(inode)->lock(filp, cmd, fl);
+out:
         unlock_kernel();
         return status;
+out_noconflict:
+        fl->fl_type = F_UNLCK;
+        goto out;
 }
 
 static int do_vfs_lock(struct file *file, struct file_lock *fl)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index d4eadeea128e..fc0f12ba89cc 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -358,6 +358,35 @@ out_no_root:
         return no_root_error;
 }
 
+static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, unsigned int timeo, unsigned int retrans)
+{
+        to->to_initval = timeo * HZ / 10;
+        to->to_retries = retrans;
+        if (!to->to_retries)
+                to->to_retries = 2;
+
+        switch (proto) {
+        case IPPROTO_TCP:
+                if (!to->to_initval)
+                        to->to_initval = 60 * HZ;
+                if (to->to_initval > NFS_MAX_TCP_TIMEOUT)
+                        to->to_initval = NFS_MAX_TCP_TIMEOUT;
+                to->to_increment = to->to_initval;
+                to->to_maxval = to->to_initval + (to->to_increment * to->to_retries);
+                to->to_exponential = 0;
+                break;
+        case IPPROTO_UDP:
+        default:
+                if (!to->to_initval)
+                        to->to_initval = 11 * HZ / 10;
+                if (to->to_initval > NFS_MAX_UDP_TIMEOUT)
+                        to->to_initval = NFS_MAX_UDP_TIMEOUT;
+                to->to_maxval = NFS_MAX_UDP_TIMEOUT;
+                to->to_exponential = 1;
+                break;
+        }
+}
+
 /*
  * Create an RPC client handle.
  */
@@ -367,22 +396,12 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
         struct rpc_timeout      timeparms;
         struct rpc_xprt         *xprt = NULL;
         struct rpc_clnt         *clnt = NULL;
-        int                     tcp   = (data->flags & NFS_MOUNT_TCP);
+        int                     proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
 
-        /* Initialize timeout values */
-        timeparms.to_initval = data->timeo * HZ / 10;
-        timeparms.to_retries = data->retrans;
-        timeparms.to_maxval  = tcp ? RPC_MAX_TCP_TIMEOUT : RPC_MAX_UDP_TIMEOUT;
-        timeparms.to_exponential = 1;
-
-        if (!timeparms.to_initval)
-                timeparms.to_initval = (tcp ? 600 : 11) * HZ / 10;
-        if (!timeparms.to_retries)
-                timeparms.to_retries = 5;
+        nfs_init_timeout_values(&timeparms, proto, data->timeo, data->retrans);
 
         /* create transport and client */
-        xprt = xprt_create_proto(tcp ? IPPROTO_TCP : IPPROTO_UDP,
-                                 &server->addr, &timeparms);
+        xprt = xprt_create_proto(proto, &server->addr, &timeparms);
         if (IS_ERR(xprt)) {
                 dprintk("%s: cannot create RPC transport. Error = %ld\n",
                                 __FUNCTION__, PTR_ERR(xprt));
@@ -576,7 +595,6 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt)
                 { NFS_MOUNT_SOFT, ",soft", ",hard" },
                 { NFS_MOUNT_INTR, ",intr", "" },
                 { NFS_MOUNT_POSIX, ",posix", "" },
-                { NFS_MOUNT_TCP, ",tcp", ",udp" },
                 { NFS_MOUNT_NOCTO, ",nocto", "" },
                 { NFS_MOUNT_NOAC, ",noac", "" },
                 { NFS_MOUNT_NONLM, ",nolock", ",lock" },
@@ -585,6 +603,8 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt)
         };
         struct proc_nfs_info *nfs_infop;
         struct nfs_server *nfss = NFS_SB(mnt->mnt_sb);
+        char buf[12];
+        char *proto;
 
         seq_printf(m, ",v%d", nfss->rpc_ops->version);
         seq_printf(m, ",rsize=%d", nfss->rsize);
@@ -603,6 +623,18 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt)
                 else
                         seq_puts(m, nfs_infop->nostr);
         }
+        switch (nfss->client->cl_xprt->prot) {
+                case IPPROTO_TCP:
+                        proto = "tcp";
+                        break;
+                case IPPROTO_UDP:
+                        proto = "udp";
+                        break;
+                default:
+                        snprintf(buf, sizeof(buf), "%u", nfss->client->cl_xprt->prot);
+                        proto = buf;
+        }
+        seq_printf(m, ",proto=%s", proto);
         seq_puts(m, ",addr=");
         seq_escape(m, nfss->hostname, " \t\n\\");
         return 0;
@@ -753,7 +785,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
                 else
                         init_special_inode(inode, inode->i_mode, fattr->rdev);
 
-                nfsi->read_cache_jiffies = fattr->timestamp;
+                nfsi->read_cache_jiffies = fattr->time_start;
+                nfsi->last_updated = jiffies;
                 inode->i_atime = fattr->atime;
                 inode->i_mtime = fattr->mtime;
                 inode->i_ctime = fattr->ctime;
@@ -821,6 +854,11 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr)
                         filemap_fdatawait(inode->i_mapping);
                 nfs_wb_all(inode);
         }
+        /*
+         * Return any delegations if we're going to change ACLs
+         */
+        if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0)
+                nfs_inode_return_delegation(inode);
         error = NFS_PROTO(inode)->setattr(dentry, &fattr, attr);
         if (error == 0)
                 nfs_refresh_inode(inode, &fattr);
@@ -1019,15 +1057,11 @@ int nfs_open(struct inode *inode, struct file *filp)
         ctx->mode = filp->f_mode;
         nfs_file_set_open_context(filp, ctx);
         put_nfs_open_context(ctx);
-        if ((filp->f_mode & FMODE_WRITE) != 0)
-                nfs_begin_data_update(inode);
         return 0;
 }
 
 int nfs_release(struct inode *inode, struct file *filp)
 {
-        if ((filp->f_mode & FMODE_WRITE) != 0)
-                nfs_end_data_update(inode);
         nfs_file_clear_open_context(filp);
         return 0;
 }
@@ -1083,14 +1117,15 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
                 goto out;
         }
 
+        spin_lock(&inode->i_lock);
         status = nfs_update_inode(inode, &fattr, verifier);
         if (status) {
+                spin_unlock(&inode->i_lock);
                 dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) refresh failed, error=%d\n",
                          inode->i_sb->s_id,
                          (long long)NFS_FILEID(inode), status);
                 goto out;
         }
-        spin_lock(&inode->i_lock);
         cache_validity = nfsi->cache_validity;
         nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;
 
@@ -1098,7 +1133,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
          * We may need to keep the attributes marked as invalid if
          * we raced with nfs_end_attr_update().
          */
-        if (verifier == nfsi->cache_change_attribute)
+        if (time_after_eq(verifier, nfsi->cache_change_attribute))
                 nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
         spin_unlock(&inode->i_lock);
 
@@ -1165,7 +1200,7 @@ void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
                 if (S_ISDIR(inode->i_mode)) {
                         memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
                         /* This ensures we revalidate child dentries */
-                        nfsi->cache_change_attribute++;
+                        nfsi->cache_change_attribute = jiffies;
                 }
                 spin_unlock(&inode->i_lock);
 
@@ -1197,20 +1232,19 @@ void nfs_end_data_update(struct inode *inode)
         struct nfs_inode *nfsi = NFS_I(inode);
 
         if (!nfs_have_delegation(inode, FMODE_READ)) {
-                /* Mark the attribute cache for revalidation */
-                spin_lock(&inode->i_lock);
-                nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
-                /* Directories and symlinks: invalidate page cache too */
-                if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
+                /* Directories and symlinks: invalidate page cache */
+                if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) {
+                        spin_lock(&inode->i_lock);
                         nfsi->cache_validity |= NFS_INO_INVALID_DATA;
-                spin_unlock(&inode->i_lock);
+                        spin_unlock(&inode->i_lock);
+                }
         }
-        nfsi->cache_change_attribute ++;
+        nfsi->cache_change_attribute = jiffies;
         atomic_dec(&nfsi->data_updates);
 }
 
 /**
- * nfs_refresh_inode - verify consistency of the inode attribute cache
+ * nfs_check_inode_attributes - verify consistency of the inode attribute cache
  * @inode - pointer to inode
  * @fattr - updated attributes
  *
@@ -1218,13 +1252,12 @@ void nfs_end_data_update(struct inode *inode)
  * so that fattr carries weak cache consistency data, then it may
  * also update the ctime/mtime/change_attribute.
  */
-int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
+static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fattr)
 {
         struct nfs_inode *nfsi = NFS_I(inode);
         loff_t cur_size, new_isize;
         int data_unstable;
 
-        spin_lock(&inode->i_lock);
 
         /* Are we in the process of updating data on the server? */
         data_unstable = nfs_caches_unstable(inode);
@@ -1241,14 +1274,12 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
         }
 
         if ((fattr->valid & NFS_ATTR_FATTR) == 0) {
-                spin_unlock(&inode->i_lock);
                 return 0;
         }
 
         /* Has the inode gone and changed behind our back? */
         if (nfsi->fileid != fattr->fileid
                         || (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) {
-                spin_unlock(&inode->i_lock);
                 return -EIO;
         }
 
@@ -1288,11 +1319,67 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
         if (!timespec_equal(&inode->i_atime, &fattr->atime))
                 nfsi->cache_validity |= NFS_INO_INVALID_ATIME;
 
-        nfsi->read_cache_jiffies = fattr->timestamp;
-        spin_unlock(&inode->i_lock);
+        nfsi->read_cache_jiffies = fattr->time_start;
         return 0;
 }
 
+/**
+ * nfs_refresh_inode - try to update the inode attribute cache
+ * @inode - pointer to inode
+ * @fattr - updated attributes
+ *
+ * Check that an RPC call that returned attributes has not overlapped with
+ * other recent updates of the inode metadata, then decide whether it is
+ * safe to do a full update of the inode attributes, or whether just to
+ * call nfs_check_inode_attributes.
+ */
+int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
+{
+        struct nfs_inode *nfsi = NFS_I(inode);
+        int status;
+
+        if ((fattr->valid & NFS_ATTR_FATTR) == 0)
+                return 0;
+        spin_lock(&inode->i_lock);
+        nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;
+        if (nfs_verify_change_attribute(inode, fattr->time_start))
+                nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
+        if (time_after(fattr->time_start, nfsi->last_updated))
+                status = nfs_update_inode(inode, fattr, fattr->time_start);
+        else
+                status = nfs_check_inode_attributes(inode, fattr);
+
+        spin_unlock(&inode->i_lock);
+        return status;
+}
+
+/**
+ * nfs_post_op_update_inode - try to update the inode attribute cache
+ * @inode - pointer to inode
+ * @fattr - updated attributes
+ *
+ * After an operation that has changed the inode metadata, mark the
+ * attribute cache as being invalid, then try to update it.
+ */
+int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr)
+{
+        struct nfs_inode *nfsi = NFS_I(inode);
+        int status = 0;
+
+        spin_lock(&inode->i_lock);
+        if (unlikely((fattr->valid & NFS_ATTR_FATTR) == 0)) {
+                nfsi->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS;
+                goto out;
+        }
+        status = nfs_update_inode(inode, fattr, fattr->time_start);
+        if (time_after_eq(fattr->time_start, nfsi->cache_change_attribute))
+                nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME|NFS_INO_REVAL_PAGECACHE);
+        nfsi->cache_change_attribute = jiffies;
+out:
+        spin_unlock(&inode->i_lock);
+        return status;
+}
+
 /*
  * Many nfs protocol calls return the new file attributes after
  * an operation.  Here we update the inode to reflect the state
@@ -1328,20 +1415,17 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
                 goto out_err;
         }
 
-        spin_lock(&inode->i_lock);
-
         /*
          * Make sure the inode's type hasn't changed.
          */
-        if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) {
-                spin_unlock(&inode->i_lock);
+        if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT))
                 goto out_changed;
-        }
 
         /*
          * Update the read time so we don't revalidate too often.
          */
-        nfsi->read_cache_jiffies = fattr->timestamp;
+        nfsi->read_cache_jiffies = fattr->time_start;
+        nfsi->last_updated = jiffies;
 
         /* Are we racing with known updates of the metadata on the server? */
         data_unstable = ! (nfs_verify_change_attribute(inode, verifier) ||
@@ -1354,7 +1438,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
                 /* Do we perhaps have any outstanding writes? */
                 if (nfsi->npages == 0) {
                         /* No, but did we race with nfs_end_data_update()? */
-                        if (verifier  ==  nfsi->cache_change_attribute) {
+                        if (time_after_eq(verifier,  nfsi->cache_change_attribute)) {
                                 inode->i_size = new_isize;
                                 invalid |= NFS_INO_INVALID_DATA;
                         }
@@ -1430,7 +1514,6 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
         if (!nfs_have_delegation(inode, FMODE_READ))
                 nfsi->cache_validity |= invalid;
 
-        spin_unlock(&inode->i_lock);
         return 0;
  out_changed:
         /*
@@ -1639,8 +1722,7 @@ static void nfs4_clear_inode(struct inode *inode)
         struct nfs_inode *nfsi = NFS_I(inode);
 
         /* If we are holding a delegation, return it! */
-        if (nfsi->delegation != NULL)
-                nfs_inode_return_delegation(inode);
+        nfs_inode_return_delegation(inode);
         /* First call standard NFS clear_inode() code */
         nfs_clear_inode(inode);
         /* Now clear out any remaining state */
@@ -1669,7 +1751,7 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data,
         struct rpc_clnt *clnt = NULL;
         struct rpc_timeout timeparms;
         rpc_authflavor_t authflavour;
-        int proto, err = -EIO;
+        int err = -EIO;
 
         sb->s_blocksize_bits = 0;
         sb->s_blocksize = 0;
@@ -1687,30 +1769,8 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data,
         server->acdirmax = data->acdirmax*HZ;
 
         server->rpc_ops = &nfs_v4_clientops;
-        /* Initialize timeout values */
-
-        timeparms.to_initval = data->timeo * HZ / 10;
-        timeparms.to_retries = data->retrans;
-        timeparms.to_exponential = 1;
-        if (!timeparms.to_retries)
-                timeparms.to_retries = 5;
 
-        proto = data->proto;
-        /* Which IP protocol do we use? */
-        switch (proto) {
-        case IPPROTO_TCP:
-                timeparms.to_maxval  = RPC_MAX_TCP_TIMEOUT;
-                if (!timeparms.to_initval)
-                        timeparms.to_initval = 600 * HZ / 10;
-                break;
-        case IPPROTO_UDP:
-                timeparms.to_maxval  = RPC_MAX_UDP_TIMEOUT;
-                if (!timeparms.to_initval)
-                        timeparms.to_initval = 11 * HZ / 10;
-                break;
-        default:
-                return -EINVAL;
-        }
+        nfs_init_timeout_values(&timeparms, data->proto, data->timeo, data->retrans);
 
         clp = nfs4_get_client(&server->addr.sin_addr);
         if (!clp) {
@@ -1735,7 +1795,7 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data,
 
         down_write(&clp->cl_sem);
         if (IS_ERR(clp->cl_rpcclient)) {
-                xprt = xprt_create_proto(proto, &server->addr, &timeparms);
+                xprt = xprt_create_proto(data->proto, &server->addr, &timeparms);
                 if (IS_ERR(xprt)) {
                         up_write(&clp->cl_sem);
                         err = PTR_ERR(xprt);
diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c
index d91b69044a4d..59049e864ca7 100644
--- a/fs/nfs/nfs2xdr.c
+++ b/fs/nfs/nfs2xdr.c
@@ -143,7 +143,6 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr)
                 fattr->mode = (fattr->mode & ~S_IFMT) | S_IFIFO;
                 fattr->rdev = 0;
         }
-        fattr->timestamp = jiffies;
         return p;
 }
 
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index edc95514046d..92c870d19ccd 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -78,7 +78,7 @@ nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("%s: call  fsinfo\n", __FUNCTION__);
-        info->fattr->valid = 0;
+        nfs_fattr_init(info->fattr);
         status = rpc_call(server->client_sys, NFS3PROC_FSINFO, fhandle, info, 0);
         dprintk("%s: reply fsinfo: %d\n", __FUNCTION__, status);
         if (!(info->fattr->valid & NFS_ATTR_FATTR)) {
@@ -98,7 +98,7 @@ nfs3_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("NFS call  getattr\n");
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call(server->client, NFS3PROC_GETATTR,
                           fhandle, fattr, 0);
         dprintk("NFS reply getattr: %d\n", status);
@@ -117,7 +117,7 @@ nfs3_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
         int     status;
 
         dprintk("NFS call  setattr\n");
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call(NFS_CLIENT(inode), NFS3PROC_SETATTR, &arg, fattr, 0);
         if (status == 0)
                 nfs_setattr_update_inode(inode, sattr);
@@ -143,8 +143,8 @@ nfs3_proc_lookup(struct inode *dir, struct qstr *name,
         int                     status;
 
         dprintk("NFS call  lookup %s\n", name->name);
-        dir_attr.valid = 0;
-        fattr->valid = 0;
+        nfs_fattr_init(&dir_attr);
+        nfs_fattr_init(fattr);
         status = rpc_call(NFS_CLIENT(dir), NFS3PROC_LOOKUP, &arg, &res, 0);
         if (status >= 0 && !(fattr->valid & NFS_ATTR_FATTR))
                 status = rpc_call(NFS_CLIENT(dir), NFS3PROC_GETATTR,
@@ -174,7 +174,6 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
         int status;
 
         dprintk("NFS call  access\n");
-        fattr.valid = 0;
 
         if (mode & MAY_READ)
                 arg.access |= NFS3_ACCESS_READ;
@@ -189,6 +188,7 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry)
                 if (mode & MAY_EXEC)
                         arg.access |= NFS3_ACCESS_EXECUTE;
         }
+        nfs_fattr_init(&fattr);
         status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
         nfs_refresh_inode(inode, &fattr);
         if (status == 0) {
@@ -217,7 +217,7 @@ static int nfs3_proc_readlink(struct inode *inode, struct page *page,
         int                     status;
 
         dprintk("NFS call  readlink\n");
-        fattr.valid = 0;
+        nfs_fattr_init(&fattr);
         status = rpc_call(NFS_CLIENT(inode), NFS3PROC_READLINK,
                           &args, &fattr, 0);
         nfs_refresh_inode(inode, &fattr);
@@ -240,7 +240,7 @@ static int nfs3_proc_read(struct nfs_read_data *rdata)
 
         dprintk("NFS call  read %d @ %Ld\n", rdata->args.count,
                         (long long) rdata->args.offset);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(NFS_CLIENT(inode), &msg, flags);
         if (status >= 0)
                 nfs_refresh_inode(inode, fattr);
@@ -263,10 +263,10 @@ static int nfs3_proc_write(struct nfs_write_data *wdata)
 
         dprintk("NFS call  write %d @ %Ld\n", wdata->args.count,
                         (long long) wdata->args.offset);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(NFS_CLIENT(inode), &msg, rpcflags);
         if (status >= 0)
-                nfs_refresh_inode(inode, fattr);
+                nfs_post_op_update_inode(inode, fattr);
         dprintk("NFS reply write: %d\n", status);
         return status < 0? status : wdata->res.count;
 }
@@ -285,10 +285,10 @@ static int nfs3_proc_commit(struct nfs_write_data *cdata)
 
         dprintk("NFS call  commit %d @ %Ld\n", cdata->args.count,
                         (long long) cdata->args.offset);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
         if (status >= 0)
-                nfs_refresh_inode(inode, fattr);
+                nfs_post_op_update_inode(inode, fattr);
         dprintk("NFS reply commit: %d\n", status);
         return status;
 }
@@ -299,7 +299,7 @@ static int nfs3_proc_commit(struct nfs_write_data *cdata)
  */
 static int
 nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
-                 int flags)
+                 int flags, struct nameidata *nd)
 {
         struct nfs_fh           fhandle;
         struct nfs_fattr        fattr;
@@ -329,10 +329,10 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
         sattr->ia_mode &= ~current->fs->umask;
 
 again:
-        dir_attr.valid = 0;
-        fattr.valid = 0;
+        nfs_fattr_init(&dir_attr);
+        nfs_fattr_init(&fattr);
         status = rpc_call(NFS_CLIENT(dir), NFS3PROC_CREATE, &arg, &res, 0);
-        nfs_refresh_inode(dir, &dir_attr);
+        nfs_post_op_update_inode(dir, &dir_attr);
 
         /* If the server doesn't support the exclusive creation semantics,
          * try again with simple 'guarded' mode. */
@@ -401,9 +401,9 @@ nfs3_proc_remove(struct inode *dir, struct qstr *name)
         int                     status;
 
         dprintk("NFS call  remove %s\n", name->name);
-        dir_attr.valid = 0;
+        nfs_fattr_init(&dir_attr);
         status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
-        nfs_refresh_inode(dir, &dir_attr);
+        nfs_post_op_update_inode(dir, &dir_attr);
         dprintk("NFS reply remove: %d\n", status);
         return status;
 }
@@ -422,7 +422,7 @@ nfs3_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr
         ptr->arg.fh = NFS_FH(dir->d_inode);
         ptr->arg.name = name->name;
         ptr->arg.len = name->len;
-        ptr->res.valid = 0;
+        nfs_fattr_init(&ptr->res);
         msg->rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE];
         msg->rpc_argp = &ptr->arg;
         msg->rpc_resp = &ptr->res;
@@ -439,7 +439,7 @@ nfs3_proc_unlink_done(struct dentry *dir, struct rpc_task *task)
                 return 1;
         if (msg->rpc_argp) {
                 dir_attr = (struct nfs_fattr*)msg->rpc_resp;
-                nfs_refresh_inode(dir->d_inode, dir_attr);
+                nfs_post_op_update_inode(dir->d_inode, dir_attr);
                 kfree(msg->rpc_argp);
         }
         return 0;
@@ -465,11 +465,11 @@ nfs3_proc_rename(struct inode *old_dir, struct qstr *old_name,
         int                     status;
 
         dprintk("NFS call  rename %s -> %s\n", old_name->name, new_name->name);
-        old_dir_attr.valid = 0;
-        new_dir_attr.valid = 0;
+        nfs_fattr_init(&old_dir_attr);
+        nfs_fattr_init(&new_dir_attr);
         status = rpc_call(NFS_CLIENT(old_dir), NFS3PROC_RENAME, &arg, &res, 0);
-        nfs_refresh_inode(old_dir, &old_dir_attr);
-        nfs_refresh_inode(new_dir, &new_dir_attr);
+        nfs_post_op_update_inode(old_dir, &old_dir_attr);
+        nfs_post_op_update_inode(new_dir, &new_dir_attr);
         dprintk("NFS reply rename: %d\n", status);
         return status;
 }
@@ -491,11 +491,11 @@ nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
         int                     status;
 
         dprintk("NFS call  link %s\n", name->name);
-        dir_attr.valid = 0;
-        fattr.valid = 0;
+        nfs_fattr_init(&dir_attr);
+        nfs_fattr_init(&fattr);
         status = rpc_call(NFS_CLIENT(inode), NFS3PROC_LINK, &arg, &res, 0);
-        nfs_refresh_inode(dir, &dir_attr);
-        nfs_refresh_inode(inode, &fattr);
+        nfs_post_op_update_inode(dir, &dir_attr);
+        nfs_post_op_update_inode(inode, &fattr);
         dprintk("NFS reply link: %d\n", status);
         return status;
 }
@@ -524,10 +524,10 @@ nfs3_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path,
         if (path->len > NFS3_MAXPATHLEN)
                 return -ENAMETOOLONG;
         dprintk("NFS call  symlink %s -> %s\n", name->name, path->name);
-        dir_attr.valid = 0;
-        fattr->valid = 0;
+        nfs_fattr_init(&dir_attr);
+        nfs_fattr_init(fattr);
         status = rpc_call(NFS_CLIENT(dir), NFS3PROC_SYMLINK, &arg, &res, 0);
-        nfs_refresh_inode(dir, &dir_attr);
+        nfs_post_op_update_inode(dir, &dir_attr);
         dprintk("NFS reply symlink: %d\n", status);
         return status;
 }
@@ -552,13 +552,13 @@ nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr)
         int status;
 
         dprintk("NFS call  mkdir %s\n", dentry->d_name.name);
-        dir_attr.valid = 0;
-        fattr.valid = 0;
 
         sattr->ia_mode &= ~current->fs->umask;
 
+        nfs_fattr_init(&dir_attr);
+        nfs_fattr_init(&fattr);
         status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKDIR, &arg, &res, 0);
-        nfs_refresh_inode(dir, &dir_attr);
+        nfs_post_op_update_inode(dir, &dir_attr);
         if (status != 0)
                 goto out;
         status = nfs_instantiate(dentry, &fhandle, &fattr);
@@ -582,9 +582,9 @@ nfs3_proc_rmdir(struct inode *dir, struct qstr *name)
         int                     status;
 
         dprintk("NFS call  rmdir %s\n", name->name);
-        dir_attr.valid = 0;
+        nfs_fattr_init(&dir_attr);
         status = rpc_call(NFS_CLIENT(dir), NFS3PROC_RMDIR, &arg, &dir_attr, 0);
-        nfs_refresh_inode(dir, &dir_attr);
+        nfs_post_op_update_inode(dir, &dir_attr);
         dprintk("NFS reply rmdir: %d\n", status);
         return status;
 }
@@ -634,7 +634,7 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
         dprintk("NFS call  readdir%s %d\n",
                         plus? "plus" : "", (unsigned int) cookie);
 
-        dir_attr.valid = 0;
+        nfs_fattr_init(&dir_attr);
         status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
         nfs_refresh_inode(dir, &dir_attr);
         dprintk("NFS reply readdir: %d\n", status);
@@ -676,10 +676,10 @@ nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 
         sattr->ia_mode &= ~current->fs->umask;
 
-        dir_attr.valid = 0;
-        fattr.valid = 0;
+        nfs_fattr_init(&dir_attr);
+        nfs_fattr_init(&fattr);
         status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKNOD, &arg, &res, 0);
-        nfs_refresh_inode(dir, &dir_attr);
+        nfs_post_op_update_inode(dir, &dir_attr);
         if (status != 0)
                 goto out;
         status = nfs_instantiate(dentry, &fh, &fattr);
@@ -698,7 +698,7 @@ nfs3_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("NFS call  fsstat\n");
-        stat->fattr->valid = 0;
+        nfs_fattr_init(stat->fattr);
         status = rpc_call(server->client, NFS3PROC_FSSTAT, fhandle, stat, 0);
         dprintk("NFS reply statfs: %d\n", status);
         return status;
@@ -711,7 +711,7 @@ nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("NFS call  fsinfo\n");
-        info->fattr->valid = 0;
+        nfs_fattr_init(info->fattr);
         status = rpc_call(server->client_sys, NFS3PROC_FSINFO, fhandle, info, 0);
         dprintk("NFS reply fsinfo: %d\n", status);
         return status;
@@ -724,7 +724,7 @@ nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("NFS call  pathconf\n");
-        info->fattr->valid = 0;
+        nfs_fattr_init(info->fattr);
         status = rpc_call(server->client, NFS3PROC_PATHCONF, fhandle, info, 0);
         dprintk("NFS reply pathconf: %d\n", status);
         return status;
@@ -735,7 +735,7 @@ extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int);
 static void
 nfs3_read_done(struct rpc_task *task)
 {
-        struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata;
+        struct nfs_read_data *data = (struct nfs_read_data *) task->tk_calldata;
 
         if (nfs3_async_handle_jukebox(task))
                 return;
@@ -775,7 +775,7 @@ nfs3_write_done(struct rpc_task *task)
                 return;
         data = (struct nfs_write_data *)task->tk_calldata;
         if (task->tk_status >= 0)
-                nfs_refresh_inode(data->inode, data->res.fattr);
+                nfs_post_op_update_inode(data->inode, data->res.fattr);
         nfs_writeback_done(task);
 }
 
@@ -819,7 +819,7 @@ nfs3_commit_done(struct rpc_task *task)
                 return;
         data = (struct nfs_write_data *)task->tk_calldata;
         if (task->tk_status >= 0)
-                nfs_refresh_inode(data->inode, data->res.fattr);
+                nfs_post_op_update_inode(data->inode, data->res.fattr);
         nfs_commit_done(task);
 }
 
diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c
index db4a904810a4..0498bd36602c 100644
--- a/fs/nfs/nfs3xdr.c
+++ b/fs/nfs/nfs3xdr.c
@@ -174,7 +174,6 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr)
 
         /* Update the mode bits */
         fattr->valid |= (NFS_ATTR_FATTR | NFS_ATTR_FATTR_V3);
-        fattr->timestamp = jiffies;
         return p;
 }
 
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index ec1a22d7b876..78a53f5a9f18 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -93,25 +93,50 @@ struct nfs4_client {
 };
 
 /*
+ * struct rpc_sequence ensures that RPC calls are sent in the exact
+ * order that they appear on the list.
+ */
+struct rpc_sequence {
+        struct rpc_wait_queue   wait;   /* RPC call delay queue */
+        spinlock_t lock;                /* Protects the list */
+        struct list_head list;          /* Defines sequence of RPC calls */
+};
+
+#define NFS_SEQID_CONFIRMED 1
+struct nfs_seqid_counter {
+        struct rpc_sequence *sequence;
+        int flags;
+        u32 counter;
+};
+
+struct nfs_seqid {
+        struct nfs_seqid_counter *sequence;
+        struct list_head list;
+};
+
+static inline void nfs_confirm_seqid(struct nfs_seqid_counter *seqid, int status)
+{
+        if (seqid_mutating_err(-status))
+                seqid->flags |= NFS_SEQID_CONFIRMED;
+}
+
+/*
  * NFS4 state_owners and lock_owners are simply labels for ordered
  * sequences of RPC calls. Their sole purpose is to provide once-only
  * semantics by allowing the server to identify replayed requests.
- *
- * The ->so_sema is held during all state_owner seqid-mutating operations:
- * OPEN, OPEN_DOWNGRADE, and CLOSE. Its purpose is to properly serialize
- * so_seqid.
  */
 struct nfs4_state_owner {
+        spinlock_t           so_lock;
         struct list_head     so_list;    /* per-clientid list of state_owners */
         struct nfs4_client   *so_client;
         u32                  so_id;      /* 32-bit identifier, unique */
-        struct semaphore     so_sema;
-        u32                  so_seqid;   /* protected by so_sema */
         atomic_t             so_count;
 
         struct rpc_cred      *so_cred;   /* Associated cred */
         struct list_head     so_states;
         struct list_head     so_delegations;
+        struct nfs_seqid_counter so_seqid;
+        struct rpc_sequence  so_sequence;
 };
 
 /*
@@ -132,7 +157,7 @@ struct nfs4_lock_state {
         fl_owner_t              ls_owner;       /* POSIX lock owner */
 #define NFS_LOCK_INITIALIZED 1
         int                     ls_flags;
-        u32                     ls_seqid;
+        struct nfs_seqid_counter        ls_seqid;
         u32                     ls_id;
         nfs4_stateid            ls_stateid;
         atomic_t                ls_count;
@@ -153,7 +178,6 @@ struct nfs4_state {
         struct inode *inode;            /* Pointer to the inode */
 
         unsigned long flags;            /* Do we hold any locks? */
-        struct semaphore lock_sema;     /* Serializes file locking operations */
         spinlock_t state_lock;          /* Protects the lock_states list */
 
         nfs4_stateid stateid;
@@ -191,8 +215,8 @@ extern int nfs4_proc_setclientid_confirm(struct nfs4_client *);
 extern int nfs4_proc_async_renew(struct nfs4_client *);
 extern int nfs4_proc_renew(struct nfs4_client *);
 extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode);
-extern struct inode *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
-extern int nfs4_open_revalidate(struct inode *, struct dentry *, int);
+extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
+extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
 
 extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
 extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops;
@@ -224,12 +248,17 @@ extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state
 extern void nfs4_put_open_state(struct nfs4_state *);
 extern void nfs4_close_state(struct nfs4_state *, mode_t);
 extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode);
-extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp);
 extern void nfs4_schedule_state_recovery(struct nfs4_client *);
+extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
-extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
 
+extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter);
+extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task);
+extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid);
+extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid);
+extern void nfs_free_seqid(struct nfs_seqid *seqid);
+
 extern const nfs4_stateid zero_stateid;
 
 /* nfs4xdr.c */
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 9701ca8c9428..933e13b383f8 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -47,6 +47,7 @@
 #include <linux/nfs_page.h>
 #include <linux/smp_lock.h>
 #include <linux/namei.h>
+#include <linux/mount.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
@@ -56,10 +57,11 @@
 #define NFS4_POLL_RETRY_MIN     (1*HZ)
 #define NFS4_POLL_RETRY_MAX     (15*HZ)
 
+static int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nfs_fh *fh, struct nfs4_state_owner *sp, nfs4_stateid *stateid, struct nfs_seqid *seqid);
 static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
-static int nfs4_async_handle_error(struct rpc_task *, struct nfs_server *);
+static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *);
 static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry);
-static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception);
+static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception);
 extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus);
 extern struct rpc_procinfo nfs4_procedures[];
 
@@ -185,8 +187,26 @@ static void update_changeattr(struct inode *inode, struct nfs4_change_info *cinf
 {
         struct nfs_inode *nfsi = NFS_I(inode);
 
+        spin_lock(&inode->i_lock);
+        nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
         if (cinfo->before == nfsi->change_attr && cinfo->atomic)
                 nfsi->change_attr = cinfo->after;
+        spin_unlock(&inode->i_lock);
+}
+
+/* Helper for asynchronous RPC calls */
+static int nfs4_call_async(struct rpc_clnt *clnt, rpc_action tk_begin,
+                rpc_action tk_exit, void *calldata)
+{
+        struct rpc_task *task;
+
+        if (!(task = rpc_new_task(clnt, tk_exit, RPC_TASK_ASYNC)))
+                return -ENOMEM;
+
+        task->tk_calldata = calldata;
+        task->tk_action = tk_begin;
+        rpc_execute(task);
+        return 0;
 }
 
 static void update_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid, int open_flags)
@@ -195,6 +215,7 @@ static void update_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid,
 
         open_flags &= (FMODE_READ|FMODE_WRITE);
         /* Protect against nfs4_find_state() */
+        spin_lock(&state->owner->so_lock);
         spin_lock(&inode->i_lock);
         state->state |= open_flags;
         /* NB! List reordering - see the reclaim code for why.  */
@@ -204,12 +225,12 @@ static void update_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid,
                 state->nreaders++;
         memcpy(&state->stateid, stateid, sizeof(state->stateid));
         spin_unlock(&inode->i_lock);
+        spin_unlock(&state->owner->so_lock);
 }
 
 /*
  * OPEN_RECLAIM:
  *      reclaim state on the server after a reboot.
- *      Assumes caller is holding the sp->so_sem
  */
 static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *state)
 {
@@ -218,7 +239,6 @@ static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *st
         struct nfs_delegation *delegation = NFS_I(inode)->delegation;
         struct nfs_openargs o_arg = {
                 .fh = NFS_FH(inode),
-                .seqid = sp->so_seqid,
                 .id = sp->so_id,
                 .open_flags = state->state,
                 .clientid = server->nfs4_state->cl_clientid,
@@ -245,8 +265,13 @@ static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *st
                 }
                 o_arg.u.delegation_type = delegation->type;
         }
+        o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid);
+        if (o_arg.seqid == NULL)
+                return -ENOMEM;
         status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
-        nfs4_increment_seqid(status, sp);
+        /* Confirm the sequence as being established */
+        nfs_confirm_seqid(&sp->so_seqid, status);
+        nfs_increment_open_seqid(status, o_arg.seqid);
         if (status == 0) {
                 memcpy(&state->stateid, &o_res.stateid, sizeof(state->stateid));
                 if (o_res.delegation_type != 0) {
@@ -256,6 +281,7 @@ static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *st
                                 nfs_async_inode_return_delegation(inode, &o_res.stateid);
                 }
         }
+        nfs_free_seqid(o_arg.seqid);
         clear_bit(NFS_DELEGATED_STATE, &state->flags);
         /* Ensure we update the inode attributes */
         NFS_CACHEINV(inode);
@@ -302,23 +328,35 @@ static int _nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state
         };
         int status = 0;
 
-        down(&sp->so_sema);
         if (!test_bit(NFS_DELEGATED_STATE, &state->flags))
                 goto out;
         if (state->state == 0)
                 goto out;
-        arg.seqid = sp->so_seqid;
+        arg.seqid = nfs_alloc_seqid(&sp->so_seqid);
+        status = -ENOMEM;
+        if (arg.seqid == NULL)
+                goto out;
         arg.open_flags = state->state;
         memcpy(arg.u.delegation.data, state->stateid.data, sizeof(arg.u.delegation.data));
         status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
-        nfs4_increment_seqid(status, sp);
+        nfs_increment_open_seqid(status, arg.seqid);
+        if (status != 0)
+                goto out_free;
+        if(res.rflags & NFS4_OPEN_RESULT_CONFIRM) {
+                status = _nfs4_proc_open_confirm(server->client, NFS_FH(inode),
+                                sp, &res.stateid, arg.seqid);
+                if (status != 0)
+                        goto out_free;
+        }
+        nfs_confirm_seqid(&sp->so_seqid, 0);
         if (status >= 0) {
                 memcpy(state->stateid.data, res.stateid.data,
                                 sizeof(state->stateid.data));
                 clear_bit(NFS_DELEGATED_STATE, &state->flags);
         }
+out_free:
+        nfs_free_seqid(arg.seqid);
 out:
-        up(&sp->so_sema);
         dput(parent);
         return status;
 }
@@ -345,11 +383,11 @@ int nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state *state)
         return err;
 }
 
-static inline int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nfs_fh *fh, struct nfs4_state_owner *sp, nfs4_stateid *stateid)
+static int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nfs_fh *fh, struct nfs4_state_owner *sp, nfs4_stateid *stateid, struct nfs_seqid *seqid)
 {
         struct nfs_open_confirmargs arg = {
                 .fh             = fh,
-                .seqid          = sp->so_seqid,
+                .seqid          = seqid,
                 .stateid        = *stateid,
         };
         struct nfs_open_confirmres res;
@@ -362,7 +400,9 @@ static inline int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nf
         int status;
 
         status = rpc_call_sync(clnt, &msg, RPC_TASK_NOINTR);
-        nfs4_increment_seqid(status, sp);
+        /* Confirm the sequence as being established */
+        nfs_confirm_seqid(&sp->so_seqid, status);
+        nfs_increment_open_seqid(status, seqid);
         if (status >= 0)
                 memcpy(stateid, &res.stateid, sizeof(*stateid));
         return status;
@@ -380,21 +420,41 @@ static int _nfs4_proc_open(struct inode *dir, struct nfs4_state_owner  *sp, stru
         int status;
 
         /* Update sequence id. The caller must serialize! */
-        o_arg->seqid = sp->so_seqid;
         o_arg->id = sp->so_id;
         o_arg->clientid = sp->so_client->cl_clientid;
 
         status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
-        nfs4_increment_seqid(status, sp);
+        if (status == 0) {
+                /* OPEN on anything except a regular file is disallowed in NFSv4 */
+                switch (o_res->f_attr->mode & S_IFMT) {
+                        case S_IFREG:
+                                break;
+                        case S_IFLNK:
+                                status = -ELOOP;
+                                break;
+                        case S_IFDIR:
+                                status = -EISDIR;
+                                break;
+                        default:
+                                status = -ENOTDIR;
+                }
+        }
+
+        nfs_increment_open_seqid(status, o_arg->seqid);
         if (status != 0)
                 goto out;
-        update_changeattr(dir, &o_res->cinfo);
+        if (o_arg->open_flags & O_CREAT) {
+                update_changeattr(dir, &o_res->cinfo);
+                nfs_post_op_update_inode(dir, o_res->dir_attr);
+        } else
+                nfs_refresh_inode(dir, o_res->dir_attr);
         if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) {
                 status = _nfs4_proc_open_confirm(server->client, &o_res->fh,
-                                sp, &o_res->stateid);
+                                sp, &o_res->stateid, o_arg->seqid);
                 if (status != 0)
                         goto out;
         }
+        nfs_confirm_seqid(&sp->so_seqid, 0);
         if (!(o_res->f_attr->valid & NFS_ATTR_FATTR))
                 status = server->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr);
 out:
@@ -441,9 +501,7 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st
         struct inode *inode = state->inode;
         struct nfs_server *server = NFS_SERVER(dir);
         struct nfs_delegation *delegation = NFS_I(inode)->delegation;
-        struct nfs_fattr        f_attr = {
-                .valid = 0,
-        };
+        struct nfs_fattr f_attr, dir_attr;
         struct nfs_openargs o_arg = {
                 .fh = NFS_FH(dir),
                 .open_flags = state->state,
@@ -453,6 +511,7 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st
         };
         struct nfs_openres o_res = {
                 .f_attr = &f_attr,
+                .dir_attr = &dir_attr,
                 .server = server,
         };
         int status = 0;
@@ -465,6 +524,12 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st
                 set_bit(NFS_DELEGATED_STATE, &state->flags);
                 goto out;
         }
+        o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid);
+        status = -ENOMEM;
+        if (o_arg.seqid == NULL)
+                goto out;
+        nfs_fattr_init(&f_attr);
+        nfs_fattr_init(&dir_attr);
         status = _nfs4_proc_open(dir, sp, &o_arg, &o_res);
         if (status != 0)
                 goto out_nodeleg;
@@ -490,6 +555,7 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st
                         nfs_inode_reclaim_delegation(inode, sp->so_cred, &o_res);
         }
 out_nodeleg:
+        nfs_free_seqid(o_arg.seqid);
         clear_bit(NFS_DELEGATED_STATE, &state->flags);
 out:
         dput(parent);
@@ -564,7 +630,6 @@ static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred
                 dprintk("%s: nfs4_get_state_owner failed!\n", __FUNCTION__);
                 goto out_err;
         }
-        down(&sp->so_sema);
         state = nfs4_get_open_state(inode, sp);
         if (state == NULL)
                 goto out_err;
@@ -589,7 +654,6 @@ static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred
         set_bit(NFS_DELEGATED_STATE, &state->flags);
         update_open_stateid(state, &delegation->stateid, open_flags);
 out_ok:
-        up(&sp->so_sema);
         nfs4_put_state_owner(sp);
         up_read(&nfsi->rwsem);
         up_read(&clp->cl_sem);
@@ -600,11 +664,12 @@ out_err:
         if (sp != NULL) {
                 if (state != NULL)
                         nfs4_put_open_state(state);
-                up(&sp->so_sema);
                 nfs4_put_state_owner(sp);
         }
         up_read(&nfsi->rwsem);
         up_read(&clp->cl_sem);
+        if (err != -EACCES)
+                nfs_inode_return_delegation(inode);
         return err;
 }
 
@@ -635,9 +700,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st
         struct nfs4_client *clp = server->nfs4_state;
         struct inode *inode = NULL;
         int                     status;
-        struct nfs_fattr        f_attr = {
-                .valid          = 0,
-        };
+        struct nfs_fattr f_attr, dir_attr;
         struct nfs_openargs o_arg = {
                 .fh             = NFS_FH(dir),
                 .open_flags     = flags,
@@ -648,6 +711,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st
         };
         struct nfs_openres o_res = {
                 .f_attr         = &f_attr,
+                .dir_attr       = &dir_attr,
                 .server         = server,
         };
 
@@ -665,8 +729,12 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st
         } else
                 o_arg.u.attrs = sattr;
         /* Serialization for the sequence id */
-        down(&sp->so_sema);
 
+        o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid);
+        if (o_arg.seqid == NULL)
+                return -ENOMEM;
+        nfs_fattr_init(&f_attr);
+        nfs_fattr_init(&dir_attr);
         status = _nfs4_proc_open(dir, sp, &o_arg, &o_res);
         if (status != 0)
                 goto out_err;
@@ -681,7 +749,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st
         update_open_stateid(state, &o_res.stateid, flags);
         if (o_res.delegation_type != 0)
                 nfs_inode_set_delegation(inode, cred, &o_res);
-        up(&sp->so_sema);
+        nfs_free_seqid(o_arg.seqid);
         nfs4_put_state_owner(sp);
         up_read(&clp->cl_sem);
         *res = state;
@@ -690,7 +758,7 @@ out_err:
         if (sp != NULL) {
                 if (state != NULL)
                         nfs4_put_open_state(state);
-                up(&sp->so_sema);
+                nfs_free_seqid(o_arg.seqid);
                 nfs4_put_state_owner(sp);
         }
         /* Note: clp->cl_sem must be released before nfs4_put_open_state()! */
@@ -718,7 +786,7 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir, struct dentry *dentry,
                  * It is actually a sign of a bug on the client or on the server.
                  *
                  * If we receive a BAD_SEQID error in the particular case of
-                 * doing an OPEN, we assume that nfs4_increment_seqid() will
+                 * doing an OPEN, we assume that nfs_increment_open_seqid() will
                  * have unhashed the old state_owner for us, and that we can
                  * therefore safely retry using a new one. We should still warn
                  * the user though...
@@ -728,6 +796,16 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir, struct dentry *dentry,
                         exception.retry = 1;
                         continue;
                 }
+                /*
+                 * BAD_STATEID on OPEN means that the server cancelled our
+                 * state before it received the OPEN_CONFIRM.
+                 * Recover by retrying the request as per the discussion
+                 * on Page 181 of RFC3530.
+                 */
+                if (status == -NFS4ERR_BAD_STATEID) {
+                        exception.retry = 1;
+                        continue;
+                }
                 res = ERR_PTR(nfs4_handle_exception(NFS_SERVER(dir),
                                         status, &exception));
         } while (exception.retry);
@@ -755,7 +833,7 @@ static int _nfs4_do_setattr(struct nfs_server *server, struct nfs_fattr *fattr,
         };
         int status;
 
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
 
         if (state != NULL) {
                 msg.rpc_cred = state->owner->so_cred;
@@ -787,19 +865,30 @@ struct nfs4_closedata {
         struct nfs4_state *state;
         struct nfs_closeargs arg;
         struct nfs_closeres res;
+        struct nfs_fattr fattr;
 };
 
+static void nfs4_free_closedata(struct nfs4_closedata *calldata)
+{
+        struct nfs4_state *state = calldata->state;
+        struct nfs4_state_owner *sp = state->owner;
+
+        nfs4_put_open_state(calldata->state);
+        nfs_free_seqid(calldata->arg.seqid);
+        nfs4_put_state_owner(sp);
+        kfree(calldata);
+}
+
 static void nfs4_close_done(struct rpc_task *task)
 {
         struct nfs4_closedata *calldata = (struct nfs4_closedata *)task->tk_calldata;
         struct nfs4_state *state = calldata->state;
-        struct nfs4_state_owner *sp = state->owner;
         struct nfs_server *server = NFS_SERVER(calldata->inode);
 
         /* hmm. we are done with the inode, and in the process of freeing
          * the state_owner. we keep this around to process errors
          */
-        nfs4_increment_seqid(task->tk_status, sp);
+        nfs_increment_open_seqid(task->tk_status, calldata->arg.seqid);
         switch (task->tk_status) {
                 case 0:
                         memcpy(&state->stateid, &calldata->res.stateid,
@@ -816,25 +905,49 @@ static void nfs4_close_done(struct rpc_task *task)
                                 return;
                         }
         }
+        nfs_refresh_inode(calldata->inode, calldata->res.fattr);
         state->state = calldata->arg.open_flags;
-        nfs4_put_open_state(state);
-        up(&sp->so_sema);
-        nfs4_put_state_owner(sp);
-        up_read(&server->nfs4_state->cl_sem);
-        kfree(calldata);
+        nfs4_free_closedata(calldata);
 }
 
-static inline int nfs4_close_call(struct rpc_clnt *clnt, struct nfs4_closedata *calldata)
+static void nfs4_close_begin(struct rpc_task *task)
 {
+        struct nfs4_closedata *calldata = (struct nfs4_closedata *)task->tk_calldata;
+        struct nfs4_state *state = calldata->state;
         struct rpc_message msg = {
                 .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE],
                 .rpc_argp = &calldata->arg,
                 .rpc_resp = &calldata->res,
-                .rpc_cred = calldata->state->owner->so_cred,
+                .rpc_cred = state->owner->so_cred,
         };
-        if (calldata->arg.open_flags != 0)
+        int mode = 0;
+        int status;
+
+        status = nfs_wait_on_sequence(calldata->arg.seqid, task);
+        if (status != 0)
+                return;
+        /* Don't reorder reads */
+        smp_rmb();
+        /* Recalculate the new open mode in case someone reopened the file
+         * while we were waiting in line to be scheduled.
+         */
+        if (state->nreaders != 0)
+                mode |= FMODE_READ;
+        if (state->nwriters != 0)
+                mode |= FMODE_WRITE;
+        if (test_bit(NFS_DELEGATED_STATE, &state->flags))
+                state->state = mode;
+        if (mode == state->state) {
+                nfs4_free_closedata(calldata);
+                task->tk_exit = NULL;
+                rpc_exit(task, 0);
+                return;
+        }
+        nfs_fattr_init(calldata->res.fattr);
+        if (mode != 0)
                 msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
-        return rpc_call_async(clnt, &msg, 0, nfs4_close_done, calldata);
+        calldata->arg.open_flags = mode;
+        rpc_call_setup(task, &msg, 0);
 }
 
 /* 
@@ -850,40 +963,57 @@ static inline int nfs4_close_call(struct rpc_clnt *clnt, struct nfs4_closedata *
  */
 int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode) 
 {
+        struct nfs_server *server = NFS_SERVER(inode);
         struct nfs4_closedata *calldata;
-        int status;
+        int status = -ENOMEM;
 
-        /* Tell caller we're done */
-        if (test_bit(NFS_DELEGATED_STATE, &state->flags)) {
-                state->state = mode;
-                return 0;
-        }
-        calldata = (struct nfs4_closedata *)kmalloc(sizeof(*calldata), GFP_KERNEL);
+        calldata = kmalloc(sizeof(*calldata), GFP_KERNEL);
         if (calldata == NULL)
-                return -ENOMEM;
+                goto out;
         calldata->inode = inode;
         calldata->state = state;
         calldata->arg.fh = NFS_FH(inode);
+        calldata->arg.stateid = &state->stateid;
         /* Serialization for the sequence id */
-        calldata->arg.seqid = state->owner->so_seqid;
-        calldata->arg.open_flags = mode;
-        memcpy(&calldata->arg.stateid, &state->stateid,
-                        sizeof(calldata->arg.stateid));
-        status = nfs4_close_call(NFS_SERVER(inode)->client, calldata);
-        /*
-         * Return -EINPROGRESS on success in order to indicate to the
-         * caller that an asynchronous RPC call has been launched, and
-         * that it will release the semaphores on completion.
-         */
-        return (status == 0) ? -EINPROGRESS : status;
+        calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid);
+        if (calldata->arg.seqid == NULL)
+                goto out_free_calldata;
+        calldata->arg.bitmask = server->attr_bitmask;
+        calldata->res.fattr = &calldata->fattr;
+        calldata->res.server = server;
+
+        status = nfs4_call_async(server->client, nfs4_close_begin,
+                        nfs4_close_done, calldata);
+        if (status == 0)
+                goto out;
+
+        nfs_free_seqid(calldata->arg.seqid);
+out_free_calldata:
+        kfree(calldata);
+out:
+        return status;
 }
 
-struct inode *
+static void nfs4_intent_set_file(struct nameidata *nd, struct dentry *dentry, struct nfs4_state *state)
+{
+        struct file *filp;
+
+        filp = lookup_instantiate_filp(nd, dentry, NULL);
+        if (!IS_ERR(filp)) {
+                struct nfs_open_context *ctx;
+                ctx = (struct nfs_open_context *)filp->private_data;
+                ctx->state = state;
+        } else
+                nfs4_close_state(state, nd->intent.open.flags);
+}
+
+struct dentry *
 nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
         struct iattr attr;
         struct rpc_cred *cred;
         struct nfs4_state *state;
+        struct dentry *res;
 
         if (nd->flags & LOOKUP_CREATE) {
                 attr.ia_mode = nd->intent.open.create_mode;
@@ -897,16 +1027,23 @@ nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 
         cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0);
         if (IS_ERR(cred))
-                return (struct inode *)cred;
+                return (struct dentry *)cred;
         state = nfs4_do_open(dir, dentry, nd->intent.open.flags, &attr, cred);
         put_rpccred(cred);
-        if (IS_ERR(state))
-                return (struct inode *)state;
-        return state->inode;
+        if (IS_ERR(state)) {
+                if (PTR_ERR(state) == -ENOENT)
+                        d_add(dentry, NULL);
+                return (struct dentry *)state;
+        }
+        res = d_add_unique(dentry, state->inode);
+        if (res != NULL)
+                dentry = res;
+        nfs4_intent_set_file(nd, dentry, state);
+        return res;
 }
 
 int
-nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags)
+nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags, struct nameidata *nd)
 {
         struct rpc_cred *cred;
         struct nfs4_state *state;
@@ -919,18 +1056,30 @@ nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags)
         if (IS_ERR(state))
                 state = nfs4_do_open(dir, dentry, openflags, NULL, cred);
         put_rpccred(cred);
-        if (state == ERR_PTR(-ENOENT) && dentry->d_inode == 0)
-                return 1;
-        if (IS_ERR(state))
-                return 0;
+        if (IS_ERR(state)) {
+                switch (PTR_ERR(state)) {
+                        case -EPERM:
+                        case -EACCES:
+                        case -EDQUOT:
+                        case -ENOSPC:
+                        case -EROFS:
+                                lookup_instantiate_filp(nd, (struct dentry *)state, NULL);
+                                return 1;
+                        case -ENOENT:
+                                if (dentry->d_inode == NULL)
+                                        return 1;
+                }
+                goto out_drop;
+        }
         inode = state->inode;
+        iput(inode);
         if (inode == dentry->d_inode) {
-                iput(inode);
+                nfs4_intent_set_file(nd, dentry, state);
                 return 1;
         }
-        d_drop(dentry);
         nfs4_close_state(state, openflags);
-        iput(inode);
+out_drop:
+        d_drop(dentry);
         return 0;
 }
 
@@ -974,13 +1123,12 @@ static int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fh
 static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
                 struct nfs_fsinfo *info)
 {
-        struct nfs_fattr *      fattr = info->fattr;
         struct nfs4_lookup_root_arg args = {
                 .bitmask = nfs4_fattr_bitmap,
         };
         struct nfs4_lookup_res res = {
                 .server = server,
-                .fattr = fattr,
+                .fattr = info->fattr,
                 .fh = fhandle,
         };
         struct rpc_message msg = {
@@ -988,7 +1136,7 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
                 .rpc_argp = &args,
                 .rpc_resp = &res,
         };
-        fattr->valid = 0;
+        nfs_fattr_init(info->fattr);
         return rpc_call_sync(server->client, &msg, 0);
 }
 
@@ -1051,7 +1199,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
                 q.len = p - q.name;
 
                 do {
-                        fattr->valid = 0;
+                        nfs_fattr_init(fattr);
                         status = nfs4_handle_exception(server,
                                         rpc_call_sync(server->client, &msg, 0),
                                         &exception);
@@ -1088,7 +1236,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
                 .rpc_resp = &res,
         };
         
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         return rpc_call_sync(server->client, &msg, 0);
 }
 
@@ -1130,7 +1278,7 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
         struct nfs4_state *state;
         int status;
 
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         
         cred = rpcauth_lookupcred(NFS_SERVER(inode)->client->cl_auth, 0);
         if (IS_ERR(cred))
@@ -1176,7 +1324,7 @@ static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name,
                 .rpc_resp = &res,
         };
         
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         
         dprintk("NFS call  lookup %s\n", name->name);
         status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
@@ -1325,7 +1473,7 @@ static int _nfs4_proc_read(struct nfs_read_data *rdata)
         dprintk("NFS call  read %d @ %Ld\n", rdata->args.count,
                         (long long) rdata->args.offset);
 
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(server->client, &msg, flags);
         if (!status)
                 renew_lease(server, timestamp);
@@ -1362,7 +1510,7 @@ static int _nfs4_proc_write(struct nfs_write_data *wdata)
         dprintk("NFS call  write %d @ %Ld\n", wdata->args.count,
                         (long long) wdata->args.offset);
 
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(server->client, &msg, rpcflags);
         dprintk("NFS reply write: %d\n", status);
         return status;
@@ -1396,7 +1544,7 @@ static int _nfs4_proc_commit(struct nfs_write_data *cdata)
         dprintk("NFS call  commit %d @ %Ld\n", cdata->args.count,
                         (long long) cdata->args.offset);
 
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(server->client, &msg, 0);
         dprintk("NFS reply commit: %d\n", status);
         return status;
@@ -1431,7 +1579,7 @@ static int nfs4_proc_commit(struct nfs_write_data *cdata)
 
 static int
 nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
-                 int flags)
+                 int flags, struct nameidata *nd)
 {
         struct nfs4_state *state;
         struct rpc_cred *cred;
@@ -1453,24 +1601,30 @@ nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
                 struct nfs_fattr fattr;
                 status = nfs4_do_setattr(NFS_SERVER(dir), &fattr,
                                      NFS_FH(state->inode), sattr, state);
-                if (status == 0) {
+                if (status == 0)
                         nfs_setattr_update_inode(state->inode, sattr);
-                        goto out;
-                }
-        } else if (flags != 0)
-                goto out;
-        nfs4_close_state(state, flags);
+        }
+        if (status == 0 && nd != NULL && (nd->flags & LOOKUP_OPEN))
+                nfs4_intent_set_file(nd, dentry, state);
+        else
+                nfs4_close_state(state, flags);
 out:
         return status;
 }
 
 static int _nfs4_proc_remove(struct inode *dir, struct qstr *name)
 {
+        struct nfs_server *server = NFS_SERVER(dir);
         struct nfs4_remove_arg args = {
                 .fh = NFS_FH(dir),
                 .name = name,
+                .bitmask = server->attr_bitmask,
+        };
+        struct nfs_fattr dir_attr;
+        struct nfs4_remove_res  res = {
+                .server = server,
+                .dir_attr = &dir_attr,
         };
-        struct nfs4_change_info res;
         struct rpc_message msg = {
                 .rpc_proc       = &nfs4_procedures[NFSPROC4_CLNT_REMOVE],
                 .rpc_argp       = &args,
@@ -1478,9 +1632,12 @@ static int _nfs4_proc_remove(struct inode *dir, struct qstr *name)
         };
         int                     status;
 
-        status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
-        if (status == 0)
-                update_changeattr(dir, &res);
+        nfs_fattr_init(res.dir_attr);
+        status = rpc_call_sync(server->client, &msg, 0);
+        if (status == 0) {
+                update_changeattr(dir, &res.cinfo);
+                nfs_post_op_update_inode(dir, res.dir_attr);
+        }
         return status;
 }
 
@@ -1498,12 +1655,14 @@ static int nfs4_proc_remove(struct inode *dir, struct qstr *name)
 
 struct unlink_desc {
         struct nfs4_remove_arg  args;
-        struct nfs4_change_info res;
+        struct nfs4_remove_res  res;
+        struct nfs_fattr dir_attr;
 };
 
 static int nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir,
                 struct qstr *name)
 {
+        struct nfs_server *server = NFS_SERVER(dir->d_inode);
         struct unlink_desc *up;
 
         up = (struct unlink_desc *) kmalloc(sizeof(*up), GFP_KERNEL);
@@ -1512,6 +1671,9 @@ static int nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir,
         
         up->args.fh = NFS_FH(dir->d_inode);
         up->args.name = name;
+        up->args.bitmask = server->attr_bitmask;
+        up->res.server = server;
+        up->res.dir_attr = &up->dir_attr;
         
         msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE];
         msg->rpc_argp = &up->args;
@@ -1526,7 +1688,8 @@ static int nfs4_proc_unlink_done(struct dentry *dir, struct rpc_task *task)
         
         if (msg->rpc_resp != NULL) {
                 up = container_of(msg->rpc_resp, struct unlink_desc, res);
-                update_changeattr(dir->d_inode, &up->res);
+                update_changeattr(dir->d_inode, &up->res.cinfo);
+                nfs_post_op_update_inode(dir->d_inode, up->res.dir_attr);
                 kfree(up);
                 msg->rpc_resp = NULL;
                 msg->rpc_argp = NULL;
@@ -1537,13 +1700,20 @@ static int nfs4_proc_unlink_done(struct dentry *dir, struct rpc_task *task)
 static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
                 struct inode *new_dir, struct qstr *new_name)
 {
+        struct nfs_server *server = NFS_SERVER(old_dir);
         struct nfs4_rename_arg arg = {
                 .old_dir = NFS_FH(old_dir),
                 .new_dir = NFS_FH(new_dir),
                 .old_name = old_name,
                 .new_name = new_name,
+                .bitmask = server->attr_bitmask,
+        };
+        struct nfs_fattr old_fattr, new_fattr;
+        struct nfs4_rename_res res = {
+                .server = server,
+                .old_fattr = &old_fattr,
+                .new_fattr = &new_fattr,
         };
-        struct nfs4_rename_res res = { };
         struct rpc_message msg = {
                 .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME],
                 .rpc_argp = &arg,
@@ -1551,11 +1721,15 @@ static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
         };
         int                     status;
         
-        status = rpc_call_sync(NFS_CLIENT(old_dir), &msg, 0);
+        nfs_fattr_init(res.old_fattr);
+        nfs_fattr_init(res.new_fattr);
+        status = rpc_call_sync(server->client, &msg, 0);
 
         if (!status) {
                 update_changeattr(old_dir, &res.old_cinfo);
+                nfs_post_op_update_inode(old_dir, res.old_fattr);
                 update_changeattr(new_dir, &res.new_cinfo);
+                nfs_post_op_update_inode(new_dir, res.new_fattr);
         }
         return status;
 }
@@ -1576,22 +1750,34 @@ static int nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
 
 static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
 {
+        struct nfs_server *server = NFS_SERVER(inode);
         struct nfs4_link_arg arg = {
                 .fh     = NFS_FH(inode),
                 .dir_fh = NFS_FH(dir),
                 .name   = name,
+                .bitmask = server->attr_bitmask,
+        };
+        struct nfs_fattr fattr, dir_attr;
+        struct nfs4_link_res res = {
+                .server = server,
+                .fattr = &fattr,
+                .dir_attr = &dir_attr,
         };
-        struct nfs4_change_info cinfo = { };
         struct rpc_message msg = {
                 .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LINK],
                 .rpc_argp = &arg,
-                .rpc_resp = &cinfo,
+                .rpc_resp = &res,
         };
         int                     status;
 
-        status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
-        if (!status)
-                update_changeattr(dir, &cinfo);
+        nfs_fattr_init(res.fattr);
+        nfs_fattr_init(res.dir_attr);
+        status = rpc_call_sync(server->client, &msg, 0);
+        if (!status) {
+                update_changeattr(dir, &res.cinfo);
+                nfs_post_op_update_inode(dir, res.dir_attr);
+                nfs_refresh_inode(inode, res.fattr);
+        }
 
         return status;
 }
@@ -1613,6 +1799,7 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name,
                 struct nfs_fattr *fattr)
 {
         struct nfs_server *server = NFS_SERVER(dir);
+        struct nfs_fattr dir_fattr;
         struct nfs4_create_arg arg = {
                 .dir_fh = NFS_FH(dir),
                 .server = server,
@@ -1625,6 +1812,7 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name,
                 .server = server,
                 .fh = fhandle,
                 .fattr = fattr,
+                .dir_fattr = &dir_fattr,
         };
         struct rpc_message msg = {
                 .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SYMLINK],
@@ -1636,11 +1824,13 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name,
         if (path->len > NFS4_MAXPATHLEN)
                 return -ENAMETOOLONG;
         arg.u.symlink = path;
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
+        nfs_fattr_init(&dir_fattr);
         
         status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
         if (!status)
                 update_changeattr(dir, &res.dir_cinfo);
+        nfs_post_op_update_inode(dir, res.dir_fattr);
         return status;
 }
 
@@ -1664,7 +1854,7 @@ static int _nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry,
 {
         struct nfs_server *server = NFS_SERVER(dir);
         struct nfs_fh fhandle;
-        struct nfs_fattr fattr;
+        struct nfs_fattr fattr, dir_fattr;
         struct nfs4_create_arg arg = {
                 .dir_fh = NFS_FH(dir),
                 .server = server,
@@ -1677,6 +1867,7 @@ static int _nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry,
                 .server = server,
                 .fh = &fhandle,
                 .fattr = &fattr,
+                .dir_fattr = &dir_fattr,
         };
         struct rpc_message msg = {
                 .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE],
@@ -1685,11 +1876,13 @@ static int _nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry,
         };
         int                     status;
 
-        fattr.valid = 0;
+        nfs_fattr_init(&fattr);
+        nfs_fattr_init(&dir_fattr);
         
         status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
         if (!status) {
                 update_changeattr(dir, &res.dir_cinfo);
+                nfs_post_op_update_inode(dir, res.dir_fattr);
                 status = nfs_instantiate(dentry, &fhandle, &fattr);
         }
         return status;
@@ -1762,7 +1955,7 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry,
 {
         struct nfs_server *server = NFS_SERVER(dir);
         struct nfs_fh fh;
-        struct nfs_fattr fattr;
+        struct nfs_fattr fattr, dir_fattr;
         struct nfs4_create_arg arg = {
                 .dir_fh = NFS_FH(dir),
                 .server = server,
@@ -1774,6 +1967,7 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry,
                 .server = server,
                 .fh = &fh,
                 .fattr = &fattr,
+                .dir_fattr = &dir_fattr,
         };
         struct rpc_message msg = {
                 .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE],
@@ -1783,7 +1977,8 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry,
         int                     status;
         int                     mode = sattr->ia_mode;
 
-        fattr.valid = 0;
+        nfs_fattr_init(&fattr);
+        nfs_fattr_init(&dir_fattr);
 
         BUG_ON(!(sattr->ia_valid & ATTR_MODE));
         BUG_ON(!S_ISFIFO(mode) && !S_ISBLK(mode) && !S_ISCHR(mode) && !S_ISSOCK(mode));
@@ -1805,6 +2000,7 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry,
         status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
         if (status == 0) {
                 update_changeattr(dir, &res.dir_cinfo);
+                nfs_post_op_update_inode(dir, res.dir_fattr);
                 status = nfs_instantiate(dentry, &fh, &fattr);
         }
         return status;
@@ -1836,7 +2032,7 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
                 .rpc_resp = fsstat,
         };
 
-        fsstat->fattr->valid = 0;
+        nfs_fattr_init(fsstat->fattr);
         return rpc_call_sync(server->client, &msg, 0);
 }
 
@@ -1883,7 +2079,7 @@ static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, str
 
 static int nfs4_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo)
 {
-        fsinfo->fattr->valid = 0;
+        nfs_fattr_init(fsinfo->fattr);
         return nfs4_do_fsinfo(server, fhandle, fsinfo);
 }
 
@@ -1906,7 +2102,7 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle
                 return 0;
         }
 
-        pathconf->fattr->valid = 0;
+        nfs_fattr_init(pathconf->fattr);
         return rpc_call_sync(server->client, &msg, 0);
 }
 
@@ -1973,8 +2169,10 @@ nfs4_write_done(struct rpc_task *task)
                 rpc_restart_call(task);
                 return;
         }
-        if (task->tk_status >= 0)
+        if (task->tk_status >= 0) {
                 renew_lease(NFS_SERVER(inode), data->timestamp);
+                nfs_post_op_update_inode(inode, data->res.fattr);
+        }
         /* Call back common NFS writeback processing */
         nfs_writeback_done(task);
 }
@@ -1990,6 +2188,7 @@ nfs4_proc_write_setup(struct nfs_write_data *data, int how)
                 .rpc_cred = data->cred,
         };
         struct inode *inode = data->inode;
+        struct nfs_server *server = NFS_SERVER(inode);
         int stable;
         int flags;
         
@@ -2001,6 +2200,8 @@ nfs4_proc_write_setup(struct nfs_write_data *data, int how)
         } else
                 stable = NFS_UNSTABLE;
         data->args.stable = stable;
+        data->args.bitmask = server->attr_bitmask;
+        data->res.server = server;
 
         data->timestamp   = jiffies;
 
@@ -2022,6 +2223,8 @@ nfs4_commit_done(struct rpc_task *task)
                 rpc_restart_call(task);
                 return;
         }
+        if (task->tk_status >= 0)
+                nfs_post_op_update_inode(inode, data->res.fattr);
         /* Call back common NFS writeback processing */
         nfs_commit_done(task);
 }
@@ -2037,8 +2240,12 @@ nfs4_proc_commit_setup(struct nfs_write_data *data, int how)
                 .rpc_cred = data->cred,
         };      
         struct inode *inode = data->inode;
+        struct nfs_server *server = NFS_SERVER(inode);
         int flags;
         
+        data->args.bitmask = server->attr_bitmask;
+        data->res.server = server;
+
         /* Set the initial flags for the task.  */
         flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
 
@@ -2106,65 +2313,6 @@ nfs4_proc_renew(struct nfs4_client *clp)
         return 0;
 }
 
-/*
- * We will need to arrange for the VFS layer to provide an atomic open.
- * Until then, this open method is prone to inefficiency and race conditions
- * due to the lookup, potential create, and open VFS calls from sys_open()
- * placed on the wire.
- */
-static int
-nfs4_proc_file_open(struct inode *inode, struct file *filp)
-{
-        struct dentry *dentry = filp->f_dentry;
-        struct nfs_open_context *ctx;
-        struct nfs4_state *state = NULL;
-        struct rpc_cred *cred;
-        int status = -ENOMEM;
-
-        dprintk("nfs4_proc_file_open: starting on (%.*s/%.*s)\n",
-                               (int)dentry->d_parent->d_name.len,
-                               dentry->d_parent->d_name.name,
-                               (int)dentry->d_name.len, dentry->d_name.name);
-
-
-        /* Find our open stateid */
-        cred = rpcauth_lookupcred(NFS_SERVER(inode)->client->cl_auth, 0);
-        if (IS_ERR(cred))
-                return PTR_ERR(cred);
-        ctx = alloc_nfs_open_context(dentry, cred);
-        put_rpccred(cred);
-        if (unlikely(ctx == NULL))
-                return -ENOMEM;
-        status = -EIO; /* ERACE actually */
-        state = nfs4_find_state(inode, cred, filp->f_mode);
-        if (unlikely(state == NULL))
-                goto no_state;
-        ctx->state = state;
-        nfs4_close_state(state, filp->f_mode);
-        ctx->mode = filp->f_mode;
-        nfs_file_set_open_context(filp, ctx);
-        put_nfs_open_context(ctx);
-        if (filp->f_mode & FMODE_WRITE)
-                nfs_begin_data_update(inode);
-        return 0;
-no_state:
-        printk(KERN_WARNING "NFS: v4 raced in function %s\n", __FUNCTION__);
-        put_nfs_open_context(ctx);
-        return status;
-}
-
-/*
- * Release our state
- */
-static int
-nfs4_proc_file_release(struct inode *inode, struct file *filp)
-{
-        if (filp->f_mode & FMODE_WRITE)
-                nfs_end_data_update(inode);
-        nfs_file_clear_open_context(filp);
-        return 0;
-}
-
 static inline int nfs4_server_supports_acls(struct nfs_server *server)
 {
         return (server->caps & NFS_CAP_ACLS)
@@ -2285,7 +2433,7 @@ static inline ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size
                         return -ENOMEM;
                 args.acl_pages[0] = localpage;
                 args.acl_pgbase = 0;
-                args.acl_len = PAGE_SIZE;
+                resp_len = args.acl_len = PAGE_SIZE;
         } else {
                 resp_buf = buf;
                 buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
@@ -2345,6 +2493,7 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen
 
         if (!nfs4_server_supports_acls(server))
                 return -EOPNOTSUPP;
+        nfs_inode_return_delegation(inode);
         buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
         ret = rpc_call_sync(NFS_SERVER(inode)->client, &msg, 0);
         if (ret == 0)
@@ -2353,7 +2502,7 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen
 }
 
 static int
-nfs4_async_handle_error(struct rpc_task *task, struct nfs_server *server)
+nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server)
 {
         struct nfs4_client *clp = server->nfs4_state;
 
@@ -2431,7 +2580,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout)
 /* This is the error handling routine for processes that are allowed
  * to sleep.
  */
-int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
+int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
 {
         struct nfs4_client *clp = server->nfs4_state;
         int ret = errorcode;
@@ -2632,7 +2781,6 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
 
         down_read(&clp->cl_sem);
         nlo.clientid = clp->cl_clientid;
-        down(&state->lock_sema);
         status = nfs4_set_lock_state(state, request);
         if (status != 0)
                 goto out;
@@ -2659,7 +2807,6 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
                 status = 0;
         }
 out:
-        up(&state->lock_sema);
         up_read(&clp->cl_sem);
         return status;
 }
@@ -2696,79 +2843,149 @@ static int do_vfs_lock(struct file *file, struct file_lock *fl)
         return res;
 }
 
-static int _nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request)
+struct nfs4_unlockdata {
+        struct nfs_lockargs arg;
+        struct nfs_locku_opargs luargs;
+        struct nfs_lockres res;
+        struct nfs4_lock_state *lsp;
+        struct nfs_open_context *ctx;
+        atomic_t refcount;
+        struct completion completion;
+};
+
+static void nfs4_locku_release_calldata(struct nfs4_unlockdata *calldata)
 {
-        struct inode *inode = state->inode;
-        struct nfs_server *server = NFS_SERVER(inode);
-        struct nfs4_client *clp = server->nfs4_state;
-        struct nfs_lockargs arg = {
-                .fh = NFS_FH(inode),
-                .type = nfs4_lck_type(cmd, request),
-                .offset = request->fl_start,
-                .length = nfs4_lck_length(request),
-        };
-        struct nfs_lockres res = {
-                .server = server,
-        };
+        if (atomic_dec_and_test(&calldata->refcount)) {
+                nfs_free_seqid(calldata->luargs.seqid);
+                nfs4_put_lock_state(calldata->lsp);
+                put_nfs_open_context(calldata->ctx);
+                kfree(calldata);
+        }
+}
+
+static void nfs4_locku_complete(struct nfs4_unlockdata *calldata)
+{
+        complete(&calldata->completion);
+        nfs4_locku_release_calldata(calldata);
+}
+
+static void nfs4_locku_done(struct rpc_task *task)
+{
+        struct nfs4_unlockdata *calldata = (struct nfs4_unlockdata *)task->tk_calldata;
+
+        nfs_increment_lock_seqid(task->tk_status, calldata->luargs.seqid);
+        switch (task->tk_status) {
+                case 0:
+                        memcpy(calldata->lsp->ls_stateid.data,
+                                        calldata->res.u.stateid.data,
+                                        sizeof(calldata->lsp->ls_stateid.data));
+                        break;
+                case -NFS4ERR_STALE_STATEID:
+                case -NFS4ERR_EXPIRED:
+                        nfs4_schedule_state_recovery(calldata->res.server->nfs4_state);
+                        break;
+                default:
+                        if (nfs4_async_handle_error(task, calldata->res.server) == -EAGAIN) {
+                                rpc_restart_call(task);
+                                return;
+                        }
+        }
+        nfs4_locku_complete(calldata);
+}
+
+static void nfs4_locku_begin(struct rpc_task *task)
+{
+        struct nfs4_unlockdata *calldata = (struct nfs4_unlockdata *)task->tk_calldata;
         struct rpc_message msg = {
                 .rpc_proc       = &nfs4_procedures[NFSPROC4_CLNT_LOCKU],
-                .rpc_argp       = &arg,
-                .rpc_resp       = &res,
-                .rpc_cred       = state->owner->so_cred,
+                .rpc_argp       = &calldata->arg,
+                .rpc_resp       = &calldata->res,
+                .rpc_cred       = calldata->lsp->ls_state->owner->so_cred,
         };
+        int status;
+
+        status = nfs_wait_on_sequence(calldata->luargs.seqid, task);
+        if (status != 0)
+                return;
+        if ((calldata->lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0) {
+                nfs4_locku_complete(calldata);
+                task->tk_exit = NULL;
+                rpc_exit(task, 0);
+                return;
+        }
+        rpc_call_setup(task, &msg, 0);
+}
+
+static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request)
+{
+        struct nfs4_unlockdata *calldata;
+        struct inode *inode = state->inode;
+        struct nfs_server *server = NFS_SERVER(inode);
         struct nfs4_lock_state *lsp;
-        struct nfs_locku_opargs luargs;
         int status;
-                        
-        down_read(&clp->cl_sem);
-        down(&state->lock_sema);
+
         status = nfs4_set_lock_state(state, request);
         if (status != 0)
-                goto out;
+                return status;
         lsp = request->fl_u.nfs4_fl.owner;
         /* We might have lost the locks! */
         if ((lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0)
-                goto out;
-        luargs.seqid = lsp->ls_seqid;
-        memcpy(&luargs.stateid, &lsp->ls_stateid, sizeof(luargs.stateid));
-        arg.u.locku = &luargs;
-        status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
-        nfs4_increment_lock_seqid(status, lsp);
-
-        if (status == 0)
-                memcpy(&lsp->ls_stateid,  &res.u.stateid, 
-                                sizeof(lsp->ls_stateid));
-out:
-        up(&state->lock_sema);
+                return 0;
+        calldata = kmalloc(sizeof(*calldata), GFP_KERNEL);
+        if (calldata == NULL)
+                return -ENOMEM;
+        calldata->luargs.seqid = nfs_alloc_seqid(&lsp->ls_seqid);
+        if (calldata->luargs.seqid == NULL) {
+                kfree(calldata);
+                return -ENOMEM;
+        }
+        calldata->luargs.stateid = &lsp->ls_stateid;
+        calldata->arg.fh = NFS_FH(inode);
+        calldata->arg.type = nfs4_lck_type(cmd, request);
+        calldata->arg.offset = request->fl_start;
+        calldata->arg.length = nfs4_lck_length(request);
+        calldata->arg.u.locku = &calldata->luargs;
+        calldata->res.server = server;
+        calldata->lsp = lsp;
+        atomic_inc(&lsp->ls_count);
+
+        /* Ensure we don't close file until we're done freeing locks! */
+        calldata->ctx = get_nfs_open_context((struct nfs_open_context*)request->fl_file->private_data);
+
+        atomic_set(&calldata->refcount, 2);
+        init_completion(&calldata->completion);
+
+        status = nfs4_call_async(NFS_SERVER(inode)->client, nfs4_locku_begin,
+                        nfs4_locku_done, calldata);
         if (status == 0)
-                do_vfs_lock(request->fl_file, request);
-        up_read(&clp->cl_sem);
+                wait_for_completion_interruptible(&calldata->completion);
+        do_vfs_lock(request->fl_file, request);
+        nfs4_locku_release_calldata(calldata);
         return status;
 }
 
-static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request)
-{
-        struct nfs4_exception exception = { };
-        int err;
-
-        do {
-                err = nfs4_handle_exception(NFS_SERVER(state->inode),
-                                _nfs4_proc_unlck(state, cmd, request),
-                                &exception);
-        } while (exception.retry);
-        return err;
-}
-
 static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *request, int reclaim)
 {
         struct inode *inode = state->inode;
         struct nfs_server *server = NFS_SERVER(inode);
         struct nfs4_lock_state *lsp = request->fl_u.nfs4_fl.owner;
+        struct nfs_lock_opargs largs = {
+                .lock_stateid = &lsp->ls_stateid,
+                .open_stateid = &state->stateid,
+                .lock_owner = {
+                        .clientid = server->nfs4_state->cl_clientid,
+                        .id = lsp->ls_id,
+                },
+                .reclaim = reclaim,
+        };
         struct nfs_lockargs arg = {
                 .fh = NFS_FH(inode),
                 .type = nfs4_lck_type(cmd, request),
                 .offset = request->fl_start,
                 .length = nfs4_lck_length(request),
+                .u = {
+                        .lock = &largs,
+                },
         };
         struct nfs_lockres res = {
                 .server = server,
@@ -2779,53 +2996,39 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r
                 .rpc_resp       = &res,
                 .rpc_cred       = state->owner->so_cred,
         };
-        struct nfs_lock_opargs largs = {
-                .reclaim = reclaim,
-                .new_lock_owner = 0,
-        };
-        int status;
+        int status = -ENOMEM;
 
-        if (!(lsp->ls_flags & NFS_LOCK_INITIALIZED)) {
+        largs.lock_seqid = nfs_alloc_seqid(&lsp->ls_seqid);
+        if (largs.lock_seqid == NULL)
+                return -ENOMEM;
+        if (!(lsp->ls_seqid.flags & NFS_SEQID_CONFIRMED)) {
                 struct nfs4_state_owner *owner = state->owner;
-                struct nfs_open_to_lock otl = {
-                        .lock_owner = {
-                                .clientid = server->nfs4_state->cl_clientid,
-                        },
-                };
-
-                otl.lock_seqid = lsp->ls_seqid;
-                otl.lock_owner.id = lsp->ls_id;
-                memcpy(&otl.open_stateid, &state->stateid, sizeof(otl.open_stateid));
-                largs.u.open_lock = &otl;
+
+                largs.open_seqid = nfs_alloc_seqid(&owner->so_seqid);
+                if (largs.open_seqid == NULL)
+                        goto out;
                 largs.new_lock_owner = 1;
-                arg.u.lock = &largs;
-                down(&owner->so_sema);
-                otl.open_seqid = owner->so_seqid;
                 status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
-                /* increment open_owner seqid on success, and 
-                * seqid mutating errors */
-                nfs4_increment_seqid(status, owner);
-                up(&owner->so_sema);
-                if (status == 0) {
-                        lsp->ls_flags |= NFS_LOCK_INITIALIZED;
-                        lsp->ls_seqid++;
+                /* increment open seqid on success, and seqid mutating errors */
+                if (largs.new_lock_owner != 0) {
+                        nfs_increment_open_seqid(status, largs.open_seqid);
+                        if (status == 0)
+                                nfs_confirm_seqid(&lsp->ls_seqid, 0);
                 }
-        } else {
-                struct nfs_exist_lock el = {
-                        .seqid = lsp->ls_seqid,
-                };
-                memcpy(&el.stateid, &lsp->ls_stateid, sizeof(el.stateid));
-                largs.u.exist_lock = &el;
-                arg.u.lock = &largs;
+                nfs_free_seqid(largs.open_seqid);
+        } else
                 status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
-                /* increment seqid on success, and * seqid mutating errors*/
-                nfs4_increment_lock_seqid(status, lsp);
-        }
+        /* increment lock seqid on success, and seqid mutating errors*/
+        nfs_increment_lock_seqid(status, largs.lock_seqid);
         /* save the returned stateid. */
-        if (status == 0)
-                memcpy(&lsp->ls_stateid, &res.u.stateid, sizeof(nfs4_stateid));
-        else if (status == -NFS4ERR_DENIED)
+        if (status == 0) {
+                memcpy(lsp->ls_stateid.data, res.u.stateid.data,
+                                sizeof(lsp->ls_stateid.data));
+                lsp->ls_flags |= NFS_LOCK_INITIALIZED;
+        } else if (status == -NFS4ERR_DENIED)
                 status = -EAGAIN;
+out:
+        nfs_free_seqid(largs.lock_seqid);
         return status;
 }
 
@@ -2865,11 +3068,9 @@ static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock
         int status;
 
         down_read(&clp->cl_sem);
-        down(&state->lock_sema);
         status = nfs4_set_lock_state(state, request);
         if (status == 0)
                 status = _nfs4_do_setlk(state, cmd, request, 0);
-        up(&state->lock_sema);
         if (status == 0) {
                 /* Note: we always want to sleep here! */
                 request->fl_flags |= FL_SLEEP;
@@ -3024,8 +3225,8 @@ struct nfs_rpc_ops	nfs_v4_clientops = {
         .read_setup     = nfs4_proc_read_setup,
         .write_setup    = nfs4_proc_write_setup,
         .commit_setup   = nfs4_proc_commit_setup,
-        .file_open      = nfs4_proc_file_open,
-        .file_release   = nfs4_proc_file_release,
+        .file_open      = nfs_open,
+        .file_release   = nfs_release,
         .lock           = nfs4_proc_lock,
         .clear_acl_cache = nfs4_zap_acl_attr,
 };
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index afe587d82f1e..2d5a6a2b9dec 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -264,13 +264,16 @@ nfs4_alloc_state_owner(void)
 {
         struct nfs4_state_owner *sp;
 
-        sp = kmalloc(sizeof(*sp),GFP_KERNEL);
+        sp = kzalloc(sizeof(*sp),GFP_KERNEL);
         if (!sp)
                 return NULL;
-        init_MUTEX(&sp->so_sema);
-        sp->so_seqid = 0;                 /* arbitrary */
+        spin_lock_init(&sp->so_lock);
         INIT_LIST_HEAD(&sp->so_states);
         INIT_LIST_HEAD(&sp->so_delegations);
+        rpc_init_wait_queue(&sp->so_sequence.wait, "Seqid_waitqueue");
+        sp->so_seqid.sequence = &sp->so_sequence;
+        spin_lock_init(&sp->so_sequence.lock);
+        INIT_LIST_HEAD(&sp->so_sequence.list);
         atomic_set(&sp->so_count, 1);
         return sp;
 }
@@ -359,7 +362,6 @@ nfs4_alloc_open_state(void)
         memset(state->stateid.data, 0, sizeof(state->stateid.data));
         atomic_set(&state->count, 1);
         INIT_LIST_HEAD(&state->lock_states);
-        init_MUTEX(&state->lock_sema);
         spin_lock_init(&state->state_lock);
         return state;
 }
@@ -437,21 +439,23 @@ nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner)
         if (state)
                 goto out;
         new = nfs4_alloc_open_state();
+        spin_lock(&owner->so_lock);
         spin_lock(&inode->i_lock);
         state = __nfs4_find_state_byowner(inode, owner);
         if (state == NULL && new != NULL) {
                 state = new;
-                /* Caller *must* be holding owner->so_sem */
-                /* Note: The reclaim code dictates that we add stateless
-                 * and read-only stateids to the end of the list */
-                list_add_tail(&state->open_states, &owner->so_states);
                 state->owner = owner;
                 atomic_inc(&owner->so_count);
                 list_add(&state->inode_states, &nfsi->open_states);
                 state->inode = igrab(inode);
                 spin_unlock(&inode->i_lock);
+                /* Note: The reclaim code dictates that we add stateless
+                 * and read-only stateids to the end of the list */
+                list_add_tail(&state->open_states, &owner->so_states);
+                spin_unlock(&owner->so_lock);
         } else {
                 spin_unlock(&inode->i_lock);
+                spin_unlock(&owner->so_lock);
                 if (new)
                         nfs4_free_open_state(new);
         }
@@ -461,19 +465,21 @@ out:
 
 /*
  * Beware! Caller must be holding exactly one
- * reference to clp->cl_sem and owner->so_sema!
+ * reference to clp->cl_sem!
  */
 void nfs4_put_open_state(struct nfs4_state *state)
 {
         struct inode *inode = state->inode;
         struct nfs4_state_owner *owner = state->owner;
 
-        if (!atomic_dec_and_lock(&state->count, &inode->i_lock))
+        if (!atomic_dec_and_lock(&state->count, &owner->so_lock))
                 return;
+        spin_lock(&inode->i_lock);
         if (!list_empty(&state->inode_states))
                 list_del(&state->inode_states);
-        spin_unlock(&inode->i_lock);
         list_del(&state->open_states);
+        spin_unlock(&inode->i_lock);
+        spin_unlock(&owner->so_lock);
         iput(inode);
         BUG_ON (state->state != 0);
         nfs4_free_open_state(state);
@@ -481,20 +487,17 @@ void nfs4_put_open_state(struct nfs4_state *state)
 }
 
 /*
- * Beware! Caller must be holding no references to clp->cl_sem!
- * of owner->so_sema!
+ * Close the current file.
  */
 void nfs4_close_state(struct nfs4_state *state, mode_t mode)
 {
         struct inode *inode = state->inode;
         struct nfs4_state_owner *owner = state->owner;
-        struct nfs4_client *clp = owner->so_client;
         int newstate;
 
         atomic_inc(&owner->so_count);
-        down_read(&clp->cl_sem);
-        down(&owner->so_sema);
         /* Protect against nfs4_find_state() */
+        spin_lock(&owner->so_lock);
         spin_lock(&inode->i_lock);
         if (mode & FMODE_READ)
                 state->nreaders--;
@@ -507,6 +510,7 @@ void nfs4_close_state(struct nfs4_state *state, mode_t mode)
                 list_move_tail(&state->open_states, &owner->so_states);
         }
         spin_unlock(&inode->i_lock);
+        spin_unlock(&owner->so_lock);
         newstate = 0;
         if (state->state != 0) {
                 if (state->nreaders)
@@ -515,14 +519,16 @@ void nfs4_close_state(struct nfs4_state *state, mode_t mode)
                         newstate |= FMODE_WRITE;
                 if (state->state == newstate)
                         goto out;
-                if (nfs4_do_close(inode, state, newstate) == -EINPROGRESS)
+                if (test_bit(NFS_DELEGATED_STATE, &state->flags)) {
+                        state->state = newstate;
+                        goto out;
+                }
+                if (nfs4_do_close(inode, state, newstate) == 0)
                         return;
         }
 out:
         nfs4_put_open_state(state);
-        up(&owner->so_sema);
         nfs4_put_state_owner(owner);
-        up_read(&clp->cl_sem);
 }
 
 /*
@@ -546,19 +552,16 @@ __nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
  * Return a compatible lock_state. If no initialized lock_state structure
  * exists, return an uninitialized one.
  *
- * The caller must be holding state->lock_sema
  */
 static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
 {
         struct nfs4_lock_state *lsp;
         struct nfs4_client *clp = state->owner->so_client;
 
-        lsp = kmalloc(sizeof(*lsp), GFP_KERNEL);
+        lsp = kzalloc(sizeof(*lsp), GFP_KERNEL);
         if (lsp == NULL)
                 return NULL;
-        lsp->ls_flags = 0;
-        lsp->ls_seqid = 0;      /* arbitrary */
-        memset(lsp->ls_stateid.data, 0, sizeof(lsp->ls_stateid.data));
+        lsp->ls_seqid.sequence = &state->owner->so_sequence;
         atomic_set(&lsp->ls_count, 1);
         lsp->ls_owner = fl_owner;
         spin_lock(&clp->cl_lock);
@@ -572,7 +575,7 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f
  * Return a compatible lock_state. If no initialized lock_state structure
  * exists, return an uninitialized one.
  *
- * The caller must be holding state->lock_sema and clp->cl_sem
+ * The caller must be holding clp->cl_sem
  */
 static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner)
 {
@@ -605,7 +608,7 @@ static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_
  * Release reference to lock_state, and free it if we see that
  * it is no longer in use
  */
-static void nfs4_put_lock_state(struct nfs4_lock_state *lsp)
+void nfs4_put_lock_state(struct nfs4_lock_state *lsp)
 {
         struct nfs4_state *state;
 
@@ -673,29 +676,94 @@ void nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t f
         nfs4_put_lock_state(lsp);
 }
 
-/*
-* Called with state->lock_sema and clp->cl_sem held.
-*/
-void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *lsp)
+struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter)
 {
-        if (status == NFS_OK || seqid_mutating_err(-status))
-                lsp->ls_seqid++;
+        struct nfs_seqid *new;
+
+        new = kmalloc(sizeof(*new), GFP_KERNEL);
+        if (new != NULL) {
+                new->sequence = counter;
+                INIT_LIST_HEAD(&new->list);
+        }
+        return new;
+}
+
+void nfs_free_seqid(struct nfs_seqid *seqid)
+{
+        struct rpc_sequence *sequence = seqid->sequence->sequence;
+
+        if (!list_empty(&seqid->list)) {
+                spin_lock(&sequence->lock);
+                list_del(&seqid->list);
+                spin_unlock(&sequence->lock);
+        }
+        rpc_wake_up_next(&sequence->wait);
+        kfree(seqid);
 }
 
 /*
-* Called with sp->so_sema and clp->cl_sem held.
-*
-* Increment the seqid if the OPEN/OPEN_DOWNGRADE/CLOSE succeeded, or
-* failed with a seqid incrementing error -
-* see comments nfs_fs.h:seqid_mutating_error()
-*/
-void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp)
-{
-        if (status == NFS_OK || seqid_mutating_err(-status))
-                sp->so_seqid++;
-        /* If the server returns BAD_SEQID, unhash state_owner here */
-        if (status == -NFS4ERR_BAD_SEQID)
+ * Increment the seqid if the OPEN/OPEN_DOWNGRADE/CLOSE succeeded, or
+ * failed with a seqid incrementing error -
+ * see comments nfs_fs.h:seqid_mutating_error()
+ */
+static inline void nfs_increment_seqid(int status, struct nfs_seqid *seqid)
+{
+        switch (status) {
+                case 0:
+                        break;
+                case -NFS4ERR_BAD_SEQID:
+                case -NFS4ERR_STALE_CLIENTID:
+                case -NFS4ERR_STALE_STATEID:
+                case -NFS4ERR_BAD_STATEID:
+                case -NFS4ERR_BADXDR:
+                case -NFS4ERR_RESOURCE:
+                case -NFS4ERR_NOFILEHANDLE:
+                        /* Non-seqid mutating errors */
+                        return;
+        };
+        /*
+         * Note: no locking needed as we are guaranteed to be first
+         * on the sequence list
+         */
+        seqid->sequence->counter++;
+}
+
+void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid)
+{
+        if (status == -NFS4ERR_BAD_SEQID) {
+                struct nfs4_state_owner *sp = container_of(seqid->sequence,
+                                struct nfs4_state_owner, so_seqid);
                 nfs4_drop_state_owner(sp);
+        }
+        return nfs_increment_seqid(status, seqid);
+}
+
+/*
+ * Increment the seqid if the LOCK/LOCKU succeeded, or
+ * failed with a seqid incrementing error -
+ * see comments nfs_fs.h:seqid_mutating_error()
+ */
+void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid)
+{
+        return nfs_increment_seqid(status, seqid);
+}
+
+int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task)
+{
+        struct rpc_sequence *sequence = seqid->sequence->sequence;
+        int status = 0;
+
+        if (sequence->list.next == &seqid->list)
+                goto out;
+        spin_lock(&sequence->lock);
+        if (!list_empty(&sequence->list)) {
+                rpc_sleep_on(&sequence->wait, task, NULL, NULL);
+                status = -EAGAIN;
+        } else
+                list_add(&seqid->list, &sequence->list);
+        spin_unlock(&sequence->lock);
+out:
+        return status;
 }
 
 static int reclaimer(void *);
@@ -791,8 +859,6 @@ static int nfs4_reclaim_open_state(struct nfs4_state_recovery_ops *ops, struct n
                 if (state->state == 0)
                         continue;
                 status = ops->recover_open(sp, state);
-                list_for_each_entry(lock, &state->lock_states, ls_locks)
-                        lock->ls_flags &= ~NFS_LOCK_INITIALIZED;
                 if (status >= 0) {
                         status = nfs4_reclaim_locks(ops, state);
                         if (status < 0)
@@ -831,6 +897,28 @@ out_err:
         return status;
 }
 
+static void nfs4_state_mark_reclaim(struct nfs4_client *clp)
+{
+        struct nfs4_state_owner *sp;
+        struct nfs4_state *state;
+        struct nfs4_lock_state *lock;
+
+        /* Reset all sequence ids to zero */
+        list_for_each_entry(sp, &clp->cl_state_owners, so_list) {
+                sp->so_seqid.counter = 0;
+                sp->so_seqid.flags = 0;
+                spin_lock(&sp->so_lock);
+                list_for_each_entry(state, &sp->so_states, open_states) {
+                        list_for_each_entry(lock, &state->lock_states, ls_locks) {
+                                lock->ls_seqid.counter = 0;
+                                lock->ls_seqid.flags = 0;
+                                lock->ls_flags &= ~NFS_LOCK_INITIALIZED;
+                        }
+                }
+                spin_unlock(&sp->so_lock);
+        }
+}
+
 static int reclaimer(void *ptr)
 {
         struct reclaimer_args *args = (struct reclaimer_args *)ptr;
@@ -864,6 +952,7 @@ restart_loop:
                 default:
                         ops = &nfs4_network_partition_recovery_ops;
         };
+        nfs4_state_mark_reclaim(clp);
         status = __nfs4_init_client(clp);
         if (status)
                 goto out_error;
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 6c564ef9489e..fbbace8a30c4 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -95,6 +95,8 @@ static int nfs_stat_to_errno(int);
 #define decode_getattr_maxsz    (op_decode_hdr_maxsz + nfs4_fattr_maxsz)
 #define encode_savefh_maxsz     (op_encode_hdr_maxsz)
 #define decode_savefh_maxsz     (op_decode_hdr_maxsz)
+#define encode_restorefh_maxsz  (op_encode_hdr_maxsz)
+#define decode_restorefh_maxsz  (op_decode_hdr_maxsz)
 #define encode_fsinfo_maxsz     (op_encode_hdr_maxsz + 2)
 #define decode_fsinfo_maxsz     (op_decode_hdr_maxsz + 11)
 #define encode_renew_maxsz      (op_encode_hdr_maxsz + 3)
@@ -157,16 +159,20 @@ static int nfs_stat_to_errno(int);
                                 op_decode_hdr_maxsz + 2)
 #define NFS4_enc_write_sz       (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
-                                op_encode_hdr_maxsz + 8)
+                                op_encode_hdr_maxsz + 8 + \
+                                encode_getattr_maxsz)
 #define NFS4_dec_write_sz       (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
-                                op_decode_hdr_maxsz + 4)
+                                op_decode_hdr_maxsz + 4 + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_commit_sz      (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
-                                op_encode_hdr_maxsz + 3)
+                                op_encode_hdr_maxsz + 3 + \
+                                encode_getattr_maxsz)
 #define NFS4_dec_commit_sz      (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
-                                op_decode_hdr_maxsz + 2)
+                                op_decode_hdr_maxsz + 2 + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_open_sz        (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
                                 op_encode_hdr_maxsz + \
@@ -196,17 +202,21 @@ static int nfs_stat_to_errno(int);
 #define NFS4_enc_open_downgrade_sz \
                                 (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
-                                op_encode_hdr_maxsz + 7)
+                                op_encode_hdr_maxsz + 7 + \
+                                encode_getattr_maxsz)
 #define NFS4_dec_open_downgrade_sz \
                                 (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
-                                op_decode_hdr_maxsz + 4)
+                                op_decode_hdr_maxsz + 4 + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_close_sz       (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
-                                op_encode_hdr_maxsz + 5)
+                                op_encode_hdr_maxsz + 5 + \
+                                encode_getattr_maxsz)
 #define NFS4_dec_close_sz       (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
-                                op_decode_hdr_maxsz + 4)
+                                op_decode_hdr_maxsz + 4 + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_setattr_sz     (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
                                 op_encode_hdr_maxsz + 4 + \
@@ -300,30 +310,44 @@ static int nfs_stat_to_errno(int);
                                 decode_getfh_maxsz)
 #define NFS4_enc_remove_sz      (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
-                                encode_remove_maxsz)
+                                encode_remove_maxsz + \
+                                encode_getattr_maxsz)
 #define NFS4_dec_remove_sz      (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
-                                op_decode_hdr_maxsz + 5)
+                                op_decode_hdr_maxsz + 5 + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_rename_sz      (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
                                 encode_savefh_maxsz + \
                                 encode_putfh_maxsz + \
-                                encode_rename_maxsz)
+                                encode_rename_maxsz + \
+                                encode_getattr_maxsz + \
+                                encode_restorefh_maxsz + \
+                                encode_getattr_maxsz)
 #define NFS4_dec_rename_sz      (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
                                 decode_savefh_maxsz + \
                                 decode_putfh_maxsz + \
-                                decode_rename_maxsz)
+                                decode_rename_maxsz + \
+                                decode_getattr_maxsz + \
+                                decode_restorefh_maxsz + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_link_sz        (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
                                 encode_savefh_maxsz + \
                                 encode_putfh_maxsz + \
-                                encode_link_maxsz)
+                                encode_link_maxsz + \
+                                decode_getattr_maxsz + \
+                                encode_restorefh_maxsz + \
+                                decode_getattr_maxsz)
 #define NFS4_dec_link_sz        (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
                                 decode_savefh_maxsz + \
                                 decode_putfh_maxsz + \
-                                decode_link_maxsz)
+                                decode_link_maxsz + \
+                                decode_getattr_maxsz + \
+                                decode_restorefh_maxsz + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_symlink_sz     (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
                                 encode_symlink_maxsz + \
@@ -336,14 +360,20 @@ static int nfs_stat_to_errno(int);
                                 decode_getfh_maxsz)
 #define NFS4_enc_create_sz      (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
+                                encode_savefh_maxsz + \
                                 encode_create_maxsz + \
+                                encode_getfh_maxsz + \
                                 encode_getattr_maxsz + \
-                                encode_getfh_maxsz)
+                                encode_restorefh_maxsz + \
+                                encode_getattr_maxsz)
 #define NFS4_dec_create_sz      (compound_decode_hdr_maxsz + \
                                 decode_putfh_maxsz + \
+                                decode_savefh_maxsz + \
                                 decode_create_maxsz + \
+                                decode_getfh_maxsz + \
                                 decode_getattr_maxsz + \
-                                decode_getfh_maxsz)
+                                decode_restorefh_maxsz + \
+                                decode_getattr_maxsz)
 #define NFS4_enc_pathconf_sz    (compound_encode_hdr_maxsz + \
                                 encode_putfh_maxsz + \
                                 encode_getattr_maxsz)
@@ -602,10 +632,10 @@ static int encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg)
 {
         uint32_t *p;
 
-        RESERVE_SPACE(8+sizeof(arg->stateid.data));
+        RESERVE_SPACE(8+sizeof(arg->stateid->data));
         WRITE32(OP_CLOSE);
-        WRITE32(arg->seqid);
-        WRITEMEM(arg->stateid.data, sizeof(arg->stateid.data));
+        WRITE32(arg->seqid->sequence->counter);
+        WRITEMEM(arg->stateid->data, sizeof(arg->stateid->data));
         
         return 0;
 }
@@ -729,22 +759,18 @@ static int encode_lock(struct xdr_stream *xdr, const struct nfs_lockargs *arg)
         WRITE64(arg->length);
         WRITE32(opargs->new_lock_owner);
         if (opargs->new_lock_owner){
-                struct nfs_open_to_lock *ol = opargs->u.open_lock;
-
                 RESERVE_SPACE(40);
-                WRITE32(ol->open_seqid);
-                WRITEMEM(&ol->open_stateid, sizeof(ol->open_stateid));
-                WRITE32(ol->lock_seqid);
-                WRITE64(ol->lock_owner.clientid);
+                WRITE32(opargs->open_seqid->sequence->counter);
+                WRITEMEM(opargs->open_stateid->data, sizeof(opargs->open_stateid->data));
+                WRITE32(opargs->lock_seqid->sequence->counter);
+                WRITE64(opargs->lock_owner.clientid);
                 WRITE32(4);
-                WRITE32(ol->lock_owner.id);
+                WRITE32(opargs->lock_owner.id);
         }
         else {
-                struct nfs_exist_lock *el = opargs->u.exist_lock;
-
                 RESERVE_SPACE(20);
-                WRITEMEM(&el->stateid, sizeof(el->stateid));
-                WRITE32(el->seqid);
+                WRITEMEM(opargs->lock_stateid->data, sizeof(opargs->lock_stateid->data));
+                WRITE32(opargs->lock_seqid->sequence->counter);
         }
 
         return 0;
@@ -775,8 +801,8 @@ static int encode_locku(struct xdr_stream *xdr, const struct nfs_lockargs *arg)
         RESERVE_SPACE(44);
         WRITE32(OP_LOCKU);
         WRITE32(arg->type);
-        WRITE32(opargs->seqid);
-        WRITEMEM(&opargs->stateid, sizeof(opargs->stateid));
+        WRITE32(opargs->seqid->sequence->counter);
+        WRITEMEM(opargs->stateid->data, sizeof(opargs->stateid->data));
         WRITE64(arg->offset);
         WRITE64(arg->length);
 
@@ -826,7 +852,7 @@ static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_opena
  */
         RESERVE_SPACE(8);
         WRITE32(OP_OPEN);
-        WRITE32(arg->seqid);
+        WRITE32(arg->seqid->sequence->counter);
         encode_share_access(xdr, arg->open_flags);
         RESERVE_SPACE(16);
         WRITE64(arg->clientid);
@@ -941,7 +967,7 @@ static int encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_con
         RESERVE_SPACE(8+sizeof(arg->stateid.data));
         WRITE32(OP_OPEN_CONFIRM);
         WRITEMEM(arg->stateid.data, sizeof(arg->stateid.data));
-        WRITE32(arg->seqid);
+        WRITE32(arg->seqid->sequence->counter);
 
         return 0;
 }
@@ -950,10 +976,10 @@ static int encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closea
 {
         uint32_t *p;
 
-        RESERVE_SPACE(8+sizeof(arg->stateid.data));
+        RESERVE_SPACE(8+sizeof(arg->stateid->data));
         WRITE32(OP_OPEN_DOWNGRADE);
-        WRITEMEM(arg->stateid.data, sizeof(arg->stateid.data));
-        WRITE32(arg->seqid);
+        WRITEMEM(arg->stateid->data, sizeof(arg->stateid->data));
+        WRITE32(arg->seqid->sequence->counter);
         encode_share_access(xdr, arg->open_flags);
         return 0;
 }
@@ -1117,6 +1143,17 @@ static int encode_renew(struct xdr_stream *xdr, const struct nfs4_client *client
 }
 
 static int
+encode_restorefh(struct xdr_stream *xdr)
+{
+        uint32_t *p;
+
+        RESERVE_SPACE(4);
+        WRITE32(OP_RESTOREFH);
+
+        return 0;
+}
+
+static int
 encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg)
 {
         uint32_t *p;
@@ -1296,14 +1333,18 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, uint32_t *p, const struct n
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops = 2,
+                .nops = 3,
         };
         int status;
 
         xdr_init_encode(&xdr, &req->rq_snd_buf, p);
         encode_compound_hdr(&xdr, &hdr);
-        if ((status = encode_putfh(&xdr, args->fh)) == 0)
-                status = encode_remove(&xdr, args->name);
+        if ((status = encode_putfh(&xdr, args->fh)) != 0)
+                goto out;
+        if ((status = encode_remove(&xdr, args->name)) != 0)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
+out:
         return status;
 }
 
@@ -1314,7 +1355,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, uint32_t *p, const struct n
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops = 4,
+                .nops = 7,
         };
         int status;
 
@@ -1326,7 +1367,13 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, uint32_t *p, const struct n
                 goto out;
         if ((status = encode_putfh(&xdr, args->new_dir)) != 0)
                 goto out;
-        status = encode_rename(&xdr, args->old_name, args->new_name);
+        if ((status = encode_rename(&xdr, args->old_name, args->new_name)) != 0)
+                goto out;
+        if ((status = encode_getfattr(&xdr, args->bitmask)) != 0)
+                goto out;
+        if ((status = encode_restorefh(&xdr)) != 0)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
 }
@@ -1338,7 +1385,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, uint32_t *p, const struct nfs
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops = 4,
+                .nops = 7,
         };
         int status;
 
@@ -1350,7 +1397,13 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, uint32_t *p, const struct nfs
                 goto out;
         if ((status = encode_putfh(&xdr, args->dir_fh)) != 0)
                 goto out;
-        status = encode_link(&xdr, args->name);
+        if ((status = encode_link(&xdr, args->name)) != 0)
+                goto out;
+        if ((status = encode_getfattr(&xdr, args->bitmask)) != 0)
+                goto out;
+        if ((status = encode_restorefh(&xdr)) != 0)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
 }
@@ -1362,7 +1415,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, uint32_t *p, const struct n
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops = 4,
+                .nops = 7,
         };
         int status;
 
@@ -1370,10 +1423,16 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, uint32_t *p, const struct n
         encode_compound_hdr(&xdr, &hdr);
         if ((status = encode_putfh(&xdr, args->dir_fh)) != 0)
                 goto out;
+        if ((status = encode_savefh(&xdr)) != 0)
+                goto out;
         if ((status = encode_create(&xdr, args)) != 0)
                 goto out;
         if ((status = encode_getfh(&xdr)) != 0)
                 goto out;
+        if ((status = encode_getfattr(&xdr, args->bitmask)) != 0)
+                goto out;
+        if ((status = encode_restorefh(&xdr)) != 0)
+                goto out;
         status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
@@ -1412,7 +1471,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, uint32_t *p, struct nfs_clos
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops   = 2,
+                .nops   = 3,
         };
         int status;
 
@@ -1422,6 +1481,9 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, uint32_t *p, struct nfs_clos
         if(status)
                 goto out;
         status = encode_close(&xdr, args);
+        if (status != 0)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
 }
@@ -1433,15 +1495,21 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, uint32_t *p, struct nfs_opena
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops = 4,
+                .nops = 7,
         };
         int status;
 
+        status = nfs_wait_on_sequence(args->seqid, req->rq_task);
+        if (status != 0)
+                goto out;
         xdr_init_encode(&xdr, &req->rq_snd_buf, p);
         encode_compound_hdr(&xdr, &hdr);
         status = encode_putfh(&xdr, args->fh);
         if (status)
                 goto out;
+        status = encode_savefh(&xdr);
+        if (status)
+                goto out;
         status = encode_open(&xdr, args);
         if (status)
                 goto out;
@@ -1449,6 +1517,12 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, uint32_t *p, struct nfs_opena
         if (status)
                 goto out;
         status = encode_getfattr(&xdr, args->bitmask);
+        if (status)
+                goto out;
+        status = encode_restorefh(&xdr);
+        if (status)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
 }
@@ -1464,6 +1538,9 @@ static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, uint32_t *p, struct n
         };
         int status;
 
+        status = nfs_wait_on_sequence(args->seqid, req->rq_task);
+        if (status != 0)
+                goto out;
         xdr_init_encode(&xdr, &req->rq_snd_buf, p);
         encode_compound_hdr(&xdr, &hdr);
         status = encode_putfh(&xdr, args->fh);
@@ -1485,6 +1562,9 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, uint32_t *p, struct nf
         };
         int status;
 
+        status = nfs_wait_on_sequence(args->seqid, req->rq_task);
+        if (status != 0)
+                goto out;
         xdr_init_encode(&xdr, &req->rq_snd_buf, p);
         encode_compound_hdr(&xdr, &hdr);
         status = encode_putfh(&xdr, args->fh);
@@ -1502,7 +1582,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, uint32_t *p, struct
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops   = 2,
+                .nops   = 3,
         };
         int status;
 
@@ -1512,6 +1592,9 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, uint32_t *p, struct
         if (status)
                 goto out;
         status = encode_open_downgrade(&xdr, args);
+        if (status != 0)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
 }
@@ -1525,8 +1608,15 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, uint32_t *p, struct nfs_locka
         struct compound_hdr hdr = {
                 .nops   = 2,
         };
+        struct nfs_lock_opargs *opargs = args->u.lock;
         int status;
 
+        status = nfs_wait_on_sequence(opargs->lock_seqid, req->rq_task);
+        if (status != 0)
+                goto out;
+        /* Do we need to do an open_to_lock_owner? */
+        if (opargs->lock_seqid->sequence->flags & NFS_SEQID_CONFIRMED)
+                opargs->new_lock_owner = 0;
         xdr_init_encode(&xdr, &req->rq_snd_buf, p);
         encode_compound_hdr(&xdr, &hdr);
         status = encode_putfh(&xdr, args->fh);
@@ -1713,7 +1803,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, uint32_t *p, struct nfs_writ
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops = 2,
+                .nops = 3,
         };
         int status;
 
@@ -1723,6 +1813,9 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, uint32_t *p, struct nfs_writ
         if (status)
                 goto out;
         status = encode_write(&xdr, args);
+        if (status)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
 }
@@ -1734,7 +1827,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, uint32_t *p, struct nfs_wri
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr = {
-                .nops = 2,
+                .nops = 3,
         };
         int status;
 
@@ -1744,6 +1837,9 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, uint32_t *p, struct nfs_wri
         if (status)
                 goto out;
         status = encode_commit(&xdr, args);
+        if (status)
+                goto out;
+        status = encode_getfattr(&xdr, args->bitmask);
 out:
         return status;
 }
@@ -2670,8 +2766,7 @@ static int decode_server_caps(struct xdr_stream *xdr, struct nfs4_server_caps_re
                 goto xdr_error;
         status = verify_attr_len(xdr, savep, attrlen);
 xdr_error:
-        if (status != 0)
-                printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status);
+        dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status);
         return status;
 }
         
@@ -2704,8 +2799,7 @@ static int decode_statfs(struct xdr_stream *xdr, struct nfs_fsstat *fsstat)
 
         status = verify_attr_len(xdr, savep, attrlen);
 xdr_error:
-        if (status != 0)
-                printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status);
+        dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status);
         return status;
 }
 
@@ -2730,8 +2824,7 @@ static int decode_pathconf(struct xdr_stream *xdr, struct nfs_pathconf *pathconf
 
         status = verify_attr_len(xdr, savep, attrlen);
 xdr_error:
-        if (status != 0)
-                printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status);
+        dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status);
         return status;
 }
 
@@ -2787,13 +2880,10 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons
                 goto xdr_error;
         if ((status = decode_attr_time_modify(xdr, bitmap, &fattr->mtime)) != 0)
                 goto xdr_error;
-        if ((status = verify_attr_len(xdr, savep, attrlen)) == 0) {
+        if ((status = verify_attr_len(xdr, savep, attrlen)) == 0)
                 fattr->valid = NFS_ATTR_FATTR | NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4;
-                fattr->timestamp = jiffies;
-        }
 xdr_error:
-        if (status != 0)
-                printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status);
+        dprintk("%s: xdr returned %d\n", __FUNCTION__, -status);
         return status;
 }
 
@@ -2826,8 +2916,7 @@ static int decode_fsinfo(struct xdr_stream *xdr, struct nfs_fsinfo *fsinfo)
 
         status = verify_attr_len(xdr, savep, attrlen);
 xdr_error:
-        if (status != 0)
-                printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status);
+        dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status);
         return status;
 }
 
@@ -2890,8 +2979,8 @@ static int decode_lock(struct xdr_stream *xdr, struct nfs_lockres *res)
 
         status = decode_op_hdr(xdr, OP_LOCK);
         if (status == 0) {
-                READ_BUF(sizeof(nfs4_stateid));
-                COPYMEM(&res->u.stateid, sizeof(res->u.stateid));
+                READ_BUF(sizeof(res->u.stateid.data));
+                COPYMEM(res->u.stateid.data, sizeof(res->u.stateid.data));
         } else if (status == -NFS4ERR_DENIED)
                 return decode_lock_denied(xdr, &res->u.denied);
         return status;
@@ -2913,8 +3002,8 @@ static int decode_locku(struct xdr_stream *xdr, struct nfs_lockres *res)
 
         status = decode_op_hdr(xdr, OP_LOCKU);
         if (status == 0) {
-                READ_BUF(sizeof(nfs4_stateid));
-                COPYMEM(&res->u.stateid, sizeof(res->u.stateid));
+                READ_BUF(sizeof(res->u.stateid.data));
+                COPYMEM(res->u.stateid.data, sizeof(res->u.stateid.data));
         }
         return status;
 }
@@ -2994,7 +3083,7 @@ static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res)
         p += bmlen;
         return decode_delegation(xdr, res);
 xdr_error:
-        printk(KERN_NOTICE "%s: xdr error!\n", __FUNCTION__);
+        dprintk("%s: Bitmap too large! Length = %u\n", __FUNCTION__, bmlen);
         return -EIO;
 }
 
@@ -3208,6 +3297,12 @@ static int decode_renew(struct xdr_stream *xdr)
         return decode_op_hdr(xdr, OP_RENEW);
 }
 
+static int
+decode_restorefh(struct xdr_stream *xdr)
+{
+        return decode_op_hdr(xdr, OP_RESTOREFH);
+}
+
 static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req,
                 size_t *acl_len)
 {
@@ -3243,7 +3338,8 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req,
                 if (attrlen <= *acl_len)
                         xdr_read_pages(xdr, attrlen);
                 *acl_len = attrlen;
-        }
+        } else
+                status = -EOPNOTSUPP;
 
 out:
         return status;
@@ -3352,6 +3448,9 @@ static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, uint32_t *p, stru
         if (status)
                 goto out;
         status = decode_open_downgrade(&xdr, res);
+        if (status != 0)
+                goto out;
+        decode_getfattr(&xdr, res->fattr, res->server);
 out:
         return status;
 }
@@ -3424,7 +3523,7 @@ out:
 /*
  * Decode REMOVE response
  */
-static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_change_info *cinfo)
+static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_remove_res *res)
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr;
@@ -3433,8 +3532,11 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_
         xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
         if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
                 goto out;
-        if ((status = decode_putfh(&xdr)) == 0)
-                status = decode_remove(&xdr, cinfo);
+        if ((status = decode_putfh(&xdr)) != 0)
+                goto out;
+        if ((status = decode_remove(&xdr, &res->cinfo)) != 0)
+                goto out;
+        decode_getfattr(&xdr, res->dir_attr, res->server);
 out:
         return status;
 }
@@ -3457,7 +3559,14 @@ static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_
                 goto out;
         if ((status = decode_putfh(&xdr)) != 0)
                 goto out;
-        status = decode_rename(&xdr, &res->old_cinfo, &res->new_cinfo);
+        if ((status = decode_rename(&xdr, &res->old_cinfo, &res->new_cinfo)) != 0)
+                goto out;
+        /* Current FH is target directory */
+        if (decode_getfattr(&xdr, res->new_fattr, res->server) != 0)
+                goto out;
+        if ((status = decode_restorefh(&xdr)) != 0)
+                goto out;
+        decode_getfattr(&xdr, res->old_fattr, res->server);
 out:
         return status;
 }
@@ -3465,7 +3574,7 @@ out:
 /*
  * Decode LINK response
  */
-static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_change_info *cinfo)
+static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_link_res *res)
 {
         struct xdr_stream xdr;
         struct compound_hdr hdr;
@@ -3480,7 +3589,17 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_ch
                 goto out;
         if ((status = decode_putfh(&xdr)) != 0)
                 goto out;
-        status = decode_link(&xdr, cinfo);
+        if ((status = decode_link(&xdr, &res->cinfo)) != 0)
+                goto out;
+        /*
+         * Note order: OP_LINK leaves the directory as the current
+         *             filehandle.
+         */
+        if (decode_getfattr(&xdr, res->dir_attr, res->server) != 0)
+                goto out;
+        if ((status = decode_restorefh(&xdr)) != 0)
+                goto out;
+        decode_getfattr(&xdr, res->fattr, res->server);
 out:
         return status;
 }
@@ -3499,13 +3618,17 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_
                 goto out;
         if ((status = decode_putfh(&xdr)) != 0)
                 goto out;
+        if ((status = decode_savefh(&xdr)) != 0)
+                goto out;
         if ((status = decode_create(&xdr,&res->dir_cinfo)) != 0)
                 goto out;
         if ((status = decode_getfh(&xdr, res->fh)) != 0)
                 goto out;
-        status = decode_getfattr(&xdr, res->fattr, res->server);
-        if (status == NFS4ERR_DELAY)
-                status = 0;
+        if (decode_getfattr(&xdr, res->fattr, res->server) != 0)
+                goto out;
+        if ((status = decode_restorefh(&xdr)) != 0)
+                goto out;
+        decode_getfattr(&xdr, res->dir_fattr, res->server);
 out:
         return status;
 }
@@ -3623,6 +3746,15 @@ static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_cl
         if (status)
                 goto out;
         status = decode_close(&xdr, res);
+        if (status != 0)
+                goto out;
+        /*
+         * Note: Server may do delete on close for this file
+         *      in which case the getattr call will fail with
+         *      an ESTALE error. Shouldn't be a problem,
+         *      though, since fattr->valid will remain unset.
+         */
+        decode_getfattr(&xdr, res->fattr, res->server);
 out:
         return status;
 }
@@ -3643,15 +3775,20 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_ope
         status = decode_putfh(&xdr);
         if (status)
                 goto out;
+        status = decode_savefh(&xdr);
+        if (status)
+                goto out;
         status = decode_open(&xdr, res);
         if (status)
                 goto out;
         status = decode_getfh(&xdr, &res->fh);
         if (status)
                 goto out;
-        status = decode_getfattr(&xdr, res->f_attr, res->server);
-        if (status == NFS4ERR_DELAY)
-                status = 0;
+        if (decode_getfattr(&xdr, res->f_attr, res->server) != 0)
+                goto out;
+        if ((status = decode_restorefh(&xdr)) != 0)
+                goto out;
+        decode_getfattr(&xdr, res->dir_attr, res->server);
 out:
         return status;
 }
@@ -3869,6 +4006,9 @@ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_wr
         if (status)
                 goto out;
         status = decode_write(&xdr, res);
+        if (status)
+                goto out;
+        decode_getfattr(&xdr, res->fattr, res->server);
         if (!status)
                 status = res->count;
 out:
@@ -3892,6 +4032,9 @@ static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_w
         if (status)
                 goto out;
         status = decode_commit(&xdr, res);
+        if (status)
+                goto out;
+        decode_getfattr(&xdr, res->fattr, res->server);
 out:
         return status;
 }
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index be23c3fb9260..a48a003242c0 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -61,7 +61,7 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
         int status;
 
         dprintk("%s: call getattr\n", __FUNCTION__);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call(server->client_sys, NFSPROC_GETATTR, fhandle, fattr, 0);
         dprintk("%s: reply getattr: %d\n", __FUNCTION__, status);
         if (status)
@@ -93,7 +93,7 @@ nfs_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("NFS call  getattr\n");
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call(server->client, NFSPROC_GETATTR,
                                 fhandle, fattr, 0);
         dprintk("NFS reply getattr: %d\n", status);
@@ -112,7 +112,7 @@ nfs_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
         int     status;
 
         dprintk("NFS call  setattr\n");
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call(NFS_CLIENT(inode), NFSPROC_SETATTR, &arg, fattr, 0);
         if (status == 0)
                 nfs_setattr_update_inode(inode, sattr);
@@ -136,7 +136,7 @@ nfs_proc_lookup(struct inode *dir, struct qstr *name,
         int                     status;
 
         dprintk("NFS call  lookup %s\n", name->name);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call(NFS_CLIENT(dir), NFSPROC_LOOKUP, &arg, &res, 0);
         dprintk("NFS reply lookup: %d\n", status);
         return status;
@@ -174,7 +174,7 @@ static int nfs_proc_read(struct nfs_read_data *rdata)
 
         dprintk("NFS call  read %d @ %Ld\n", rdata->args.count,
                         (long long) rdata->args.offset);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(NFS_CLIENT(inode), &msg, flags);
         if (status >= 0) {
                 nfs_refresh_inode(inode, fattr);
@@ -203,10 +203,10 @@ static int nfs_proc_write(struct nfs_write_data *wdata)
 
         dprintk("NFS call  write %d @ %Ld\n", wdata->args.count,
                         (long long) wdata->args.offset);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         status = rpc_call_sync(NFS_CLIENT(inode), &msg, flags);
         if (status >= 0) {
-                nfs_refresh_inode(inode, fattr);
+                nfs_post_op_update_inode(inode, fattr);
                 wdata->res.count = wdata->args.count;
                 wdata->verf.committed = NFS_FILE_SYNC;
         }
@@ -216,7 +216,7 @@ static int nfs_proc_write(struct nfs_write_data *wdata)
 
 static int
 nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
-                int flags)
+                int flags, struct nameidata *nd)
 {
         struct nfs_fh           fhandle;
         struct nfs_fattr        fattr;
@@ -232,7 +232,7 @@ nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
         };
         int                     status;
 
-        fattr.valid = 0;
+        nfs_fattr_init(&fattr);
         dprintk("NFS call  create %s\n", dentry->d_name.name);
         status = rpc_call(NFS_CLIENT(dir), NFSPROC_CREATE, &arg, &res, 0);
         if (status == 0)
@@ -273,12 +273,13 @@ nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
                 sattr->ia_size = new_encode_dev(rdev);/* get out your barf bag */
         }
 
-        fattr.valid = 0;
+        nfs_fattr_init(&fattr);
         status = rpc_call(NFS_CLIENT(dir), NFSPROC_CREATE, &arg, &res, 0);
+        nfs_mark_for_revalidate(dir);
 
         if (status == -EINVAL && S_ISFIFO(mode)) {
                 sattr->ia_mode = mode;
-                fattr.valid = 0;
+                nfs_fattr_init(&fattr);
                 status = rpc_call(NFS_CLIENT(dir), NFSPROC_CREATE, &arg, &res, 0);
         }
         if (status == 0)
@@ -305,6 +306,7 @@ nfs_proc_remove(struct inode *dir, struct qstr *name)
 
         dprintk("NFS call  remove %s\n", name->name);
         status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+        nfs_mark_for_revalidate(dir);
 
         dprintk("NFS reply remove: %d\n", status);
         return status;
@@ -331,8 +333,10 @@ nfs_proc_unlink_done(struct dentry *dir, struct rpc_task *task)
 {
         struct rpc_message *msg = &task->tk_msg;
         
-        if (msg->rpc_argp)
+        if (msg->rpc_argp) {
+                nfs_mark_for_revalidate(dir->d_inode);
                 kfree(msg->rpc_argp);
+        }
         return 0;
 }
 
@@ -352,6 +356,8 @@ nfs_proc_rename(struct inode *old_dir, struct qstr *old_name,
 
         dprintk("NFS call  rename %s -> %s\n", old_name->name, new_name->name);
         status = rpc_call(NFS_CLIENT(old_dir), NFSPROC_RENAME, &arg, NULL, 0);
+        nfs_mark_for_revalidate(old_dir);
+        nfs_mark_for_revalidate(new_dir);
         dprintk("NFS reply rename: %d\n", status);
         return status;
 }
@@ -369,6 +375,7 @@ nfs_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
 
         dprintk("NFS call  link %s\n", name->name);
         status = rpc_call(NFS_CLIENT(inode), NFSPROC_LINK, &arg, NULL, 0);
+        nfs_mark_for_revalidate(dir);
         dprintk("NFS reply link: %d\n", status);
         return status;
 }
@@ -391,9 +398,10 @@ nfs_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path,
         if (path->len > NFS2_MAXPATHLEN)
                 return -ENAMETOOLONG;
         dprintk("NFS call  symlink %s -> %s\n", name->name, path->name);
-        fattr->valid = 0;
+        nfs_fattr_init(fattr);
         fhandle->size = 0;
         status = rpc_call(NFS_CLIENT(dir), NFSPROC_SYMLINK, &arg, NULL, 0);
+        nfs_mark_for_revalidate(dir);
         dprintk("NFS reply symlink: %d\n", status);
         return status;
 }
@@ -416,8 +424,9 @@ nfs_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr)
         int                     status;
 
         dprintk("NFS call  mkdir %s\n", dentry->d_name.name);
-        fattr.valid = 0;
+        nfs_fattr_init(&fattr);
         status = rpc_call(NFS_CLIENT(dir), NFSPROC_MKDIR, &arg, &res, 0);
+        nfs_mark_for_revalidate(dir);
         if (status == 0)
                 status = nfs_instantiate(dentry, &fhandle, &fattr);
         dprintk("NFS reply mkdir: %d\n", status);
@@ -436,6 +445,7 @@ nfs_proc_rmdir(struct inode *dir, struct qstr *name)
 
         dprintk("NFS call  rmdir %s\n", name->name);
         status = rpc_call(NFS_CLIENT(dir), NFSPROC_RMDIR, &arg, NULL, 0);
+        nfs_mark_for_revalidate(dir);
         dprintk("NFS reply rmdir: %d\n", status);
         return status;
 }
@@ -484,7 +494,7 @@ nfs_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("NFS call  statfs\n");
-        stat->fattr->valid = 0;
+        nfs_fattr_init(stat->fattr);
         status = rpc_call(server->client, NFSPROC_STATFS, fhandle, &fsinfo, 0);
         dprintk("NFS reply statfs: %d\n", status);
         if (status)
@@ -507,7 +517,7 @@ nfs_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
         int     status;
 
         dprintk("NFS call  fsinfo\n");
-        info->fattr->valid = 0;
+        nfs_fattr_init(info->fattr);
         status = rpc_call(server->client, NFSPROC_STATFS, fhandle, &fsinfo, 0);
         dprintk("NFS reply fsinfo: %d\n", status);
         if (status)
@@ -579,7 +589,7 @@ nfs_write_done(struct rpc_task *task)
         struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata;
 
         if (task->tk_status >= 0)
-                nfs_refresh_inode(data->inode, data->res.fattr);
+                nfs_post_op_update_inode(data->inode, data->res.fattr);
         nfs_writeback_done(task);
 }
 
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 9758ebd49905..43b03b19731b 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -215,6 +215,7 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
         data->res.fattr   = &data->fattr;
         data->res.count   = count;
         data->res.eof     = 0;
+        nfs_fattr_init(&data->fattr);
 
         NFS_PROTO(inode)->read_setup(data);
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 5130eda231d7..819a65f5071f 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -870,6 +870,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req,
         data->res.fattr   = &data->fattr;
         data->res.count   = count;
         data->res.verf    = &data->verf;
+        nfs_fattr_init(&data->fattr);
 
         NFS_PROTO(inode)->write_setup(data, how);
 
@@ -1237,6 +1238,7 @@ static void nfs_commit_rpcsetup(struct list_head *head,
         data->res.count   = 0;
         data->res.fattr   = &data->fattr;
         data->res.verf    = &data->verf;
+        nfs_fattr_init(&data->fattr);
         
         NFS_PROTO(inode)->commit_setup(data, how);
 
diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog
index de58579a1d0e..50a7749cfca1 100644
--- a/fs/ntfs/ChangeLog
+++ b/fs/ntfs/ChangeLog
@@ -1,18 +1,15 @@
 ToDo/Notes:
         - Find and fix bugs.
-        - In between ntfs_prepare/commit_write, need exclusion between
-          simultaneous file extensions.  This is given to us by holding i_sem
-          on the inode.  The only places in the kernel when a file is resized
-          are prepare/commit write and truncate for both of which i_sem is
-          held.  Just have to be careful in readpage/writepage and all other
-          helpers not running under i_sem that we play nice...
-          Also need to be careful with initialized_size extention in
-          ntfs_prepare_write. Basically, just be _very_ careful in this code...
-          UPDATE: The only things that need to be checked are read/writepage
-          which do not hold i_sem.  Note writepage cannot change i_size but it
-          needs to cope with a concurrent i_size change, just like readpage.
-          Also both need to cope with concurrent changes to the other sizes,
-          i.e. initialized/allocated/compressed size, as well.
+        - The only places in the kernel where a file is resized are
+          ntfs_file_write*() and ntfs_truncate() for both of which i_sem is
+          held.  Just have to be careful in read-/writepage and other helpers
+          not running under i_sem that we play nice...  Also need to be careful
+          with initialized_size extension in ntfs_file_write*() and writepage.
+          UPDATE: The only things that need to be checked are the compressed
+          write and the other attribute resize/write cases like index
+          attributes, etc.  For now none of these are implemented so are safe.
+        - Implement filling in of holes in aops.c::ntfs_writepage() and its
+          helpers.
         - Implement mft.c::sync_mft_mirror_umount().  We currently will just
           leave the volume dirty on umount if the final iput(vol->mft_ino)
           causes a write of any mirrored mft records due to the mft mirror
@@ -22,6 +19,68 @@ ToDo/Notes:
         - Enable the code for setting the NT4 compatibility flag when we start
           making NTFS 1.2 specific modifications.
 
+2.1.25 - (Almost) fully implement write(2) and truncate(2).
+
+        - Change ntfs_map_runlist_nolock(), ntfs_attr_find_vcn_nolock() and
+          {__,}ntfs_cluster_free() to also take an optional attribute search
+          context as argument.  This allows calling these functions with the
+          mft record mapped.  Update all callers.
+        - Fix potential deadlock in ntfs_mft_data_extend_allocation_nolock()
+          error handling by passing in the active search context when calling
+          ntfs_cluster_free().
+        - Change ntfs_cluster_alloc() to take an extra boolean parameter
+          specifying whether the cluster are being allocated to extend an
+          attribute or to fill a hole.
+        - Change ntfs_attr_make_non_resident() to call ntfs_cluster_alloc()
+          with @is_extension set to TRUE and remove the runlist terminator
+          fixup code as this is now done by ntfs_cluster_alloc().
+        - Change ntfs_attr_make_non_resident to take the attribute value size
+          as an extra parameter.  This is needed since we need to know the size
+          before we can map the mft record and our callers always know it.  The
+          reason we cannot simply read the size from the vfs inode i_size is
+          that this is not necessarily uptodate.  This happens when
+          ntfs_attr_make_non_resident() is called in the ->truncate call path.
+        - Fix ntfs_attr_make_non_resident() to update the vfs inode i_blocks
+          which is zero for a resident attribute but should no longer be zero
+          once the attribute is non-resident as it then has real clusters
+          allocated.
+        - Add fs/ntfs/attrib.[hc]::ntfs_attr_extend_allocation(), a function to
+          extend the allocation of an attributes.  Optionally, the data size,
+          but not the initialized size can be extended, too.
+        - Implement fs/ntfs/inode.[hc]::ntfs_truncate().  It only supports
+          uncompressed and unencrypted files and it never creates sparse files
+          at least for the moment (making a file sparse requires us to modify
+          its directory entries and we do not support directory operations at
+          the moment).  Also, support for highly fragmented files, i.e. ones
+          whose data attribute is split across multiple extents, is severly
+          limited.  When such a case is encountered, EOPNOTSUPP is returned.
+        - Enable ATTR_SIZE attribute changes in ntfs_setattr().  This completes
+          the initial implementation of file truncation.  Now both open(2)ing
+          a file with the O_TRUNC flag and the {,f}truncate(2) system calls
+          will resize a file appropriately.  The limitations are that only
+          uncompressed and unencrypted files are supported.  Also, there is
+          only very limited support for highly fragmented files (the ones whose
+          $DATA attribute is split into multiple attribute extents).
+        - In attrib.c::ntfs_attr_set() call balance_dirty_pages_ratelimited()
+          and cond_resched() in the main loop as we could be dirtying a lot of
+          pages and this ensures we play nice with the VM and the system as a
+          whole.
+        - Implement file operations ->write, ->aio_write, ->writev for regular
+          files.  This replaces the old use of generic_file_write(), et al and
+          the address space operations ->prepare_write and ->commit_write.
+          This means that both sparse and non-sparse (unencrypted and
+          uncompressed) files can now be extended using the normal write(2)
+          code path.  There are two limitations at present and these are that
+          we never create sparse files and that we only have limited support
+          for highly fragmented files, i.e. ones whose data attribute is split
+          across multiple extents.   When such a case is encountered,
+          EOPNOTSUPP is returned.
+        - $EA attributes can be both resident and non-resident.
+        - Use %z for size_t to fix compilation warnings.  (Andrew Morton)
+        - Fix compilation warnings with gcc-4.0.2 on SUSE 10.0.
+        - Document extended attribute ($EA) NEED_EA flag.  (Based on libntfs
+          patch by Yura Pakhuchiy.)
+
 2.1.24 - Lots of bug fixes and support more clean journal states.
 
         - Support journals ($LogFile) which have been modified by chkdsk.  This
diff --git a/fs/ntfs/Makefile b/fs/ntfs/Makefile
index 894b2b876d35..d0d45d1c853a 100644
--- a/fs/ntfs/Makefile
+++ b/fs/ntfs/Makefile
@@ -6,7 +6,7 @@ ntfs-objs := aops.o attrib.o collate.o compress.o debug.o dir.o file.o \
              index.o inode.o mft.o mst.o namei.o runlist.o super.o sysctl.o \
              unistr.o upcase.o
 
-EXTRA_CFLAGS = -DNTFS_VERSION=\"2.1.24\"
+EXTRA_CFLAGS = -DNTFS_VERSION=\"2.1.25\"
 
 ifeq ($(CONFIG_NTFS_DEBUG),y)
 EXTRA_CFLAGS += -DDEBUG
diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
index 5e80c07c6a4d..1c0a4315876a 100644
--- a/fs/ntfs/aops.c
+++ b/fs/ntfs/aops.c
@@ -1391,8 +1391,7 @@ retry_writepage:
                 if (NInoEncrypted(ni)) {
                         unlock_page(page);
                         BUG_ON(ni->type != AT_DATA);
-                        ntfs_debug("Denying write access to encrypted "
-                                        "file.");
+                        ntfs_debug("Denying write access to encrypted file.");
                         return -EACCES;
                 }
                 /* Compressed data streams are handled in compress.c. */
@@ -1508,8 +1507,8 @@ retry_writepage:
         /* Zero out of bounds area in the page cache page. */
         memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len);
         kunmap_atomic(kaddr, KM_USER0);
-        flush_dcache_mft_record_page(ctx->ntfs_ino);
         flush_dcache_page(page);
+        flush_dcache_mft_record_page(ctx->ntfs_ino);
         /* We are done with the page. */
         end_page_writeback(page);
         /* Finally, mark the mft record dirty, so it gets written back. */
@@ -1542,830 +1541,6 @@ err_out:
         return err;
 }
 
-/**
- * ntfs_prepare_nonresident_write -
- *
- */
-static int ntfs_prepare_nonresident_write(struct page *page,
-                unsigned from, unsigned to)
-{
-        VCN vcn;
-        LCN lcn;
-        s64 initialized_size;
-        loff_t i_size;
-        sector_t block, ablock, iblock;
-        struct inode *vi;
-        ntfs_inode *ni;
-        ntfs_volume *vol;
-        runlist_element *rl;
-        struct buffer_head *bh, *head, *wait[2], **wait_bh = wait;
-        unsigned long flags;
-        unsigned int vcn_ofs, block_start, block_end, blocksize;
-        int err;
-        BOOL is_retry;
-        unsigned char blocksize_bits;
-
-        vi = page->mapping->host;
-        ni = NTFS_I(vi);
-        vol = ni->vol;
-
-        ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index "
-                        "0x%lx, from = %u, to = %u.", ni->mft_no, ni->type,
-                        page->index, from, to);
-
-        BUG_ON(!NInoNonResident(ni));
-
-        blocksize_bits = vi->i_blkbits;
-        blocksize = 1 << blocksize_bits;
-
-        /*
-         * create_empty_buffers() will create uptodate/dirty buffers if the
-         * page is uptodate/dirty.
-         */
-        if (!page_has_buffers(page))
-                create_empty_buffers(page, blocksize, 0);
-        bh = head = page_buffers(page);
-        if (unlikely(!bh))
-                return -ENOMEM;
-
-        /* The first block in the page. */
-        block = (s64)page->index << (PAGE_CACHE_SHIFT - blocksize_bits);
-
-        read_lock_irqsave(&ni->size_lock, flags);
-        /*
-         * The first out of bounds block for the allocated size.  No need to
-         * round up as allocated_size is in multiples of cluster size and the
-         * minimum cluster size is 512 bytes, which is equal to the smallest
-         * blocksize.
-         */
-        ablock = ni->allocated_size >> blocksize_bits;
-        i_size = i_size_read(vi);
-        initialized_size = ni->initialized_size;
-        read_unlock_irqrestore(&ni->size_lock, flags);
-
-        /* The last (fully or partially) initialized block. */
-        iblock = initialized_size >> blocksize_bits;
-
-        /* Loop through all the buffers in the page. */
-        block_start = 0;
-        rl = NULL;
-        err = 0;
-        do {
-                block_end = block_start + blocksize;
-                /*
-                 * If buffer @bh is outside the write, just mark it uptodate
-                 * if the page is uptodate and continue with the next buffer.
-                 */
-                if (block_end <= from || block_start >= to) {
-                        if (PageUptodate(page)) {
-                                if (!buffer_uptodate(bh))
-                                        set_buffer_uptodate(bh);
-                        }
-                        continue;
-                }
-                /*
-                 * @bh is at least partially being written to.
-                 * Make sure it is not marked as new.
-                 */
-                //if (buffer_new(bh))
-                //      clear_buffer_new(bh);
-
-                if (block >= ablock) {
-                        // TODO: block is above allocated_size, need to
-                        // allocate it. Best done in one go to accommodate not
-                        // only block but all above blocks up to and including:
-                        // ((page->index << PAGE_CACHE_SHIFT) + to + blocksize
-                        // - 1) >> blobksize_bits. Obviously will need to round
-                        // up to next cluster boundary, too. This should be
-                        // done with a helper function, so it can be reused.
-                        ntfs_error(vol->sb, "Writing beyond allocated size "
-                                        "is not supported yet. Sorry.");
-                        err = -EOPNOTSUPP;
-                        goto err_out;
-                        // Need to update ablock.
-                        // Need to set_buffer_new() on all block bhs that are
-                        // newly allocated.
-                }
-                /*
-                 * Now we have enough allocated size to fulfill the whole
-                 * request, i.e. block < ablock is true.
-                 */
-                if (unlikely((block >= iblock) &&
-                                (initialized_size < i_size))) {
-                        /*
-                         * If this page is fully outside initialized size, zero
-                         * out all pages between the current initialized size
-                         * and the current page. Just use ntfs_readpage() to do
-                         * the zeroing transparently.
-                         */
-                        if (block > iblock) {
-                                // TODO:
-                                // For each page do:
-                                // - read_cache_page()
-                                // Again for each page do:
-                                // - wait_on_page_locked()
-                                // - Check (PageUptodate(page) &&
-                                //                      !PageError(page))
-                                // Update initialized size in the attribute and
-                                // in the inode.
-                                // Again, for each page do:
-                                //      __set_page_dirty_buffers();
-                                // page_cache_release()
-                                // We don't need to wait on the writes.
-                                // Update iblock.
-                        }
-                        /*
-                         * The current page straddles initialized size. Zero
-                         * all non-uptodate buffers and set them uptodate (and
-                         * dirty?). Note, there aren't any non-uptodate buffers
-                         * if the page is uptodate.
-                         * FIXME: For an uptodate page, the buffers may need to
-                         * be written out because they were not initialized on
-                         * disk before.
-                         */
-                        if (!PageUptodate(page)) {
-                                // TODO:
-                                // Zero any non-uptodate buffers up to i_size.
-                                // Set them uptodate and dirty.
-                        }
-                        // TODO:
-                        // Update initialized size in the attribute and in the
-                        // inode (up to i_size).
-                        // Update iblock.
-                        // FIXME: This is inefficient. Try to batch the two
-                        // size changes to happen in one go.
-                        ntfs_error(vol->sb, "Writing beyond initialized size "
-                                        "is not supported yet. Sorry.");
-                        err = -EOPNOTSUPP;
-                        goto err_out;
-                        // Do NOT set_buffer_new() BUT DO clear buffer range
-                        // outside write request range.
-                        // set_buffer_uptodate() on complete buffers as well as
-                        // set_buffer_dirty().
-                }
-
-                /* Need to map unmapped buffers. */
-                if (!buffer_mapped(bh)) {
-                        /* Unmapped buffer. Need to map it. */
-                        bh->b_bdev = vol->sb->s_bdev;
-
-                        /* Convert block into corresponding vcn and offset. */
-                        vcn = (VCN)block << blocksize_bits >>
-                                        vol->cluster_size_bits;
-                        vcn_ofs = ((VCN)block << blocksize_bits) &
-                                        vol->cluster_size_mask;
-
-                        is_retry = FALSE;
-                        if (!rl) {
-lock_retry_remap:
-                                down_read(&ni->runlist.lock);
-                                rl = ni->runlist.rl;
-                        }
-                        if (likely(rl != NULL)) {
-                                /* Seek to element containing target vcn. */
-                                while (rl->length && rl[1].vcn <= vcn)
-                                        rl++;
-                                lcn = ntfs_rl_vcn_to_lcn(rl, vcn);
-                        } else
-                                lcn = LCN_RL_NOT_MAPPED;
-                        if (unlikely(lcn < 0)) {
-                                /*
-                                 * We extended the attribute allocation above.
-                                 * If we hit an ENOENT here it means that the
-                                 * allocation was insufficient which is a bug.
-                                 */
-                                BUG_ON(lcn == LCN_ENOENT);
-
-                                /* It is a hole, need to instantiate it. */
-                                if (lcn == LCN_HOLE) {
-                                        // TODO: Instantiate the hole.
-                                        // clear_buffer_new(bh);
-                                        // unmap_underlying_metadata(bh->b_bdev,
-                                        //              bh->b_blocknr);
-                                        // For non-uptodate buffers, need to
-                                        // zero out the region outside the
-                                        // request in this bh or all bhs,
-                                        // depending on what we implemented
-                                        // above.
-                                        // Need to flush_dcache_page().
-                                        // Or could use set_buffer_new()
-                                        // instead?
-                                        ntfs_error(vol->sb, "Writing into "
-                                                        "sparse regions is "
-                                                        "not supported yet. "
-                                                        "Sorry.");
-                                        err = -EOPNOTSUPP;
-                                        if (!rl)
-                                                up_read(&ni->runlist.lock);
-                                        goto err_out;
-                                } else if (!is_retry &&
-                                                lcn == LCN_RL_NOT_MAPPED) {
-                                        is_retry = TRUE;
-                                        /*
-                                         * Attempt to map runlist, dropping
-                                         * lock for the duration.
-                                         */
-                                        up_read(&ni->runlist.lock);
-                                        err = ntfs_map_runlist(ni, vcn);
-                                        if (likely(!err))
-                                                goto lock_retry_remap;
-                                        rl = NULL;
-                                } else if (!rl)
-                                        up_read(&ni->runlist.lock);
-                                /*
-                                 * Failed to map the buffer, even after
-                                 * retrying.
-                                 */
-                                if (!err)
-                                        err = -EIO;
-                                bh->b_blocknr = -1;
-                                ntfs_error(vol->sb, "Failed to write to inode "
-                                                "0x%lx, attribute type 0x%x, "
-                                                "vcn 0x%llx, offset 0x%x "
-                                                "because its location on disk "
-                                                "could not be determined%s "
-                                                "(error code %i).",
-                                                ni->mft_no, ni->type,
-                                                (unsigned long long)vcn,
-                                                vcn_ofs, is_retry ? " even "
-                                                "after retrying" : "", err);
-                                goto err_out;
-                        }
-                        /* We now have a successful remap, i.e. lcn >= 0. */
-
-                        /* Setup buffer head to correct block. */
-                        bh->b_blocknr = ((lcn << vol->cluster_size_bits)
-                                        + vcn_ofs) >> blocksize_bits;
-                        set_buffer_mapped(bh);
-
-                        // FIXME: Something analogous to this is needed for
-                        // each newly allocated block, i.e. BH_New.
-                        // FIXME: Might need to take this out of the
-                        // if (!buffer_mapped(bh)) {}, depending on how we
-                        // implement things during the allocated_size and
-                        // initialized_size extension code above.
-                        if (buffer_new(bh)) {
-                                clear_buffer_new(bh);
-                                unmap_underlying_metadata(bh->b_bdev,
-                                                bh->b_blocknr);
-                                if (PageUptodate(page)) {
-                                        set_buffer_uptodate(bh);
-                                        continue;
-                                }
-                                /*
-                                 * Page is _not_ uptodate, zero surrounding
-                                 * region. NOTE: This is how we decide if to
-                                 * zero or not!
-                                 */
-                                if (block_end > to || block_start < from) {
-                                        void *kaddr;
-
-                                        kaddr = kmap_atomic(page, KM_USER0);
-                                        if (block_end > to)
-                                                memset(kaddr + to, 0,
-                                                                block_end - to);
-                                        if (block_start < from)
-                                                memset(kaddr + block_start, 0,
-                                                                from -
-                                                                block_start);
-                                        flush_dcache_page(page);
-                                        kunmap_atomic(kaddr, KM_USER0);
-                                }
-                                continue;
-                        }
-                }
-                /* @bh is mapped, set it uptodate if the page is uptodate. */
-                if (PageUptodate(page)) {
-                        if (!buffer_uptodate(bh))
-                                set_buffer_uptodate(bh);
-                        continue;
-                }
-                /*
-                 * The page is not uptodate. The buffer is mapped. If it is not
-                 * uptodate, and it is only partially being written to, we need
-                 * to read the buffer in before the write, i.e. right now.
-                 */
-                if (!buffer_uptodate(bh) &&
-                                (block_start < from || block_end > to)) {
-                        ll_rw_block(READ, 1, &bh);
-                        *wait_bh++ = bh;
-                }
-        } while (block++, block_start = block_end,
-                        (bh = bh->b_this_page) != head);
-
-        /* Release the lock if we took it. */
-        if (rl) {
-                up_read(&ni->runlist.lock);
-                rl = NULL;
-        }
-
-        /* If we issued read requests, let them complete. */
-        while (wait_bh > wait) {
-                wait_on_buffer(*--wait_bh);
-                if (!buffer_uptodate(*wait_bh))
-                        return -EIO;
-        }
-
-        ntfs_debug("Done.");
-        return 0;
-err_out:
-        /*
-         * Zero out any newly allocated blocks to avoid exposing stale data.
-         * If BH_New is set, we know that the block was newly allocated in the
-         * above loop.
-         * FIXME: What about initialized_size increments? Have we done all the
-         * required zeroing above? If not this error handling is broken, and
-         * in particular the if (block_end <= from) check is completely bogus.
-         */
-        bh = head;
-        block_start = 0;
-        is_retry = FALSE;
-        do {
-                block_end = block_start + blocksize;
-                if (block_end <= from)
-                        continue;
-                if (block_start >= to)
-                        break;
-                if (buffer_new(bh)) {
-                        void *kaddr;
-
-                        clear_buffer_new(bh);
-                        kaddr = kmap_atomic(page, KM_USER0);
-                        memset(kaddr + block_start, 0, bh->b_size);
-                        kunmap_atomic(kaddr, KM_USER0);
-                        set_buffer_uptodate(bh);
-                        mark_buffer_dirty(bh);
-                        is_retry = TRUE;
-                }
-        } while (block_start = block_end, (bh = bh->b_this_page) != head);
-        if (is_retry)
-                flush_dcache_page(page);
-        if (rl)
-                up_read(&ni->runlist.lock);
-        return err;
-}
-
-/**
- * ntfs_prepare_write - prepare a page for receiving data
- *
- * This is called from generic_file_write() with i_sem held on the inode
- * (@page->mapping->host).  The @page is locked but not kmap()ped.  The source
- * data has not yet been copied into the @page.
- *
- * Need to extend the attribute/fill in holes if necessary, create blocks and
- * make partially overwritten blocks uptodate,
- *
- * i_size is not to be modified yet.
- *
- * Return 0 on success or -errno on error.
- *
- * Should be using block_prepare_write() [support for sparse files] or
- * cont_prepare_write() [no support for sparse files].  Cannot do that due to
- * ntfs specifics but can look at them for implementation guidance.
- *
- * Note: In the range, @from is inclusive and @to is exclusive, i.e. @from is
- * the first byte in the page that will be written to and @to is the first byte
- * after the last byte that will be written to.
- */
-static int ntfs_prepare_write(struct file *file, struct page *page,
-                unsigned from, unsigned to)
-{
-        s64 new_size;
-        loff_t i_size;
-        struct inode *vi = page->mapping->host;
-        ntfs_inode *base_ni = NULL, *ni = NTFS_I(vi);
-        ntfs_volume *vol = ni->vol;
-        ntfs_attr_search_ctx *ctx = NULL;
-        MFT_RECORD *m = NULL;
-        ATTR_RECORD *a;
-        u8 *kaddr;
-        u32 attr_len;
-        int err;
-
-        ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index "
-                        "0x%lx, from = %u, to = %u.", vi->i_ino, ni->type,
-                        page->index, from, to);
-        BUG_ON(!PageLocked(page));
-        BUG_ON(from > PAGE_CACHE_SIZE);
-        BUG_ON(to > PAGE_CACHE_SIZE);
-        BUG_ON(from > to);
-        BUG_ON(NInoMstProtected(ni));
-        /*
-         * If a previous ntfs_truncate() failed, repeat it and abort if it
-         * fails again.
-         */
-        if (unlikely(NInoTruncateFailed(ni))) {
-                down_write(&vi->i_alloc_sem);
-                err = ntfs_truncate(vi);
-                up_write(&vi->i_alloc_sem);
-                if (err || NInoTruncateFailed(ni)) {
-                        if (!err)
-                                err = -EIO;
-                        goto err_out;
-                }
-        }
-        /* If the attribute is not resident, deal with it elsewhere. */
-        if (NInoNonResident(ni)) {
-                /*
-                 * Only unnamed $DATA attributes can be compressed, encrypted,
-                 * and/or sparse.
-                 */
-                if (ni->type == AT_DATA && !ni->name_len) {
-                        /* If file is encrypted, deny access, just like NT4. */
-                        if (NInoEncrypted(ni)) {
-                                ntfs_debug("Denying write access to encrypted "
-                                                "file.");
-                                return -EACCES;
-                        }
-                        /* Compressed data streams are handled in compress.c. */
-                        if (NInoCompressed(ni)) {
-                                // TODO: Implement and replace this check with
-                                // return ntfs_write_compressed_block(page);
-                                ntfs_error(vi->i_sb, "Writing to compressed "
-                                                "files is not supported yet. "
-                                                "Sorry.");
-                                return -EOPNOTSUPP;
-                        }
-                        // TODO: Implement and remove this check.
-                        if (NInoSparse(ni)) {
-                                ntfs_error(vi->i_sb, "Writing to sparse files "
-                                                "is not supported yet. Sorry.");
-                                return -EOPNOTSUPP;
-                        }
-                }
-                /* Normal data stream. */
-                return ntfs_prepare_nonresident_write(page, from, to);
-        }
-        /*
-         * Attribute is resident, implying it is not compressed, encrypted, or
-         * sparse.
-         */
-        BUG_ON(page_has_buffers(page));
-        new_size = ((s64)page->index << PAGE_CACHE_SHIFT) + to;
-        /* If we do not need to resize the attribute allocation we are done. */
-        if (new_size <= i_size_read(vi))
-                goto done;
-        /* Map, pin, and lock the (base) mft record. */
-        if (!NInoAttr(ni))
-                base_ni = ni;
-        else
-                base_ni = ni->ext.base_ntfs_ino;
-        m = map_mft_record(base_ni);
-        if (IS_ERR(m)) {
-                err = PTR_ERR(m);
-                m = NULL;
-                ctx = NULL;
-                goto err_out;
-        }
-        ctx = ntfs_attr_get_search_ctx(base_ni, m);
-        if (unlikely(!ctx)) {
-                err = -ENOMEM;
-                goto err_out;
-        }
-        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
-                        CASE_SENSITIVE, 0, NULL, 0, ctx);
-        if (unlikely(err)) {
-                if (err == -ENOENT)
-                        err = -EIO;
-                goto err_out;
-        }
-        m = ctx->mrec;
-        a = ctx->attr;
-        /* The total length of the attribute value. */
-        attr_len = le32_to_cpu(a->data.resident.value_length);
-        /* Fix an eventual previous failure of ntfs_commit_write(). */
-        i_size = i_size_read(vi);
-        if (unlikely(attr_len > i_size)) {
-                attr_len = i_size;
-                a->data.resident.value_length = cpu_to_le32(attr_len);
-        }
-        /* If we do not need to resize the attribute allocation we are done. */
-        if (new_size <= attr_len)
-                goto done_unm;
-        /* Check if new size is allowed in $AttrDef. */
-        err = ntfs_attr_size_bounds_check(vol, ni->type, new_size);
-        if (unlikely(err)) {
-                if (err == -ERANGE) {
-                        ntfs_error(vol->sb, "Write would cause the inode "
-                                        "0x%lx to exceed the maximum size for "
-                                        "its attribute type (0x%x).  Aborting "
-                                        "write.", vi->i_ino,
-                                        le32_to_cpu(ni->type));
-                } else {
-                        ntfs_error(vol->sb, "Inode 0x%lx has unknown "
-                                        "attribute type 0x%x.  Aborting "
-                                        "write.", vi->i_ino,
-                                        le32_to_cpu(ni->type));
-                        err = -EIO;
-                }
-                goto err_out2;
-        }
-        /*
-         * Extend the attribute record to be able to store the new attribute
-         * size.
-         */
-        if (new_size >= vol->mft_record_size || ntfs_attr_record_resize(m, a,
-                        le16_to_cpu(a->data.resident.value_offset) +
-                        new_size)) {
-                /* Not enough space in the mft record. */
-                ntfs_error(vol->sb, "Not enough space in the mft record for "
-                                "the resized attribute value.  This is not "
-                                "supported yet.  Aborting write.");
-                err = -EOPNOTSUPP;
-                goto err_out2;
-        }
-        /*
-         * We have enough space in the mft record to fit the write.  This
-         * implies the attribute is smaller than the mft record and hence the
-         * attribute must be in a single page and hence page->index must be 0.
-         */
-        BUG_ON(page->index);
-        /*
-         * If the beginning of the write is past the old size, enlarge the
-         * attribute value up to the beginning of the write and fill it with
-         * zeroes.
-         */
-        if (from > attr_len) {
-                memset((u8*)a + le16_to_cpu(a->data.resident.value_offset) +
-                                attr_len, 0, from - attr_len);
-                a->data.resident.value_length = cpu_to_le32(from);
-                /* Zero the corresponding area in the page as well. */
-                if (PageUptodate(page)) {
-                        kaddr = kmap_atomic(page, KM_USER0);
-                        memset(kaddr + attr_len, 0, from - attr_len);
-                        kunmap_atomic(kaddr, KM_USER0);
-                        flush_dcache_page(page);
-                }
-        }
-        flush_dcache_mft_record_page(ctx->ntfs_ino);
-        mark_mft_record_dirty(ctx->ntfs_ino);
-done_unm:
-        ntfs_attr_put_search_ctx(ctx);
-        unmap_mft_record(base_ni);
-        /*
-         * Because resident attributes are handled by memcpy() to/from the
-         * corresponding MFT record, and because this form of i/o is byte
-         * aligned rather than block aligned, there is no need to bring the
-         * page uptodate here as in the non-resident case where we need to
-         * bring the buffers straddled by the write uptodate before
-         * generic_file_write() does the copying from userspace.
-         *
-         * We thus defer the uptodate bringing of the page region outside the
-         * region written to to ntfs_commit_write(), which makes the code
-         * simpler and saves one atomic kmap which is good.
-         */
-done:
-        ntfs_debug("Done.");
-        return 0;
-err_out:
-        if (err == -ENOMEM)
-                ntfs_warning(vi->i_sb, "Error allocating memory required to "
-                                "prepare the write.");
-        else {
-                ntfs_error(vi->i_sb, "Resident attribute prepare write failed "
-                                "with error %i.", err);
-                NVolSetErrors(vol);
-                make_bad_inode(vi);
-        }
-err_out2:
-        if (ctx)
-                ntfs_attr_put_search_ctx(ctx);
-        if (m)
-                unmap_mft_record(base_ni);
-        return err;
-}
-
-/**
- * ntfs_commit_nonresident_write -
- *
- */
-static int ntfs_commit_nonresident_write(struct page *page,
-                unsigned from, unsigned to)
-{
-        s64 pos = ((s64)page->index << PAGE_CACHE_SHIFT) + to;
-        struct inode *vi = page->mapping->host;
-        struct buffer_head *bh, *head;
-        unsigned int block_start, block_end, blocksize;
-        BOOL partial;
-
-        ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index "
-                        "0x%lx, from = %u, to = %u.", vi->i_ino,
-                        NTFS_I(vi)->type, page->index, from, to);
-        blocksize = 1 << vi->i_blkbits;
-
-        // FIXME: We need a whole slew of special cases in here for compressed
-        // files for example...
-        // For now, we know ntfs_prepare_write() would have failed so we can't
-        // get here in any of the cases which we have to special case, so we
-        // are just a ripped off, unrolled generic_commit_write().
-
-        bh = head = page_buffers(page);
-        block_start = 0;
-        partial = FALSE;
-        do {
-                block_end = block_start + blocksize;
-                if (block_end <= from || block_start >= to) {
-                        if (!buffer_uptodate(bh))
-                                partial = TRUE;
-                } else {
-                        set_buffer_uptodate(bh);
-                        mark_buffer_dirty(bh);
-                }
-        } while (block_start = block_end, (bh = bh->b_this_page) != head);
-        /*
-         * If this is a partial write which happened to make all buffers
-         * uptodate then we can optimize away a bogus ->readpage() for the next
-         * read().  Here we 'discover' whether the page went uptodate as a
-         * result of this (potentially partial) write.
-         */
-        if (!partial)
-                SetPageUptodate(page);
-        /*
-         * Not convinced about this at all.  See disparity comment above.  For
-         * now we know ntfs_prepare_write() would have failed in the write
-         * exceeds i_size case, so this will never trigger which is fine.
-         */
-        if (pos > i_size_read(vi)) {
-                ntfs_error(vi->i_sb, "Writing beyond the existing file size is "
-                                "not supported yet.  Sorry.");
-                return -EOPNOTSUPP;
-                // vi->i_size = pos;
-                // mark_inode_dirty(vi);
-        }
-        ntfs_debug("Done.");
-        return 0;
-}
-
-/**
- * ntfs_commit_write - commit the received data
- *
- * This is called from generic_file_write() with i_sem held on the inode
- * (@page->mapping->host).  The @page is locked but not kmap()ped.  The source
- * data has already been copied into the @page.  ntfs_prepare_write() has been
- * called before the data copied and it returned success so we can take the
- * results of various BUG checks and some error handling for granted.
- *
- * Need to mark modified blocks dirty so they get written out later when
- * ntfs_writepage() is invoked by the VM.
- *
- * Return 0 on success or -errno on error.
- *
- * Should be using generic_commit_write().  This marks buffers uptodate and
- * dirty, sets the page uptodate if all buffers in the page are uptodate, and
- * updates i_size if the end of io is beyond i_size.  In that case, it also
- * marks the inode dirty.
- *
- * Cannot use generic_commit_write() due to ntfs specialities but can look at
- * it for implementation guidance.
- *
- * If things have gone as outlined in ntfs_prepare_write(), then we do not
- * need to do any page content modifications here at all, except in the write
- * to resident attribute case, where we need to do the uptodate bringing here
- * which we combine with the copying into the mft record which means we save
- * one atomic kmap.
- */
-static int ntfs_commit_write(struct file *file, struct page *page,
-                unsigned from, unsigned to)
-{
-        struct inode *vi = page->mapping->host;
-        ntfs_inode *base_ni, *ni = NTFS_I(vi);
-        char *kaddr, *kattr;
-        ntfs_attr_search_ctx *ctx;
-        MFT_RECORD *m;
-        ATTR_RECORD *a;
-        u32 attr_len;
-        int err;
-
-        ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index "
-                        "0x%lx, from = %u, to = %u.", vi->i_ino, ni->type,
-                        page->index, from, to);
-        /* If the attribute is not resident, deal with it elsewhere. */
-        if (NInoNonResident(ni)) {
-                /* Only unnamed $DATA attributes can be compressed/encrypted. */
-                if (ni->type == AT_DATA && !ni->name_len) {
-                        /* Encrypted files need separate handling. */
-                        if (NInoEncrypted(ni)) {
-                                // We never get here at present!
-                                BUG();
-                        }
-                        /* Compressed data streams are handled in compress.c. */
-                        if (NInoCompressed(ni)) {
-                                // TODO: Implement this!
-                                // return ntfs_write_compressed_block(page);
-                                // We never get here at present!
-                                BUG();
-                        }
-                }
-                /* Normal data stream. */
-                return ntfs_commit_nonresident_write(page, from, to);
-        }
-        /*
-         * Attribute is resident, implying it is not compressed, encrypted, or
-         * sparse.
-         */
-        if (!NInoAttr(ni))
-                base_ni = ni;
-        else
-                base_ni = ni->ext.base_ntfs_ino;
-        /* Map, pin, and lock the mft record. */
-        m = map_mft_record(base_ni);
-        if (IS_ERR(m)) {
-                err = PTR_ERR(m);
-                m = NULL;
-                ctx = NULL;
-                goto err_out;
-        }
-        ctx = ntfs_attr_get_search_ctx(base_ni, m);
-        if (unlikely(!ctx)) {
-                err = -ENOMEM;
-                goto err_out;
-        }
-        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
-                        CASE_SENSITIVE, 0, NULL, 0, ctx);
-        if (unlikely(err)) {
-                if (err == -ENOENT)
-                        err = -EIO;
-                goto err_out;
-        }
-        a = ctx->attr;
-        /* The total length of the attribute value. */
-        attr_len = le32_to_cpu(a->data.resident.value_length);
-        BUG_ON(from > attr_len);
-        kattr = (u8*)a + le16_to_cpu(a->data.resident.value_offset);
-        kaddr = kmap_atomic(page, KM_USER0);
-        /* Copy the received data from the page to the mft record. */
-        memcpy(kattr + from, kaddr + from, to - from);
-        /* Update the attribute length if necessary. */
-        if (to > attr_len) {
-                attr_len = to;
-                a->data.resident.value_length = cpu_to_le32(attr_len);
-        }
-        /*
-         * If the page is not uptodate, bring the out of bounds area(s)
-         * uptodate by copying data from the mft record to the page.
-         */
-        if (!PageUptodate(page)) {
-                if (from > 0)
-                        memcpy(kaddr, kattr, from);
-                if (to < attr_len)
-                        memcpy(kaddr + to, kattr + to, attr_len - to);
-                /* Zero the region outside the end of the attribute value. */
-                if (attr_len < PAGE_CACHE_SIZE)
-                        memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len);
-                /*
-                 * The probability of not having done any of the above is
-                 * extremely small, so we just flush unconditionally.
-                 */
-                flush_dcache_page(page);
-                SetPageUptodate(page);
-        }
-        kunmap_atomic(kaddr, KM_USER0);
-        /* Update i_size if necessary. */
-        if (i_size_read(vi) < attr_len) {
-                unsigned long flags;
-
-                write_lock_irqsave(&ni->size_lock, flags);
-                ni->allocated_size = ni->initialized_size = attr_len;
-                i_size_write(vi, attr_len);
-                write_unlock_irqrestore(&ni->size_lock, flags);
-        }
-        /* Mark the mft record dirty, so it gets written back. */
-        flush_dcache_mft_record_page(ctx->ntfs_ino);
-        mark_mft_record_dirty(ctx->ntfs_ino);
-        ntfs_attr_put_search_ctx(ctx);
-        unmap_mft_record(base_ni);
-        ntfs_debug("Done.");
-        return 0;
-err_out:
-        if (err == -ENOMEM) {
-                ntfs_warning(vi->i_sb, "Error allocating memory required to "
-                                "commit the write.");
-                if (PageUptodate(page)) {
-                        ntfs_warning(vi->i_sb, "Page is uptodate, setting "
-                                        "dirty so the write will be retried "
-                                        "later on by the VM.");
-                        /*
-                         * Put the page on mapping->dirty_pages, but leave its
-                         * buffers' dirty state as-is.
-                         */
-                        __set_page_dirty_nobuffers(page);
-                        err = 0;
-                } else
-                        ntfs_error(vi->i_sb, "Page is not uptodate.  Written "
-                                        "data has been lost.");
-        } else {
-                ntfs_error(vi->i_sb, "Resident attribute commit write failed "
-                                "with error %i.", err);
-                NVolSetErrors(ni->vol);
-                make_bad_inode(vi);
-        }
-        if (ctx)
-                ntfs_attr_put_search_ctx(ctx);
-        if (m)
-                unmap_mft_record(base_ni);
-        return err;
-}
-
 #endif  /* NTFS_RW */
 
 /**
@@ -2377,9 +1552,6 @@ struct address_space_operations ntfs_aops = {
                                                    disk request queue. */
 #ifdef NTFS_RW
         .writepage      = ntfs_writepage,       /* Write dirty page to disk. */
-        .prepare_write  = ntfs_prepare_write,   /* Prepare page and buffers
-                                                   ready to receive data. */
-        .commit_write   = ntfs_commit_write,    /* Commit received data. */
 #endif /* NTFS_RW */
 };
 
diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c
index 3f9a4ff42ee5..eda056bac256 100644
--- a/fs/ntfs/attrib.c
+++ b/fs/ntfs/attrib.c
@@ -21,7 +21,9 @@
  */
 
 #include <linux/buffer_head.h>
+#include <linux/sched.h>
 #include <linux/swap.h>
+#include <linux/writeback.h>
 
 #include "attrib.h"
 #include "debug.h"
@@ -36,9 +38,27 @@
  * ntfs_map_runlist_nolock - map (a part of) a runlist of an ntfs inode
  * @ni:         ntfs inode for which to map (part of) a runlist
  * @vcn:        map runlist part containing this vcn
+ * @ctx:        active attribute search context if present or NULL if not
  *
  * Map the part of a runlist containing the @vcn of the ntfs inode @ni.
  *
+ * If @ctx is specified, it is an active search context of @ni and its base mft
+ * record.  This is needed when ntfs_map_runlist_nolock() encounters unmapped
+ * runlist fragments and allows their mapping.  If you do not have the mft
+ * record mapped, you can specify @ctx as NULL and ntfs_map_runlist_nolock()
+ * will perform the necessary mapping and unmapping.
+ *
+ * Note, ntfs_map_runlist_nolock() saves the state of @ctx on entry and
+ * restores it before returning.  Thus, @ctx will be left pointing to the same
+ * attribute on return as on entry.  However, the actual pointers in @ctx may
+ * point to different memory locations on return, so you must remember to reset
+ * any cached pointers from the @ctx, i.e. after the call to
+ * ntfs_map_runlist_nolock(), you will probably want to do:
+ *      m = ctx->mrec;
+ *      a = ctx->attr;
+ * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that
+ * you cache ctx->mrec in a variable @m of type MFT_RECORD *.
+ *
  * Return 0 on success and -errno on error.  There is one special error code
  * which is not an error as such.  This is -ENOENT.  It means that @vcn is out
  * of bounds of the runlist.
@@ -46,19 +66,32 @@
  * Note the runlist can be NULL after this function returns if @vcn is zero and
  * the attribute has zero allocated size, i.e. there simply is no runlist.
  *
- * Locking: - The runlist must be locked for writing.
- *          - This function modifies the runlist.
+ * WARNING: If @ctx is supplied, regardless of whether success or failure is
+ *          returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx
+ *          is no longer valid, i.e. you need to either call
+ *          ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it.
+ *          In that case PTR_ERR(@ctx->mrec) will give you the error code for
+ *          why the mapping of the old inode failed.
+ *
+ * Locking: - The runlist described by @ni must be locked for writing on entry
+ *            and is locked on return.  Note the runlist will be modified.
+ *          - If @ctx is NULL, the base mft record of @ni must not be mapped on
+ *            entry and it will be left unmapped on return.
+ *          - If @ctx is not NULL, the base mft record must be mapped on entry
+ *            and it will be left mapped on return.
  */
-int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn)
+int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn, ntfs_attr_search_ctx *ctx)
 {
         VCN end_vcn;
+        unsigned long flags;
         ntfs_inode *base_ni;
         MFT_RECORD *m;
         ATTR_RECORD *a;
-        ntfs_attr_search_ctx *ctx;
         runlist_element *rl;
-        unsigned long flags;
+        struct page *put_this_page = NULL;
         int err = 0;
+        BOOL ctx_is_temporary, ctx_needs_reset;
+        ntfs_attr_search_ctx old_ctx = { NULL, };
 
         ntfs_debug("Mapping runlist part containing vcn 0x%llx.",
                         (unsigned long long)vcn);
@@ -66,20 +99,77 @@ int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn)
                 base_ni = ni;
         else
                 base_ni = ni->ext.base_ntfs_ino;
-        m = map_mft_record(base_ni);
-        if (IS_ERR(m))
-                return PTR_ERR(m);
-        ctx = ntfs_attr_get_search_ctx(base_ni, m);
-        if (unlikely(!ctx)) {
-                err = -ENOMEM;
-                goto err_out;
+        if (!ctx) {
+                ctx_is_temporary = ctx_needs_reset = TRUE;
+                m = map_mft_record(base_ni);
+                if (IS_ERR(m))
+                        return PTR_ERR(m);
+                ctx = ntfs_attr_get_search_ctx(base_ni, m);
+                if (unlikely(!ctx)) {
+                        err = -ENOMEM;
+                        goto err_out;
+                }
+        } else {
+                VCN allocated_size_vcn;
+
+                BUG_ON(IS_ERR(ctx->mrec));
+                a = ctx->attr;
+                BUG_ON(!a->non_resident);
+                ctx_is_temporary = FALSE;
+                end_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn);
+                read_lock_irqsave(&ni->size_lock, flags);
+                allocated_size_vcn = ni->allocated_size >>
+                                ni->vol->cluster_size_bits;
+                read_unlock_irqrestore(&ni->size_lock, flags);
+                if (!a->data.non_resident.lowest_vcn && end_vcn <= 0)
+                        end_vcn = allocated_size_vcn - 1;
+                /*
+                 * If we already have the attribute extent containing @vcn in
+                 * @ctx, no need to look it up again.  We slightly cheat in
+                 * that if vcn exceeds the allocated size, we will refuse to
+                 * map the runlist below, so there is definitely no need to get
+                 * the right attribute extent.
+                 */
+                if (vcn >= allocated_size_vcn || (a->type == ni->type &&
+                                a->name_length == ni->name_len &&
+                                !memcmp((u8*)a + le16_to_cpu(a->name_offset),
+                                ni->name, ni->name_len) &&
+                                sle64_to_cpu(a->data.non_resident.lowest_vcn)
+                                <= vcn && end_vcn >= vcn))
+                        ctx_needs_reset = FALSE;
+                else {
+                        /* Save the old search context. */
+                        old_ctx = *ctx;
+                        /*
+                         * If the currently mapped (extent) inode is not the
+                         * base inode we will unmap it when we reinitialize the
+                         * search context which means we need to get a
+                         * reference to the page containing the mapped mft
+                         * record so we do not accidentally drop changes to the
+                         * mft record when it has not been marked dirty yet.
+                         */
+                        if (old_ctx.base_ntfs_ino && old_ctx.ntfs_ino !=
+                                        old_ctx.base_ntfs_ino) {
+                                put_this_page = old_ctx.ntfs_ino->page;
+                                page_cache_get(put_this_page);
+                        }
+                        /*
+                         * Reinitialize the search context so we can lookup the
+                         * needed attribute extent.
+                         */
+                        ntfs_attr_reinit_search_ctx(ctx);
+                        ctx_needs_reset = TRUE;
+                }
         }
-        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
-                        CASE_SENSITIVE, vcn, NULL, 0, ctx);
-        if (unlikely(err)) {
-                if (err == -ENOENT)
-                        err = -EIO;
-                goto err_out;
+        if (ctx_needs_reset) {
+                err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                                CASE_SENSITIVE, vcn, NULL, 0, ctx);
+                if (unlikely(err)) {
+                        if (err == -ENOENT)
+                                err = -EIO;
+                        goto err_out;
+                }
+                BUG_ON(!ctx->attr->non_resident);
         }
         a = ctx->attr;
         /*
@@ -89,11 +179,9 @@ int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn)
          * ntfs_mapping_pairs_decompress() fails.
          */
         end_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn) + 1;
-        if (unlikely(!a->data.non_resident.lowest_vcn && end_vcn <= 1)) {
-                read_lock_irqsave(&ni->size_lock, flags);
-                end_vcn = ni->allocated_size >> ni->vol->cluster_size_bits;
-                read_unlock_irqrestore(&ni->size_lock, flags);
-        }
+        if (!a->data.non_resident.lowest_vcn && end_vcn == 1)
+                end_vcn = sle64_to_cpu(a->data.non_resident.allocated_size) >>
+                                ni->vol->cluster_size_bits;
         if (unlikely(vcn >= end_vcn)) {
                 err = -ENOENT;
                 goto err_out;
@@ -104,9 +192,93 @@ int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn)
         else
                 ni->runlist.rl = rl;
 err_out:
-        if (likely(ctx))
-                ntfs_attr_put_search_ctx(ctx);
-        unmap_mft_record(base_ni);
+        if (ctx_is_temporary) {
+                if (likely(ctx))
+                        ntfs_attr_put_search_ctx(ctx);
+                unmap_mft_record(base_ni);
+        } else if (ctx_needs_reset) {
+                /*
+                 * If there is no attribute list, restoring the search context
+                 * is acomplished simply by copying the saved context back over
+                 * the caller supplied context.  If there is an attribute list,
+                 * things are more complicated as we need to deal with mapping
+                 * of mft records and resulting potential changes in pointers.
+                 */
+                if (NInoAttrList(base_ni)) {
+                        /*
+                         * If the currently mapped (extent) inode is not the
+                         * one we had before, we need to unmap it and map the
+                         * old one.
+                         */
+                        if (ctx->ntfs_ino != old_ctx.ntfs_ino) {
+                                /*
+                                 * If the currently mapped inode is not the
+                                 * base inode, unmap it.
+                                 */
+                                if (ctx->base_ntfs_ino && ctx->ntfs_ino !=
+                                                ctx->base_ntfs_ino) {
+                                        unmap_extent_mft_record(ctx->ntfs_ino);
+                                        ctx->mrec = ctx->base_mrec;
+                                        BUG_ON(!ctx->mrec);
+                                }
+                                /*
+                                 * If the old mapped inode is not the base
+                                 * inode, map it.
+                                 */
+                                if (old_ctx.base_ntfs_ino &&
+                                                old_ctx.ntfs_ino !=
+                                                old_ctx.base_ntfs_ino) {
+retry_map:
+                                        ctx->mrec = map_mft_record(
+                                                        old_ctx.ntfs_ino);
+                                        /*
+                                         * Something bad has happened.  If out
+                                         * of memory retry till it succeeds.
+                                         * Any other errors are fatal and we
+                                         * return the error code in ctx->mrec.
+                                         * Let the caller deal with it...  We
+                                         * just need to fudge things so the
+                                         * caller can reinit and/or put the
+                                         * search context safely.
+                                         */
+                                        if (IS_ERR(ctx->mrec)) {
+                                                if (PTR_ERR(ctx->mrec) ==
+                                                                -ENOMEM) {
+                                                        schedule();
+                                                        goto retry_map;
+                                                } else
+                                                        old_ctx.ntfs_ino =
+                                                                old_ctx.
+                                                                base_ntfs_ino;
+                                        }
+                                }
+                        }
+                        /* Update the changed pointers in the saved context. */
+                        if (ctx->mrec != old_ctx.mrec) {
+                                if (!IS_ERR(ctx->mrec))
+                                        old_ctx.attr = (ATTR_RECORD*)(
+                                                        (u8*)ctx->mrec +
+                                                        ((u8*)old_ctx.attr -
+                                                        (u8*)old_ctx.mrec));
+                                old_ctx.mrec = ctx->mrec;
+                        }
+                }
+                /* Restore the search context to the saved one. */
+                *ctx = old_ctx;
+                /*
+                 * We drop the reference on the page we took earlier.  In the
+                 * case that IS_ERR(ctx->mrec) is true this means we might lose
+                 * some changes to the mft record that had been made between
+                 * the last time it was marked dirty/written out and now.  This
+                 * at this stage is not a problem as the mapping error is fatal
+                 * enough that the mft record cannot be written out anyway and
+                 * the caller is very likely to shutdown the whole inode
+                 * immediately and mark the volume dirty for chkdsk to pick up
+                 * the pieces anyway.
+                 */
+                if (put_this_page)
+                        page_cache_release(put_this_page);
+        }
         return err;
 }
 
@@ -122,8 +294,8 @@ err_out:
  * of bounds of the runlist.
  *
  * Locking: - The runlist must be unlocked on entry and is unlocked on return.
- *          - This function takes the runlist lock for writing and modifies the
- *            runlist.
+ *          - This function takes the runlist lock for writing and may modify
+ *            the runlist.
  */
 int ntfs_map_runlist(ntfs_inode *ni, VCN vcn)
 {
@@ -133,7 +305,7 @@ int ntfs_map_runlist(ntfs_inode *ni, VCN vcn)
         /* Make sure someone else didn't do the work while we were sleeping. */
         if (likely(ntfs_rl_vcn_to_lcn(ni->runlist.rl, vcn) <=
                         LCN_RL_NOT_MAPPED))
-                err = ntfs_map_runlist_nolock(ni, vcn);
+                err = ntfs_map_runlist_nolock(ni, vcn, NULL);
         up_write(&ni->runlist.lock);
         return err;
 }
@@ -212,7 +384,7 @@ retry_remap:
                                 goto retry_remap;
                         }
                 }
-                err = ntfs_map_runlist_nolock(ni, vcn);
+                err = ntfs_map_runlist_nolock(ni, vcn, NULL);
                 if (!write_locked) {
                         up_write(&ni->runlist.lock);
                         down_read(&ni->runlist.lock);
@@ -236,9 +408,9 @@ retry_remap:
 
 /**
  * ntfs_attr_find_vcn_nolock - find a vcn in the runlist of an ntfs inode
- * @ni:                 ntfs inode describing the runlist to search
- * @vcn:                vcn to find
- * @write_locked:       true if the runlist is locked for writing
+ * @ni:         ntfs inode describing the runlist to search
+ * @vcn:        vcn to find
+ * @ctx:        active attribute search context if present or NULL if not
  *
  * Find the virtual cluster number @vcn in the runlist described by the ntfs
  * inode @ni and return the address of the runlist element containing the @vcn.
@@ -246,9 +418,22 @@ retry_remap:
  * If the @vcn is not mapped yet, the attempt is made to map the attribute
  * extent containing the @vcn and the vcn to lcn conversion is retried.
  *
- * If @write_locked is true the caller has locked the runlist for writing and
- * if false for reading.
- *
+ * If @ctx is specified, it is an active search context of @ni and its base mft
+ * record.  This is needed when ntfs_attr_find_vcn_nolock() encounters unmapped
+ * runlist fragments and allows their mapping.  If you do not have the mft
+ * record mapped, you can specify @ctx as NULL and ntfs_attr_find_vcn_nolock()
+ * will perform the necessary mapping and unmapping.
+ *
+ * Note, ntfs_attr_find_vcn_nolock() saves the state of @ctx on entry and
+ * restores it before returning.  Thus, @ctx will be left pointing to the same
+ * attribute on return as on entry.  However, the actual pointers in @ctx may
+ * point to different memory locations on return, so you must remember to reset
+ * any cached pointers from the @ctx, i.e. after the call to
+ * ntfs_attr_find_vcn_nolock(), you will probably want to do:
+ *      m = ctx->mrec;
+ *      a = ctx->attr;
+ * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that
+ * you cache ctx->mrec in a variable @m of type MFT_RECORD *.
  * Note you need to distinguish between the lcn of the returned runlist element
  * being >= 0 and LCN_HOLE.  In the later case you have to return zeroes on
  * read and allocate clusters on write.
@@ -263,22 +448,31 @@ retry_remap:
  *      -ENOMEM - Not enough memory to map runlist.
  *      -EIO    - Critical error (runlist/file is corrupt, i/o error, etc).
  *
- * Locking: - The runlist must be locked on entry and is left locked on return.
- *          - If @write_locked is FALSE, i.e. the runlist is locked for reading,
- *            the lock may be dropped inside the function so you cannot rely on
- *            the runlist still being the same when this function returns.
+ * WARNING: If @ctx is supplied, regardless of whether success or failure is
+ *          returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx
+ *          is no longer valid, i.e. you need to either call
+ *          ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it.
+ *          In that case PTR_ERR(@ctx->mrec) will give you the error code for
+ *          why the mapping of the old inode failed.
+ *
+ * Locking: - The runlist described by @ni must be locked for writing on entry
+ *            and is locked on return.  Note the runlist may be modified when
+ *            needed runlist fragments need to be mapped.
+ *          - If @ctx is NULL, the base mft record of @ni must not be mapped on
+ *            entry and it will be left unmapped on return.
+ *          - If @ctx is not NULL, the base mft record must be mapped on entry
+ *            and it will be left mapped on return.
  */
 runlist_element *ntfs_attr_find_vcn_nolock(ntfs_inode *ni, const VCN vcn,
-                const BOOL write_locked)
+                ntfs_attr_search_ctx *ctx)
 {
         unsigned long flags;
         runlist_element *rl;
         int err = 0;
         BOOL is_retry = FALSE;
 
-        ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, %s_locked.",
-                        ni->mft_no, (unsigned long long)vcn,
-                        write_locked ? "write" : "read");
+        ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, with%s ctx.",
+                        ni->mft_no, (unsigned long long)vcn, ctx ? "" : "out");
         BUG_ON(!ni);
         BUG_ON(!NInoNonResident(ni));
         BUG_ON(vcn < 0);
@@ -312,33 +506,22 @@ retry_remap:
         }
         if (!err && !is_retry) {
                 /*
-                 * The @vcn is in an unmapped region, map the runlist and
-                 * retry.
+                 * If the search context is invalid we cannot map the unmapped
+                 * region.
                  */
-                if (!write_locked) {
-                        up_read(&ni->runlist.lock);
-                        down_write(&ni->runlist.lock);
-                        if (unlikely(ntfs_rl_vcn_to_lcn(ni->runlist.rl, vcn) !=
-                                        LCN_RL_NOT_MAPPED)) {
-                                up_write(&ni->runlist.lock);
-                                down_read(&ni->runlist.lock);
+                if (IS_ERR(ctx->mrec))
+                        err = PTR_ERR(ctx->mrec);
+                else {
+                        /*
+                         * The @vcn is in an unmapped region, map the runlist
+                         * and retry.
+                         */
+                        err = ntfs_map_runlist_nolock(ni, vcn, ctx);
+                        if (likely(!err)) {
+                                is_retry = TRUE;
                                 goto retry_remap;
                         }
                 }
-                err = ntfs_map_runlist_nolock(ni, vcn);
-                if (!write_locked) {
-                        up_write(&ni->runlist.lock);
-                        down_read(&ni->runlist.lock);
-                }
-                if (likely(!err)) {
-                        is_retry = TRUE;
-                        goto retry_remap;
-                }
-                /*
-                 * -EINVAL coming from a failed mapping attempt is equivalent
-                 * to i/o error for us as it should not happen in our code
-                 * paths.
-                 */
                 if (err == -EINVAL)
                         err = -EIO;
         } else if (!err)
@@ -1011,6 +1194,7 @@ int ntfs_attr_lookup(const ATTR_TYPE type, const ntfschar *name,
         ntfs_inode *base_ni;
 
         ntfs_debug("Entering.");
+        BUG_ON(IS_ERR(ctx->mrec));
         if (ctx->base_ntfs_ino)
                 base_ni = ctx->base_ntfs_ino;
         else
@@ -1227,7 +1411,7 @@ int ntfs_attr_can_be_non_resident(const ntfs_volume *vol, const ATTR_TYPE type)
  */
 int ntfs_attr_can_be_resident(const ntfs_volume *vol, const ATTR_TYPE type)
 {
-        if (type == AT_INDEX_ALLOCATION || type == AT_EA)
+        if (type == AT_INDEX_ALLOCATION)
                 return -EPERM;
         return 0;
 }
@@ -1319,10 +1503,17 @@ int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a,
 /**
  * ntfs_attr_make_non_resident - convert a resident to a non-resident attribute
  * @ni:         ntfs inode describing the attribute to convert
+ * @data_size:  size of the resident data to copy to the non-resident attribute
  *
  * Convert the resident ntfs attribute described by the ntfs inode @ni to a
  * non-resident one.
  *
+ * @data_size must be equal to the attribute value size.  This is needed since
+ * we need to know the size before we can map the mft record and our callers
+ * always know it.  The reason we cannot simply read the size from the vfs
+ * inode i_size is that this is not necessarily uptodate.  This happens when
+ * ntfs_attr_make_non_resident() is called in the ->truncate call path(s).
+ *
  * Return 0 on success and -errno on error.  The following error return codes
  * are defined:
  *      -EPERM  - The attribute is not allowed to be non-resident.
@@ -1343,7 +1534,7 @@ int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a,
  *
  * Locking: - The caller must hold i_sem on the inode.
  */
-int ntfs_attr_make_non_resident(ntfs_inode *ni)
+int ntfs_attr_make_non_resident(ntfs_inode *ni, const u32 data_size)
 {
         s64 new_size;
         struct inode *vi = VFS_I(ni);
@@ -1381,11 +1572,9 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni)
          * The size needs to be aligned to a cluster boundary for allocation
          * purposes.
          */
-        new_size = (i_size_read(vi) + vol->cluster_size - 1) &
+        new_size = (data_size + vol->cluster_size - 1) &
                         ~(vol->cluster_size - 1);
         if (new_size > 0) {
-                runlist_element *rl2;
-
                 /*
                  * Will need the page later and since the page lock nests
                  * outside all ntfs locks, we need to get the page now.
@@ -1396,7 +1585,7 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni)
                         return -ENOMEM;
                 /* Start by allocating clusters to hold the attribute value. */
                 rl = ntfs_cluster_alloc(vol, 0, new_size >>
-                                vol->cluster_size_bits, -1, DATA_ZONE);
+                                vol->cluster_size_bits, -1, DATA_ZONE, TRUE);
                 if (IS_ERR(rl)) {
                         err = PTR_ERR(rl);
                         ntfs_debug("Failed to allocate cluster%s, error code "
@@ -1405,12 +1594,6 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni)
                                         err);
                         goto page_err_out;
                 }
-                /* Change the runlist terminator to LCN_ENOENT. */
-                rl2 = rl;
-                while (rl2->length)
-                        rl2++;
-                BUG_ON(rl2->lcn != LCN_RL_NOT_MAPPED);
-                rl2->lcn = LCN_ENOENT;
         } else {
                 rl = NULL;
                 page = NULL;
@@ -1473,7 +1656,7 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni)
          * attribute value.
          */
         attr_size = le32_to_cpu(a->data.resident.value_length);
-        BUG_ON(attr_size != i_size_read(vi));
+        BUG_ON(attr_size != data_size);
         if (page && !PageUptodate(page)) {
                 kaddr = kmap_atomic(page, KM_USER0);
                 memcpy(kaddr, (u8*)a +
@@ -1538,7 +1721,9 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni)
                                 ffs(ni->itype.compressed.block_size) - 1;
                 ni->itype.compressed.block_clusters = 1U <<
                                 a->data.non_resident.compression_unit;
-        }
+                vi->i_blocks = ni->itype.compressed.size >> 9;
+        } else
+                vi->i_blocks = ni->allocated_size >> 9;
         write_unlock_irqrestore(&ni->size_lock, flags);
         /*
          * This needs to be last since the address space operations ->readpage
@@ -1652,6 +1837,640 @@ page_err_out:
 }
 
 /**
+ * ntfs_attr_extend_allocation - extend the allocated space of an attribute
+ * @ni:                 ntfs inode of the attribute whose allocation to extend
+ * @new_alloc_size:     new size in bytes to which to extend the allocation to
+ * @new_data_size:      new size in bytes to which to extend the data to
+ * @data_start:         beginning of region which is required to be non-sparse
+ *
+ * Extend the allocated space of an attribute described by the ntfs inode @ni
+ * to @new_alloc_size bytes.  If @data_start is -1, the whole extension may be
+ * implemented as a hole in the file (as long as both the volume and the ntfs
+ * inode @ni have sparse support enabled).  If @data_start is >= 0, then the
+ * region between the old allocated size and @data_start - 1 may be made sparse
+ * but the regions between @data_start and @new_alloc_size must be backed by
+ * actual clusters.
+ *
+ * If @new_data_size is -1, it is ignored.  If it is >= 0, then the data size
+ * of the attribute is extended to @new_data_size.  Note that the i_size of the
+ * vfs inode is not updated.  Only the data size in the base attribute record
+ * is updated.  The caller has to update i_size separately if this is required.
+ * WARNING: It is a BUG() for @new_data_size to be smaller than the old data
+ * size as well as for @new_data_size to be greater than @new_alloc_size.
+ *
+ * For resident attributes this involves resizing the attribute record and if
+ * necessary moving it and/or other attributes into extent mft records and/or
+ * converting the attribute to a non-resident attribute which in turn involves
+ * extending the allocation of a non-resident attribute as described below.
+ *
+ * For non-resident attributes this involves allocating clusters in the data
+ * zone on the volume (except for regions that are being made sparse) and
+ * extending the run list to describe the allocated clusters as well as
+ * updating the mapping pairs array of the attribute.  This in turn involves
+ * resizing the attribute record and if necessary moving it and/or other
+ * attributes into extent mft records and/or splitting the attribute record
+ * into multiple extent attribute records.
+ *
+ * Also, the attribute list attribute is updated if present and in some of the
+ * above cases (the ones where extent mft records/attributes come into play),
+ * an attribute list attribute is created if not already present.
+ *
+ * Return the new allocated size on success and -errno on error.  In the case
+ * that an error is encountered but a partial extension at least up to
+ * @data_start (if present) is possible, the allocation is partially extended
+ * and this is returned.  This means the caller must check the returned size to
+ * determine if the extension was partial.  If @data_start is -1 then partial
+ * allocations are not performed.
+ *
+ * WARNING: Do not call ntfs_attr_extend_allocation() for $MFT/$DATA.
+ *
+ * Locking: This function takes the runlist lock of @ni for writing as well as
+ * locking the mft record of the base ntfs inode.  These locks are maintained
+ * throughout execution of the function.  These locks are required so that the
+ * attribute can be resized safely and so that it can for example be converted
+ * from resident to non-resident safely.
+ *
+ * TODO: At present attribute list attribute handling is not implemented.
+ *
+ * TODO: At present it is not safe to call this function for anything other
+ * than the $DATA attribute(s) of an uncompressed and unencrypted file.
+ */
+s64 ntfs_attr_extend_allocation(ntfs_inode *ni, s64 new_alloc_size,
+                const s64 new_data_size, const s64 data_start)
+{
+        VCN vcn;
+        s64 ll, allocated_size, start = data_start;
+        struct inode *vi = VFS_I(ni);
+        ntfs_volume *vol = ni->vol;
+        ntfs_inode *base_ni;
+        MFT_RECORD *m;
+        ATTR_RECORD *a;
+        ntfs_attr_search_ctx *ctx;
+        runlist_element *rl, *rl2;
+        unsigned long flags;
+        int err, mp_size;
+        u32 attr_len = 0; /* Silence stupid gcc warning. */
+        BOOL mp_rebuilt;
+
+#ifdef NTFS_DEBUG
+        read_lock_irqsave(&ni->size_lock, flags);
+        allocated_size = ni->allocated_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, "
+                        "old_allocated_size 0x%llx, "
+                        "new_allocated_size 0x%llx, new_data_size 0x%llx, "
+                        "data_start 0x%llx.", vi->i_ino,
+                        (unsigned)le32_to_cpu(ni->type),
+                        (unsigned long long)allocated_size,
+                        (unsigned long long)new_alloc_size,
+                        (unsigned long long)new_data_size,
+                        (unsigned long long)start);
+#endif
+retry_extend:
+        /*
+         * For non-resident attributes, @start and @new_size need to be aligned
+         * to cluster boundaries for allocation purposes.
+         */
+        if (NInoNonResident(ni)) {
+                if (start > 0)
+                        start &= ~(s64)vol->cluster_size_mask;
+                new_alloc_size = (new_alloc_size + vol->cluster_size - 1) &
+                                ~(s64)vol->cluster_size_mask;
+        }
+        BUG_ON(new_data_size >= 0 && new_data_size > new_alloc_size);
+        /* Check if new size is allowed in $AttrDef. */
+        err = ntfs_attr_size_bounds_check(vol, ni->type, new_alloc_size);
+        if (unlikely(err)) {
+                /* Only emit errors when the write will fail completely. */
+                read_lock_irqsave(&ni->size_lock, flags);
+                allocated_size = ni->allocated_size;
+                read_unlock_irqrestore(&ni->size_lock, flags);
+                if (start < 0 || start >= allocated_size) {
+                        if (err == -ERANGE) {
+                                ntfs_error(vol->sb, "Cannot extend allocation "
+                                                "of inode 0x%lx, attribute "
+                                                "type 0x%x, because the new "
+                                                "allocation would exceed the "
+                                                "maximum allowed size for "
+                                                "this attribute type.",
+                                                vi->i_ino, (unsigned)
+                                                le32_to_cpu(ni->type));
+                        } else {
+                                ntfs_error(vol->sb, "Cannot extend allocation "
+                                                "of inode 0x%lx, attribute "
+                                                "type 0x%x, because this "
+                                                "attribute type is not "
+                                                "defined on the NTFS volume.  "
+                                                "Possible corruption!  You "
+                                                "should run chkdsk!",
+                                                vi->i_ino, (unsigned)
+                                                le32_to_cpu(ni->type));
+                        }
+                }
+                /* Translate error code to be POSIX conformant for write(2). */
+                if (err == -ERANGE)
+                        err = -EFBIG;
+                else
+                        err = -EIO;
+                return err;
+        }
+        if (!NInoAttr(ni))
+                base_ni = ni;
+        else
+                base_ni = ni->ext.base_ntfs_ino;
+        /*
+         * We will be modifying both the runlist (if non-resident) and the mft
+         * record so lock them both down.
+         */
+        down_write(&ni->runlist.lock);
+        m = map_mft_record(base_ni);
+        if (IS_ERR(m)) {
+                err = PTR_ERR(m);
+                m = NULL;
+                ctx = NULL;
+                goto err_out;
+        }
+        ctx = ntfs_attr_get_search_ctx(base_ni, m);
+        if (unlikely(!ctx)) {
+                err = -ENOMEM;
+                goto err_out;
+        }
+        read_lock_irqsave(&ni->size_lock, flags);
+        allocated_size = ni->allocated_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        /*
+         * If non-resident, seek to the last extent.  If resident, there is
+         * only one extent, so seek to that.
+         */
+        vcn = NInoNonResident(ni) ? allocated_size >> vol->cluster_size_bits :
+                        0;
+        /*
+         * Abort if someone did the work whilst we waited for the locks.  If we
+         * just converted the attribute from resident to non-resident it is
+         * likely that exactly this has happened already.  We cannot quite
+         * abort if we need to update the data size.
+         */
+        if (unlikely(new_alloc_size <= allocated_size)) {
+                ntfs_debug("Allocated size already exceeds requested size.");
+                new_alloc_size = allocated_size;
+                if (new_data_size < 0)
+                        goto done;
+                /*
+                 * We want the first attribute extent so that we can update the
+                 * data size.
+                 */
+                vcn = 0;
+        }
+        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                        CASE_SENSITIVE, vcn, NULL, 0, ctx);
+        if (unlikely(err)) {
+                if (err == -ENOENT)
+                        err = -EIO;
+                goto err_out;
+        }
+        m = ctx->mrec;
+        a = ctx->attr;
+        /* Use goto to reduce indentation. */
+        if (a->non_resident)
+                goto do_non_resident_extend;
+        BUG_ON(NInoNonResident(ni));
+        /* The total length of the attribute value. */
+        attr_len = le32_to_cpu(a->data.resident.value_length);
+        /*
+         * Extend the attribute record to be able to store the new attribute
+         * size.  ntfs_attr_record_resize() will not do anything if the size is
+         * not changing.
+         */
+        if (new_alloc_size < vol->mft_record_size &&
+                        !ntfs_attr_record_resize(m, a,
+                        le16_to_cpu(a->data.resident.value_offset) +
+                        new_alloc_size)) {
+                /* The resize succeeded! */
+                write_lock_irqsave(&ni->size_lock, flags);
+                ni->allocated_size = le32_to_cpu(a->length) -
+                                le16_to_cpu(a->data.resident.value_offset);
+                write_unlock_irqrestore(&ni->size_lock, flags);
+                if (new_data_size >= 0) {
+                        BUG_ON(new_data_size < attr_len);
+                        a->data.resident.value_length =
+                                        cpu_to_le32((u32)new_data_size);
+                }
+                goto flush_done;
+        }
+        /*
+         * We have to drop all the locks so we can call
+         * ntfs_attr_make_non_resident().  This could be optimised by try-
+         * locking the first page cache page and only if that fails dropping
+         * the locks, locking the page, and redoing all the locking and
+         * lookups.  While this would be a huge optimisation, it is not worth
+         * it as this is definitely a slow code path.
+         */
+        ntfs_attr_put_search_ctx(ctx);
+        unmap_mft_record(base_ni);
+        up_write(&ni->runlist.lock);
+        /*
+         * Not enough space in the mft record, try to make the attribute
+         * non-resident and if successful restart the extension process.
+         */
+        err = ntfs_attr_make_non_resident(ni, attr_len);
+        if (likely(!err))
+                goto retry_extend;
+        /*
+         * Could not make non-resident.  If this is due to this not being
+         * permitted for this attribute type or there not being enough space,
+         * try to make other attributes non-resident.  Otherwise fail.
+         */
+        if (unlikely(err != -EPERM && err != -ENOSPC)) {
+                /* Only emit errors when the write will fail completely. */
+                read_lock_irqsave(&ni->size_lock, flags);
+                allocated_size = ni->allocated_size;
+                read_unlock_irqrestore(&ni->size_lock, flags);
+                if (start < 0 || start >= allocated_size)
+                        ntfs_error(vol->sb, "Cannot extend allocation of "
+                                        "inode 0x%lx, attribute type 0x%x, "
+                                        "because the conversion from resident "
+                                        "to non-resident attribute failed "
+                                        "with error code %i.", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                if (err != -ENOMEM)
+                        err = -EIO;
+                goto conv_err_out;
+        }
+        /* TODO: Not implemented from here, abort. */
+        read_lock_irqsave(&ni->size_lock, flags);
+        allocated_size = ni->allocated_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        if (start < 0 || start >= allocated_size) {
+                if (err == -ENOSPC)
+                        ntfs_error(vol->sb, "Not enough space in the mft "
+                                        "record/on disk for the non-resident "
+                                        "attribute value.  This case is not "
+                                        "implemented yet.");
+                else /* if (err == -EPERM) */
+                        ntfs_error(vol->sb, "This attribute type may not be "
+                                        "non-resident.  This case is not "
+                                        "implemented yet.");
+        }
+        err = -EOPNOTSUPP;
+        goto conv_err_out;
+#if 0
+        // TODO: Attempt to make other attributes non-resident.
+        if (!err)
+                goto do_resident_extend;
+        /*
+         * Both the attribute list attribute and the standard information
+         * attribute must remain in the base inode.  Thus, if this is one of
+         * these attributes, we have to try to move other attributes out into
+         * extent mft records instead.
+         */
+        if (ni->type == AT_ATTRIBUTE_LIST ||
+                        ni->type == AT_STANDARD_INFORMATION) {
+                // TODO: Attempt to move other attributes into extent mft
+                // records.
+                err = -EOPNOTSUPP;
+                if (!err)
+                        goto do_resident_extend;
+                goto err_out;
+        }
+        // TODO: Attempt to move this attribute to an extent mft record, but
+        // only if it is not already the only attribute in an mft record in
+        // which case there would be nothing to gain.
+        err = -EOPNOTSUPP;
+        if (!err)
+                goto do_resident_extend;
+        /* There is nothing we can do to make enough space. )-: */
+        goto err_out;
+#endif
+do_non_resident_extend:
+        BUG_ON(!NInoNonResident(ni));
+        if (new_alloc_size == allocated_size) {
+                BUG_ON(vcn);
+                goto alloc_done;
+        }
+        /*
+         * If the data starts after the end of the old allocation, this is a
+         * $DATA attribute and sparse attributes are enabled on the volume and
+         * for this inode, then create a sparse region between the old
+         * allocated size and the start of the data.  Otherwise simply proceed
+         * with filling the whole space between the old allocated size and the
+         * new allocated size with clusters.
+         */
+        if ((start >= 0 && start <= allocated_size) || ni->type != AT_DATA ||
+                        !NVolSparseEnabled(vol) || NInoSparseDisabled(ni))
+                goto skip_sparse;
+        // TODO: This is not implemented yet.  We just fill in with real
+        // clusters for now...
+        ntfs_debug("Inserting holes is not-implemented yet.  Falling back to "
+                        "allocating real clusters instead.");
+skip_sparse:
+        rl = ni->runlist.rl;
+        if (likely(rl)) {
+                /* Seek to the end of the runlist. */
+                while (rl->length)
+                        rl++;
+        }
+        /* If this attribute extent is not mapped, map it now. */
+        if (unlikely(!rl || rl->lcn == LCN_RL_NOT_MAPPED ||
+                        (rl->lcn == LCN_ENOENT && rl > ni->runlist.rl &&
+                        (rl-1)->lcn == LCN_RL_NOT_MAPPED))) {
+                if (!rl && !allocated_size)
+                        goto first_alloc;
+                rl = ntfs_mapping_pairs_decompress(vol, a, ni->runlist.rl);
+                if (IS_ERR(rl)) {
+                        err = PTR_ERR(rl);
+                        if (start < 0 || start >= allocated_size)
+                                ntfs_error(vol->sb, "Cannot extend allocation "
+                                                "of inode 0x%lx, attribute "
+                                                "type 0x%x, because the "
+                                                "mapping of a runlist "
+                                                "fragment failed with error "
+                                                "code %i.", vi->i_ino,
+                                                (unsigned)le32_to_cpu(ni->type),
+                                                err);
+                        if (err != -ENOMEM)
+                                err = -EIO;
+                        goto err_out;
+                }
+                ni->runlist.rl = rl;
+                /* Seek to the end of the runlist. */
+                while (rl->length)
+                        rl++;
+        }
+        /*
+         * We now know the runlist of the last extent is mapped and @rl is at
+         * the end of the runlist.  We want to begin allocating clusters
+         * starting at the last allocated cluster to reduce fragmentation.  If
+         * there are no valid LCNs in the attribute we let the cluster
+         * allocator choose the starting cluster.
+         */
+        /* If the last LCN is a hole or simillar seek back to last real LCN. */
+        while (rl->lcn < 0 && rl > ni->runlist.rl)
+                rl--;
+first_alloc:
+        // FIXME: Need to implement partial allocations so at least part of the
+        // write can be performed when start >= 0.  (Needed for POSIX write(2)
+        // conformance.)
+        rl2 = ntfs_cluster_alloc(vol, allocated_size >> vol->cluster_size_bits,
+                        (new_alloc_size - allocated_size) >>
+                        vol->cluster_size_bits, (rl && (rl->lcn >= 0)) ?
+                        rl->lcn + rl->length : -1, DATA_ZONE, TRUE);
+        if (IS_ERR(rl2)) {
+                err = PTR_ERR(rl2);
+                if (start < 0 || start >= allocated_size)
+                        ntfs_error(vol->sb, "Cannot extend allocation of "
+                                        "inode 0x%lx, attribute type 0x%x, "
+                                        "because the allocation of clusters "
+                                        "failed with error code %i.", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                if (err != -ENOMEM && err != -ENOSPC)
+                        err = -EIO;
+                goto err_out;
+        }
+        rl = ntfs_runlists_merge(ni->runlist.rl, rl2);
+        if (IS_ERR(rl)) {
+                err = PTR_ERR(rl);
+                if (start < 0 || start >= allocated_size)
+                        ntfs_error(vol->sb, "Cannot extend allocation of "
+                                        "inode 0x%lx, attribute type 0x%x, "
+                                        "because the runlist merge failed "
+                                        "with error code %i.", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                if (err != -ENOMEM)
+                        err = -EIO;
+                if (ntfs_cluster_free_from_rl(vol, rl2)) {
+                        ntfs_error(vol->sb, "Failed to release allocated "
+                                        "cluster(s) in error code path.  Run "
+                                        "chkdsk to recover the lost "
+                                        "cluster(s).");
+                        NVolSetErrors(vol);
+                }
+                ntfs_free(rl2);
+                goto err_out;
+        }
+        ni->runlist.rl = rl;
+        ntfs_debug("Allocated 0x%llx clusters.", (long long)(new_alloc_size -
+                        allocated_size) >> vol->cluster_size_bits);
+        /* Find the runlist element with which the attribute extent starts. */
+        ll = sle64_to_cpu(a->data.non_resident.lowest_vcn);
+        rl2 = ntfs_rl_find_vcn_nolock(rl, ll);
+        BUG_ON(!rl2);
+        BUG_ON(!rl2->length);
+        BUG_ON(rl2->lcn < LCN_HOLE);
+        mp_rebuilt = FALSE;
+        /* Get the size for the new mapping pairs array for this extent. */
+        mp_size = ntfs_get_size_for_mapping_pairs(vol, rl2, ll, -1);
+        if (unlikely(mp_size <= 0)) {
+                err = mp_size;
+                if (start < 0 || start >= allocated_size)
+                        ntfs_error(vol->sb, "Cannot extend allocation of "
+                                        "inode 0x%lx, attribute type 0x%x, "
+                                        "because determining the size for the "
+                                        "mapping pairs failed with error code "
+                                        "%i.", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                err = -EIO;
+                goto undo_alloc;
+        }
+        /* Extend the attribute record to fit the bigger mapping pairs array. */
+        attr_len = le32_to_cpu(a->length);
+        err = ntfs_attr_record_resize(m, a, mp_size +
+                        le16_to_cpu(a->data.non_resident.mapping_pairs_offset));
+        if (unlikely(err)) {
+                BUG_ON(err != -ENOSPC);
+                // TODO: Deal with this by moving this extent to a new mft
+                // record or by starting a new extent in a new mft record,
+                // possibly by extending this extent partially and filling it
+                // and creating a new extent for the remainder, or by making
+                // other attributes non-resident and/or by moving other
+                // attributes out of this mft record.
+                if (start < 0 || start >= allocated_size)
+                        ntfs_error(vol->sb, "Not enough space in the mft "
+                                        "record for the extended attribute "
+                                        "record.  This case is not "
+                                        "implemented yet.");
+                err = -EOPNOTSUPP;
+                goto undo_alloc;
+        }
+        mp_rebuilt = TRUE;
+        /* Generate the mapping pairs array directly into the attr record. */
+        err = ntfs_mapping_pairs_build(vol, (u8*)a +
+                        le16_to_cpu(a->data.non_resident.mapping_pairs_offset),
+                        mp_size, rl2, ll, -1, NULL);
+        if (unlikely(err)) {
+                if (start < 0 || start >= allocated_size)
+                        ntfs_error(vol->sb, "Cannot extend allocation of "
+                                        "inode 0x%lx, attribute type 0x%x, "
+                                        "because building the mapping pairs "
+                                        "failed with error code %i.", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                err = -EIO;
+                goto undo_alloc;
+        }
+        /* Update the highest_vcn. */
+        a->data.non_resident.highest_vcn = cpu_to_sle64((new_alloc_size >>
+                        vol->cluster_size_bits) - 1);
+        /*
+         * We now have extended the allocated size of the attribute.  Reflect
+         * this in the ntfs_inode structure and the attribute record.
+         */
+        if (a->data.non_resident.lowest_vcn) {
+                /*
+                 * We are not in the first attribute extent, switch to it, but
+                 * first ensure the changes will make it to disk later.
+                 */
+                flush_dcache_mft_record_page(ctx->ntfs_ino);
+                mark_mft_record_dirty(ctx->ntfs_ino);
+                ntfs_attr_reinit_search_ctx(ctx);
+                err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                                CASE_SENSITIVE, 0, NULL, 0, ctx);
+                if (unlikely(err))
+                        goto restore_undo_alloc;
+                /* @m is not used any more so no need to set it. */
+                a = ctx->attr;
+        }
+        write_lock_irqsave(&ni->size_lock, flags);
+        ni->allocated_size = new_alloc_size;
+        a->data.non_resident.allocated_size = cpu_to_sle64(new_alloc_size);
+        /*
+         * FIXME: This would fail if @ni is a directory, $MFT, or an index,
+         * since those can have sparse/compressed set.  For example can be
+         * set compressed even though it is not compressed itself and in that
+         * case the bit means that files are to be created compressed in the
+         * directory...  At present this is ok as this code is only called for
+         * regular files, and only for their $DATA attribute(s).
+         * FIXME: The calculation is wrong if we created a hole above.  For now
+         * it does not matter as we never create holes.
+         */
+        if (NInoSparse(ni) || NInoCompressed(ni)) {
+                ni->itype.compressed.size += new_alloc_size - allocated_size;
+                a->data.non_resident.compressed_size =
+                                cpu_to_sle64(ni->itype.compressed.size);
+                vi->i_blocks = ni->itype.compressed.size >> 9;
+        } else
+                vi->i_blocks = new_alloc_size >> 9;
+        write_unlock_irqrestore(&ni->size_lock, flags);
+alloc_done:
+        if (new_data_size >= 0) {
+                BUG_ON(new_data_size <
+                                sle64_to_cpu(a->data.non_resident.data_size));
+                a->data.non_resident.data_size = cpu_to_sle64(new_data_size);
+        }
+flush_done:
+        /* Ensure the changes make it to disk. */
+        flush_dcache_mft_record_page(ctx->ntfs_ino);
+        mark_mft_record_dirty(ctx->ntfs_ino);
+done:
+        ntfs_attr_put_search_ctx(ctx);
+        unmap_mft_record(base_ni);
+        up_write(&ni->runlist.lock);
+        ntfs_debug("Done, new_allocated_size 0x%llx.",
+                        (unsigned long long)new_alloc_size);
+        return new_alloc_size;
+restore_undo_alloc:
+        if (start < 0 || start >= allocated_size)
+                ntfs_error(vol->sb, "Cannot complete extension of allocation "
+                                "of inode 0x%lx, attribute type 0x%x, because "
+                                "lookup of first attribute extent failed with "
+                                "error code %i.", vi->i_ino,
+                                (unsigned)le32_to_cpu(ni->type), err);
+        if (err == -ENOENT)
+                err = -EIO;
+        ntfs_attr_reinit_search_ctx(ctx);
+        if (ntfs_attr_lookup(ni->type, ni->name, ni->name_len, CASE_SENSITIVE,
+                        allocated_size >> vol->cluster_size_bits, NULL, 0,
+                        ctx)) {
+                ntfs_error(vol->sb, "Failed to find last attribute extent of "
+                                "attribute in error code path.  Run chkdsk to "
+                                "recover.");
+                write_lock_irqsave(&ni->size_lock, flags);
+                ni->allocated_size = new_alloc_size;
+                /*
+                 * FIXME: This would fail if @ni is a directory...  See above.
+                 * FIXME: The calculation is wrong if we created a hole above.
+                 * For now it does not matter as we never create holes.
+                 */
+                if (NInoSparse(ni) || NInoCompressed(ni)) {
+                        ni->itype.compressed.size += new_alloc_size -
+                                        allocated_size;
+                        vi->i_blocks = ni->itype.compressed.size >> 9;
+                } else
+                        vi->i_blocks = new_alloc_size >> 9;
+                write_unlock_irqrestore(&ni->size_lock, flags);
+                ntfs_attr_put_search_ctx(ctx);
+                unmap_mft_record(base_ni);
+                up_write(&ni->runlist.lock);
+                /*
+                 * The only thing that is now wrong is the allocated size of the
+                 * base attribute extent which chkdsk should be able to fix.
+                 */
+                NVolSetErrors(vol);
+                return err;
+        }
+        ctx->attr->data.non_resident.highest_vcn = cpu_to_sle64(
+                        (allocated_size >> vol->cluster_size_bits) - 1);
+undo_alloc:
+        ll = allocated_size >> vol->cluster_size_bits;
+        if (ntfs_cluster_free(ni, ll, -1, ctx) < 0) {
+                ntfs_error(vol->sb, "Failed to release allocated cluster(s) "
+                                "in error code path.  Run chkdsk to recover "
+                                "the lost cluster(s).");
+                NVolSetErrors(vol);
+        }
+        m = ctx->mrec;
+        a = ctx->attr;
+        /*
+         * If the runlist truncation fails and/or the search context is no
+         * longer valid, we cannot resize the attribute record or build the
+         * mapping pairs array thus we mark the inode bad so that no access to
+         * the freed clusters can happen.
+         */
+        if (ntfs_rl_truncate_nolock(vol, &ni->runlist, ll) || IS_ERR(m)) {
+                ntfs_error(vol->sb, "Failed to %s in error code path.  Run "
+                                "chkdsk to recover.", IS_ERR(m) ?
+                                "restore attribute search context" :
+                                "truncate attribute runlist");
+                make_bad_inode(vi);
+                make_bad_inode(VFS_I(base_ni));
+                NVolSetErrors(vol);
+        } else if (mp_rebuilt) {
+                if (ntfs_attr_record_resize(m, a, attr_len)) {
+                        ntfs_error(vol->sb, "Failed to restore attribute "
+                                        "record in error code path.  Run "
+                                        "chkdsk to recover.");
+                        make_bad_inode(vi);
+                        make_bad_inode(VFS_I(base_ni));
+                        NVolSetErrors(vol);
+                } else /* if (success) */ {
+                        if (ntfs_mapping_pairs_build(vol, (u8*)a + le16_to_cpu(
+                                        a->data.non_resident.
+                                        mapping_pairs_offset), attr_len -
+                                        le16_to_cpu(a->data.non_resident.
+                                        mapping_pairs_offset), rl2, ll, -1,
+                                        NULL)) {
+                                ntfs_error(vol->sb, "Failed to restore "
+                                                "mapping pairs array in error "
+                                                "code path.  Run chkdsk to "
+                                                "recover.");
+                                make_bad_inode(vi);
+                                make_bad_inode(VFS_I(base_ni));
+                                NVolSetErrors(vol);
+                        }
+                        flush_dcache_mft_record_page(ctx->ntfs_ino);
+                        mark_mft_record_dirty(ctx->ntfs_ino);
+                }
+        }
+err_out:
+        if (ctx)
+                ntfs_attr_put_search_ctx(ctx);
+        if (m)
+                unmap_mft_record(base_ni);
+        up_write(&ni->runlist.lock);
+conv_err_out:
+        ntfs_debug("Failed.  Returning error code %i.", err);
+        return err;
+}
+
+/**
  * ntfs_attr_set - fill (a part of) an attribute with a byte
  * @ni:         ntfs inode describing the attribute to fill
  * @ofs:        offset inside the attribute at which to start to fill
@@ -1773,6 +2592,8 @@ int ntfs_attr_set(ntfs_inode *ni, const s64 ofs, const s64 cnt, const u8 val)
                 /* Finally unlock and release the page. */
                 unlock_page(page);
                 page_cache_release(page);
+                balance_dirty_pages_ratelimited(mapping);
+                cond_resched();
         }
         /* If there is a last partial page, need to do it the slow way. */
         if (end_ofs) {
diff --git a/fs/ntfs/attrib.h b/fs/ntfs/attrib.h
index 0618ed6fd7b3..9074886b44ba 100644
--- a/fs/ntfs/attrib.h
+++ b/fs/ntfs/attrib.h
@@ -60,14 +60,15 @@ typedef struct {
         ATTR_RECORD *base_attr;
 } ntfs_attr_search_ctx;
 
-extern int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn);
+extern int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn,
+                ntfs_attr_search_ctx *ctx);
 extern int ntfs_map_runlist(ntfs_inode *ni, VCN vcn);
 
 extern LCN ntfs_attr_vcn_to_lcn_nolock(ntfs_inode *ni, const VCN vcn,
                 const BOOL write_locked);
 
 extern runlist_element *ntfs_attr_find_vcn_nolock(ntfs_inode *ni,
-                const VCN vcn, const BOOL write_locked);
+                const VCN vcn, ntfs_attr_search_ctx *ctx);
 
 int ntfs_attr_lookup(const ATTR_TYPE type, const ntfschar *name,
                 const u32 name_len, const IGNORE_CASE_BOOL ic,
@@ -102,7 +103,10 @@ extern int ntfs_attr_record_resize(MFT_RECORD *m, ATTR_RECORD *a, u32 new_size);
 extern int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a,
                 const u32 new_size);
 
-extern int ntfs_attr_make_non_resident(ntfs_inode *ni);
+extern int ntfs_attr_make_non_resident(ntfs_inode *ni, const u32 data_size);
+
+extern s64 ntfs_attr_extend_allocation(ntfs_inode *ni, s64 new_alloc_size,
+                const s64 new_data_size, const s64 data_start);
 
 extern int ntfs_attr_set(ntfs_inode *ni, const s64 ofs, const s64 cnt,
                 const u8 val);
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index be9fd1dd423d..cf3e6ced2d01 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -19,11 +19,24 @@
  * Foundation,Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  */
 
-#include <linux/pagemap.h>
 #include <linux/buffer_head.h>
+#include <linux/pagemap.h>
+#include <linux/pagevec.h>
+#include <linux/sched.h>
+#include <linux/swap.h>
+#include <linux/uio.h>
+#include <linux/writeback.h>
 
+#include <asm/page.h>
+#include <asm/uaccess.h>
+
+#include "attrib.h"
+#include "bitmap.h"
 #include "inode.h"
 #include "debug.h"
+#include "lcnalloc.h"
+#include "malloc.h"
+#include "mft.h"
 #include "ntfs.h"
 
 /**
@@ -56,6 +69,2184 @@ static int ntfs_file_open(struct inode *vi, struct file *filp)
 #ifdef NTFS_RW
 
 /**
+ * ntfs_attr_extend_initialized - extend the initialized size of an attribute
+ * @ni:                 ntfs inode of the attribute to extend
+ * @new_init_size:      requested new initialized size in bytes
+ * @cached_page:        store any allocated but unused page here
+ * @lru_pvec:           lru-buffering pagevec of the caller
+ *
+ * Extend the initialized size of an attribute described by the ntfs inode @ni
+ * to @new_init_size bytes.  This involves zeroing any non-sparse space between
+ * the old initialized size and @new_init_size both in the page cache and on
+ * disk (if relevant complete pages are already uptodate in the page cache then
+ * these are simply marked dirty).
+ *
+ * As a side-effect, the file size (vfs inode->i_size) may be incremented as,
+ * in the resident attribute case, it is tied to the initialized size and, in
+ * the non-resident attribute case, it may not fall below the initialized size.
+ *
+ * Note that if the attribute is resident, we do not need to touch the page
+ * cache at all.  This is because if the page cache page is not uptodate we
+ * bring it uptodate later, when doing the write to the mft record since we
+ * then already have the page mapped.  And if the page is uptodate, the
+ * non-initialized region will already have been zeroed when the page was
+ * brought uptodate and the region may in fact already have been overwritten
+ * with new data via mmap() based writes, so we cannot just zero it.  And since
+ * POSIX specifies that the behaviour of resizing a file whilst it is mmap()ped
+ * is unspecified, we choose not to do zeroing and thus we do not need to touch
+ * the page at all.  For a more detailed explanation see ntfs_truncate() in
+ * fs/ntfs/inode.c.
+ *
+ * @cached_page and @lru_pvec are just optimizations for dealing with multiple
+ * pages.
+ *
+ * Return 0 on success and -errno on error.  In the case that an error is
+ * encountered it is possible that the initialized size will already have been
+ * incremented some way towards @new_init_size but it is guaranteed that if
+ * this is the case, the necessary zeroing will also have happened and that all
+ * metadata is self-consistent.
+ *
+ * Locking: i_sem on the vfs inode corrseponsind to the ntfs inode @ni must be
+ *          held by the caller.
+ */
+static int ntfs_attr_extend_initialized(ntfs_inode *ni, const s64 new_init_size,
+                struct page **cached_page, struct pagevec *lru_pvec)
+{
+        s64 old_init_size;
+        loff_t old_i_size;
+        pgoff_t index, end_index;
+        unsigned long flags;
+        struct inode *vi = VFS_I(ni);
+        ntfs_inode *base_ni;
+        MFT_RECORD *m = NULL;
+        ATTR_RECORD *a;
+        ntfs_attr_search_ctx *ctx = NULL;
+        struct address_space *mapping;
+        struct page *page = NULL;
+        u8 *kattr;
+        int err;
+        u32 attr_len;
+
+        read_lock_irqsave(&ni->size_lock, flags);
+        old_init_size = ni->initialized_size;
+        old_i_size = i_size_read(vi);
+        BUG_ON(new_init_size > ni->allocated_size);
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, "
+                        "old_initialized_size 0x%llx, "
+                        "new_initialized_size 0x%llx, i_size 0x%llx.",
+                        vi->i_ino, (unsigned)le32_to_cpu(ni->type),
+                        (unsigned long long)old_init_size,
+                        (unsigned long long)new_init_size, old_i_size);
+        if (!NInoAttr(ni))
+                base_ni = ni;
+        else
+                base_ni = ni->ext.base_ntfs_ino;
+        /* Use goto to reduce indentation and we need the label below anyway. */
+        if (NInoNonResident(ni))
+                goto do_non_resident_extend;
+        BUG_ON(old_init_size != old_i_size);
+        m = map_mft_record(base_ni);
+        if (IS_ERR(m)) {
+                err = PTR_ERR(m);
+                m = NULL;
+                goto err_out;
+        }
+        ctx = ntfs_attr_get_search_ctx(base_ni, m);
+        if (unlikely(!ctx)) {
+                err = -ENOMEM;
+                goto err_out;
+        }
+        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                        CASE_SENSITIVE, 0, NULL, 0, ctx);
+        if (unlikely(err)) {
+                if (err == -ENOENT)
+                        err = -EIO;
+                goto err_out;
+        }
+        m = ctx->mrec;
+        a = ctx->attr;
+        BUG_ON(a->non_resident);
+        /* The total length of the attribute value. */
+        attr_len = le32_to_cpu(a->data.resident.value_length);
+        BUG_ON(old_i_size != (loff_t)attr_len);
+        /*
+         * Do the zeroing in the mft record and update the attribute size in
+         * the mft record.
+         */
+        kattr = (u8*)a + le16_to_cpu(a->data.resident.value_offset);
+        memset(kattr + attr_len, 0, new_init_size - attr_len);
+        a->data.resident.value_length = cpu_to_le32((u32)new_init_size);
+        /* Finally, update the sizes in the vfs and ntfs inodes. */
+        write_lock_irqsave(&ni->size_lock, flags);
+        i_size_write(vi, new_init_size);
+        ni->initialized_size = new_init_size;
+        write_unlock_irqrestore(&ni->size_lock, flags);
+        goto done;
+do_non_resident_extend:
+        /*
+         * If the new initialized size @new_init_size exceeds the current file
+         * size (vfs inode->i_size), we need to extend the file size to the
+         * new initialized size.
+         */
+        if (new_init_size > old_i_size) {
+                m = map_mft_record(base_ni);
+                if (IS_ERR(m)) {
+                        err = PTR_ERR(m);
+                        m = NULL;
+                        goto err_out;
+                }
+                ctx = ntfs_attr_get_search_ctx(base_ni, m);
+                if (unlikely(!ctx)) {
+                        err = -ENOMEM;
+                        goto err_out;
+                }
+                err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                                CASE_SENSITIVE, 0, NULL, 0, ctx);
+                if (unlikely(err)) {
+                        if (err == -ENOENT)
+                                err = -EIO;
+                        goto err_out;
+                }
+                m = ctx->mrec;
+                a = ctx->attr;
+                BUG_ON(!a->non_resident);
+                BUG_ON(old_i_size != (loff_t)
+                                sle64_to_cpu(a->data.non_resident.data_size));
+                a->data.non_resident.data_size = cpu_to_sle64(new_init_size);
+                flush_dcache_mft_record_page(ctx->ntfs_ino);
+                mark_mft_record_dirty(ctx->ntfs_ino);
+                /* Update the file size in the vfs inode. */
+                i_size_write(vi, new_init_size);
+                ntfs_attr_put_search_ctx(ctx);
+                ctx = NULL;
+                unmap_mft_record(base_ni);
+                m = NULL;
+        }
+        mapping = vi->i_mapping;
+        index = old_init_size >> PAGE_CACHE_SHIFT;
+        end_index = (new_init_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+        do {
+                /*
+                 * Read the page.  If the page is not present, this will zero
+                 * the uninitialized regions for us.
+                 */
+                page = read_cache_page(mapping, index,
+                                (filler_t*)mapping->a_ops->readpage, NULL);
+                if (IS_ERR(page)) {
+                        err = PTR_ERR(page);
+                        goto init_err_out;
+                }
+                wait_on_page_locked(page);
+                if (unlikely(!PageUptodate(page) || PageError(page))) {
+                        page_cache_release(page);
+                        err = -EIO;
+                        goto init_err_out;
+                }
+                /*
+                 * Update the initialized size in the ntfs inode.  This is
+                 * enough to make ntfs_writepage() work.
+                 */
+                write_lock_irqsave(&ni->size_lock, flags);
+                ni->initialized_size = (index + 1) << PAGE_CACHE_SHIFT;
+                if (ni->initialized_size > new_init_size)
+                        ni->initialized_size = new_init_size;
+                write_unlock_irqrestore(&ni->size_lock, flags);
+                /* Set the page dirty so it gets written out. */
+                set_page_dirty(page);
+                page_cache_release(page);
+                /*
+                 * Play nice with the vm and the rest of the system.  This is
+                 * very much needed as we can potentially be modifying the
+                 * initialised size from a very small value to a really huge
+                 * value, e.g.
+                 *      f = open(somefile, O_TRUNC);
+                 *      truncate(f, 10GiB);
+                 *      seek(f, 10GiB);
+                 *      write(f, 1);
+                 * And this would mean we would be marking dirty hundreds of
+                 * thousands of pages or as in the above example more than
+                 * two and a half million pages!
+                 *
+                 * TODO: For sparse pages could optimize this workload by using
+                 * the FsMisc / MiscFs page bit as a "PageIsSparse" bit.  This
+                 * would be set in readpage for sparse pages and here we would
+                 * not need to mark dirty any pages which have this bit set.
+                 * The only caveat is that we have to clear the bit everywhere
+                 * where we allocate any clusters that lie in the page or that
+                 * contain the page.
+                 *
+                 * TODO: An even greater optimization would be for us to only
+                 * call readpage() on pages which are not in sparse regions as
+                 * determined from the runlist.  This would greatly reduce the
+                 * number of pages we read and make dirty in the case of sparse
+                 * files.
+                 */
+                balance_dirty_pages_ratelimited(mapping);
+                cond_resched();
+        } while (++index < end_index);
+        read_lock_irqsave(&ni->size_lock, flags);
+        BUG_ON(ni->initialized_size != new_init_size);
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        /* Now bring in sync the initialized_size in the mft record. */
+        m = map_mft_record(base_ni);
+        if (IS_ERR(m)) {
+                err = PTR_ERR(m);
+                m = NULL;
+                goto init_err_out;
+        }
+        ctx = ntfs_attr_get_search_ctx(base_ni, m);
+        if (unlikely(!ctx)) {
+                err = -ENOMEM;
+                goto init_err_out;
+        }
+        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                        CASE_SENSITIVE, 0, NULL, 0, ctx);
+        if (unlikely(err)) {
+                if (err == -ENOENT)
+                        err = -EIO;
+                goto init_err_out;
+        }
+        m = ctx->mrec;
+        a = ctx->attr;
+        BUG_ON(!a->non_resident);
+        a->data.non_resident.initialized_size = cpu_to_sle64(new_init_size);
+done:
+        flush_dcache_mft_record_page(ctx->ntfs_ino);
+        mark_mft_record_dirty(ctx->ntfs_ino);
+        if (ctx)
+                ntfs_attr_put_search_ctx(ctx);
+        if (m)
+                unmap_mft_record(base_ni);
+        ntfs_debug("Done, initialized_size 0x%llx, i_size 0x%llx.",
+                        (unsigned long long)new_init_size, i_size_read(vi));
+        return 0;
+init_err_out:
+        write_lock_irqsave(&ni->size_lock, flags);
+        ni->initialized_size = old_init_size;
+        write_unlock_irqrestore(&ni->size_lock, flags);
+err_out:
+        if (ctx)
+                ntfs_attr_put_search_ctx(ctx);
+        if (m)
+                unmap_mft_record(base_ni);
+        ntfs_debug("Failed.  Returning error code %i.", err);
+        return err;
+}
+
+/**
+ * ntfs_fault_in_pages_readable -
+ *
+ * Fault a number of userspace pages into pagetables.
+ *
+ * Unlike include/linux/pagemap.h::fault_in_pages_readable(), this one copes
+ * with more than two userspace pages as well as handling the single page case
+ * elegantly.
+ *
+ * If you find this difficult to understand, then think of the while loop being
+ * the following code, except that we do without the integer variable ret:
+ *
+ *      do {
+ *              ret = __get_user(c, uaddr);
+ *              uaddr += PAGE_SIZE;
+ *      } while (!ret && uaddr < end);
+ *
+ * Note, the final __get_user() may well run out-of-bounds of the user buffer,
+ * but _not_ out-of-bounds of the page the user buffer belongs to, and since
+ * this is only a read and not a write, and since it is still in the same page,
+ * it should not matter and this makes the code much simpler.
+ */
+static inline void ntfs_fault_in_pages_readable(const char __user *uaddr,
+                int bytes)
+{
+        const char __user *end;
+        volatile char c;
+
+        /* Set @end to the first byte outside the last page we care about. */
+        end = (const char __user*)PAGE_ALIGN((ptrdiff_t __user)uaddr + bytes);
+
+        while (!__get_user(c, uaddr) && (uaddr += PAGE_SIZE, uaddr < end))
+                ;
+}
+
+/**
+ * ntfs_fault_in_pages_readable_iovec -
+ *
+ * Same as ntfs_fault_in_pages_readable() but operates on an array of iovecs.
+ */
+static inline void ntfs_fault_in_pages_readable_iovec(const struct iovec *iov,
+                size_t iov_ofs, int bytes)
+{
+        do {
+                const char __user *buf;
+                unsigned len;
+
+                buf = iov->iov_base + iov_ofs;
+                len = iov->iov_len - iov_ofs;
+                if (len > bytes)
+                        len = bytes;
+                ntfs_fault_in_pages_readable(buf, len);
+                bytes -= len;
+                iov++;
+                iov_ofs = 0;
+        } while (bytes);
+}
+
+/**
+ * __ntfs_grab_cache_pages - obtain a number of locked pages
+ * @mapping:    address space mapping from which to obtain page cache pages
+ * @index:      starting index in @mapping at which to begin obtaining pages
+ * @nr_pages:   number of page cache pages to obtain
+ * @pages:      array of pages in which to return the obtained page cache pages
+ * @cached_page: allocated but as yet unused page
+ * @lru_pvec:   lru-buffering pagevec of caller
+ *
+ * Obtain @nr_pages locked page cache pages from the mapping @maping and
+ * starting at index @index.
+ *
+ * If a page is newly created, increment its refcount and add it to the
+ * caller's lru-buffering pagevec @lru_pvec.
+ *
+ * This is the same as mm/filemap.c::__grab_cache_page(), except that @nr_pages
+ * are obtained at once instead of just one page and that 0 is returned on
+ * success and -errno on error.
+ *
+ * Note, the page locks are obtained in ascending page index order.
+ */
+static inline int __ntfs_grab_cache_pages(struct address_space *mapping,
+                pgoff_t index, const unsigned nr_pages, struct page **pages,
+                struct page **cached_page, struct pagevec *lru_pvec)
+{
+        int err, nr;
+
+        BUG_ON(!nr_pages);
+        err = nr = 0;
+        do {
+                pages[nr] = find_lock_page(mapping, index);
+                if (!pages[nr]) {
+                        if (!*cached_page) {
+                                *cached_page = page_cache_alloc(mapping);
+                                if (unlikely(!*cached_page)) {
+                                        err = -ENOMEM;
+                                        goto err_out;
+                                }
+                        }
+                        err = add_to_page_cache(*cached_page, mapping, index,
+                                        GFP_KERNEL);
+                        if (unlikely(err)) {
+                                if (err == -EEXIST)
+                                        continue;
+                                goto err_out;
+                        }
+                        pages[nr] = *cached_page;
+                        page_cache_get(*cached_page);
+                        if (unlikely(!pagevec_add(lru_pvec, *cached_page)))
+                                __pagevec_lru_add(lru_pvec);
+                        *cached_page = NULL;
+                }
+                index++;
+                nr++;
+        } while (nr < nr_pages);
+out:
+        return err;
+err_out:
+        while (nr > 0) {
+                unlock_page(pages[--nr]);
+                page_cache_release(pages[nr]);
+        }
+        goto out;
+}
+
+static inline int ntfs_submit_bh_for_read(struct buffer_head *bh)
+{
+        lock_buffer(bh);
+        get_bh(bh);
+        bh->b_end_io = end_buffer_read_sync;
+        return submit_bh(READ, bh);
+}
+
+/**
+ * ntfs_prepare_pages_for_non_resident_write - prepare pages for receiving data
+ * @pages:      array of destination pages
+ * @nr_pages:   number of pages in @pages
+ * @pos:        byte position in file at which the write begins
+ * @bytes:      number of bytes to be written
+ *
+ * This is called for non-resident attributes from ntfs_file_buffered_write()
+ * with i_sem held on the inode (@pages[0]->mapping->host).  There are
+ * @nr_pages pages in @pages which are locked but not kmap()ped.  The source
+ * data has not yet been copied into the @pages.
+ * 
+ * Need to fill any holes with actual clusters, allocate buffers if necessary,
+ * ensure all the buffers are mapped, and bring uptodate any buffers that are
+ * only partially being written to.
+ *
+ * If @nr_pages is greater than one, we are guaranteed that the cluster size is
+ * greater than PAGE_CACHE_SIZE, that all pages in @pages are entirely inside
+ * the same cluster and that they are the entirety of that cluster, and that
+ * the cluster is sparse, i.e. we need to allocate a cluster to fill the hole.
+ *
+ * i_size is not to be modified yet.
+ *
+ * Return 0 on success or -errno on error.
+ */
+static int ntfs_prepare_pages_for_non_resident_write(struct page **pages,
+                unsigned nr_pages, s64 pos, size_t bytes)
+{
+        VCN vcn, highest_vcn = 0, cpos, cend, bh_cpos, bh_cend;
+        LCN lcn;
+        s64 bh_pos, vcn_len, end, initialized_size;
+        sector_t lcn_block;
+        struct page *page;
+        struct inode *vi;
+        ntfs_inode *ni, *base_ni = NULL;
+        ntfs_volume *vol;
+        runlist_element *rl, *rl2;
+        struct buffer_head *bh, *head, *wait[2], **wait_bh = wait;
+        ntfs_attr_search_ctx *ctx = NULL;
+        MFT_RECORD *m = NULL;
+        ATTR_RECORD *a = NULL;
+        unsigned long flags;
+        u32 attr_rec_len = 0;
+        unsigned blocksize, u;
+        int err, mp_size;
+        BOOL rl_write_locked, was_hole, is_retry;
+        unsigned char blocksize_bits;
+        struct {
+                u8 runlist_merged:1;
+                u8 mft_attr_mapped:1;
+                u8 mp_rebuilt:1;
+                u8 attr_switched:1;
+        } status = { 0, 0, 0, 0 };
+
+        BUG_ON(!nr_pages);
+        BUG_ON(!pages);
+        BUG_ON(!*pages);
+        vi = pages[0]->mapping->host;
+        ni = NTFS_I(vi);
+        vol = ni->vol;
+        ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, start page "
+                        "index 0x%lx, nr_pages 0x%x, pos 0x%llx, bytes 0x%zx.",
+                        vi->i_ino, ni->type, pages[0]->index, nr_pages,
+                        (long long)pos, bytes);
+        blocksize_bits = vi->i_blkbits;
+        blocksize = 1 << blocksize_bits;
+        u = 0;
+        do {
+                struct page *page = pages[u];
+                /*
+                 * create_empty_buffers() will create uptodate/dirty buffers if
+                 * the page is uptodate/dirty.
+                 */
+                if (!page_has_buffers(page)) {
+                        create_empty_buffers(page, blocksize, 0);
+                        if (unlikely(!page_has_buffers(page)))
+                                return -ENOMEM;
+                }
+        } while (++u < nr_pages);
+        rl_write_locked = FALSE;
+        rl = NULL;
+        err = 0;
+        vcn = lcn = -1;
+        vcn_len = 0;
+        lcn_block = -1;
+        was_hole = FALSE;
+        cpos = pos >> vol->cluster_size_bits;
+        end = pos + bytes;
+        cend = (end + vol->cluster_size - 1) >> vol->cluster_size_bits;
+        /*
+         * Loop over each page and for each page over each buffer.  Use goto to
+         * reduce indentation.
+         */
+        u = 0;
+do_next_page:
+        page = pages[u];
+        bh_pos = (s64)page->index << PAGE_CACHE_SHIFT;
+        bh = head = page_buffers(page);
+        do {
+                VCN cdelta;
+                s64 bh_end;
+                unsigned bh_cofs;
+
+                /* Clear buffer_new on all buffers to reinitialise state. */
+                if (buffer_new(bh))
+                        clear_buffer_new(bh);
+                bh_end = bh_pos + blocksize;
+                bh_cpos = bh_pos >> vol->cluster_size_bits;
+                bh_cofs = bh_pos & vol->cluster_size_mask;
+                if (buffer_mapped(bh)) {
+                        /*
+                         * The buffer is already mapped.  If it is uptodate,
+                         * ignore it.
+                         */
+                        if (buffer_uptodate(bh))
+                                continue;
+                        /*
+                         * The buffer is not uptodate.  If the page is uptodate
+                         * set the buffer uptodate and otherwise ignore it.
+                         */
+                        if (PageUptodate(page)) {
+                                set_buffer_uptodate(bh);
+                                continue;
+                        }
+                        /*
+                         * Neither the page nor the buffer are uptodate.  If
+                         * the buffer is only partially being written to, we
+                         * need to read it in before the write, i.e. now.
+                         */
+                        if ((bh_pos < pos && bh_end > pos) ||
+                                        (bh_pos < end && bh_end > end)) {
+                                /*
+                                 * If the buffer is fully or partially within
+                                 * the initialized size, do an actual read.
+                                 * Otherwise, simply zero the buffer.
+                                 */
+                                read_lock_irqsave(&ni->size_lock, flags);
+                                initialized_size = ni->initialized_size;
+                                read_unlock_irqrestore(&ni->size_lock, flags);
+                                if (bh_pos < initialized_size) {
+                                        ntfs_submit_bh_for_read(bh);
+                                        *wait_bh++ = bh;
+                                } else {
+                                        u8 *kaddr = kmap_atomic(page, KM_USER0);
+                                        memset(kaddr + bh_offset(bh), 0,
+                                                        blocksize);
+                                        kunmap_atomic(kaddr, KM_USER0);
+                                        flush_dcache_page(page);
+                                        set_buffer_uptodate(bh);
+                                }
+                        }
+                        continue;
+                }
+                /* Unmapped buffer.  Need to map it. */
+                bh->b_bdev = vol->sb->s_bdev;
+                /*
+                 * If the current buffer is in the same clusters as the map
+                 * cache, there is no need to check the runlist again.  The
+                 * map cache is made up of @vcn, which is the first cached file
+                 * cluster, @vcn_len which is the number of cached file
+                 * clusters, @lcn is the device cluster corresponding to @vcn,
+                 * and @lcn_block is the block number corresponding to @lcn.
+                 */
+                cdelta = bh_cpos - vcn;
+                if (likely(!cdelta || (cdelta > 0 && cdelta < vcn_len))) {
+map_buffer_cached:
+                        BUG_ON(lcn < 0);
+                        bh->b_blocknr = lcn_block +
+                                        (cdelta << (vol->cluster_size_bits -
+                                        blocksize_bits)) +
+                                        (bh_cofs >> blocksize_bits);
+                        set_buffer_mapped(bh);
+                        /*
+                         * If the page is uptodate so is the buffer.  If the
+                         * buffer is fully outside the write, we ignore it if
+                         * it was already allocated and we mark it dirty so it
+                         * gets written out if we allocated it.  On the other
+                         * hand, if we allocated the buffer but we are not
+                         * marking it dirty we set buffer_new so we can do
+                         * error recovery.
+                         */
+                        if (PageUptodate(page)) {
+                                if (!buffer_uptodate(bh))
+                                        set_buffer_uptodate(bh);
+                                if (unlikely(was_hole)) {
+                                        /* We allocated the buffer. */
+                                        unmap_underlying_metadata(bh->b_bdev,
+                                                        bh->b_blocknr);
+                                        if (bh_end <= pos || bh_pos >= end)
+                                                mark_buffer_dirty(bh);
+                                        else
+                                                set_buffer_new(bh);
+                                }
+                                continue;
+                        }
+                        /* Page is _not_ uptodate. */
+                        if (likely(!was_hole)) {
+                                /*
+                                 * Buffer was already allocated.  If it is not
+                                 * uptodate and is only partially being written
+                                 * to, we need to read it in before the write,
+                                 * i.e. now.
+                                 */
+                                if (!buffer_uptodate(bh) && ((bh_pos < pos &&
+                                                bh_end > pos) ||
+                                                (bh_end > end &&
+                                                bh_end > end))) {
+                                        /*
+                                         * If the buffer is fully or partially
+                                         * within the initialized size, do an
+                                         * actual read.  Otherwise, simply zero
+                                         * the buffer.
+                                         */
+                                        read_lock_irqsave(&ni->size_lock,
+                                                        flags);
+                                        initialized_size = ni->initialized_size;
+                                        read_unlock_irqrestore(&ni->size_lock,
+                                                        flags);
+                                        if (bh_pos < initialized_size) {
+                                                ntfs_submit_bh_for_read(bh);
+                                                *wait_bh++ = bh;
+                                        } else {
+                                                u8 *kaddr = kmap_atomic(page,
+                                                                KM_USER0);
+                                                memset(kaddr + bh_offset(bh),
+                                                                0, blocksize);
+                                                kunmap_atomic(kaddr, KM_USER0);
+                                                flush_dcache_page(page);
+                                                set_buffer_uptodate(bh);
+                                        }
+                                }
+                                continue;
+                        }
+                        /* We allocated the buffer. */
+                        unmap_underlying_metadata(bh->b_bdev, bh->b_blocknr);
+                        /*
+                         * If the buffer is fully outside the write, zero it,
+                         * set it uptodate, and mark it dirty so it gets
+                         * written out.  If it is partially being written to,
+                         * zero region surrounding the write but leave it to
+                         * commit write to do anything else.  Finally, if the
+                         * buffer is fully being overwritten, do nothing.
+                         */
+                        if (bh_end <= pos || bh_pos >= end) {
+                                if (!buffer_uptodate(bh)) {
+                                        u8 *kaddr = kmap_atomic(page, KM_USER0);
+                                        memset(kaddr + bh_offset(bh), 0,
+                                                        blocksize);
+                                        kunmap_atomic(kaddr, KM_USER0);
+                                        flush_dcache_page(page);
+                                        set_buffer_uptodate(bh);
+                                }
+                                mark_buffer_dirty(bh);
+                                continue;
+                        }
+                        set_buffer_new(bh);
+                        if (!buffer_uptodate(bh) &&
+                                        (bh_pos < pos || bh_end > end)) {
+                                u8 *kaddr;
+                                unsigned pofs;
+                                        
+                                kaddr = kmap_atomic(page, KM_USER0);
+                                if (bh_pos < pos) {
+                                        pofs = bh_pos & ~PAGE_CACHE_MASK;
+                                        memset(kaddr + pofs, 0, pos - bh_pos);
+                                }
+                                if (bh_end > end) {
+                                        pofs = end & ~PAGE_CACHE_MASK;
+                                        memset(kaddr + pofs, 0, bh_end - end);
+                                }
+                                kunmap_atomic(kaddr, KM_USER0);
+                                flush_dcache_page(page);
+                        }
+                        continue;
+                }
+                /*
+                 * Slow path: this is the first buffer in the cluster.  If it
+                 * is outside allocated size and is not uptodate, zero it and
+                 * set it uptodate.
+                 */
+                read_lock_irqsave(&ni->size_lock, flags);
+                initialized_size = ni->allocated_size;
+                read_unlock_irqrestore(&ni->size_lock, flags);
+                if (bh_pos > initialized_size) {
+                        if (PageUptodate(page)) {
+                                if (!buffer_uptodate(bh))
+                                        set_buffer_uptodate(bh);
+                        } else if (!buffer_uptodate(bh)) {
+                                u8 *kaddr = kmap_atomic(page, KM_USER0);
+                                memset(kaddr + bh_offset(bh), 0, blocksize);
+                                kunmap_atomic(kaddr, KM_USER0);
+                                flush_dcache_page(page);
+                                set_buffer_uptodate(bh);
+                        }
+                        continue;
+                }
+                is_retry = FALSE;
+                if (!rl) {
+                        down_read(&ni->runlist.lock);
+retry_remap:
+                        rl = ni->runlist.rl;
+                }
+                if (likely(rl != NULL)) {
+                        /* Seek to element containing target cluster. */
+                        while (rl->length && rl[1].vcn <= bh_cpos)
+                                rl++;
+                        lcn = ntfs_rl_vcn_to_lcn(rl, bh_cpos);
+                        if (likely(lcn >= 0)) {
+                                /*
+                                 * Successful remap, setup the map cache and
+                                 * use that to deal with the buffer.
+                                 */
+                                was_hole = FALSE;
+                                vcn = bh_cpos;
+                                vcn_len = rl[1].vcn - vcn;
+                                lcn_block = lcn << (vol->cluster_size_bits -
+                                                blocksize_bits);
+                                cdelta = 0;
+                                /*
+                                 * If the number of remaining clusters in the
+                                 * @pages is smaller or equal to the number of
+                                 * cached clusters, unlock the runlist as the
+                                 * map cache will be used from now on.
+                                 */
+                                if (likely(vcn + vcn_len >= cend)) {
+                                        if (rl_write_locked) {
+                                                up_write(&ni->runlist.lock);
+                                                rl_write_locked = FALSE;
+                                        } else
+                                                up_read(&ni->runlist.lock);
+                                        rl = NULL;
+                                }
+                                goto map_buffer_cached;
+                        }
+                } else
+                        lcn = LCN_RL_NOT_MAPPED;
+                /*
+                 * If it is not a hole and not out of bounds, the runlist is
+                 * probably unmapped so try to map it now.
+                 */
+                if (unlikely(lcn != LCN_HOLE && lcn != LCN_ENOENT)) {
+                        if (likely(!is_retry && lcn == LCN_RL_NOT_MAPPED)) {
+                                /* Attempt to map runlist. */
+                                if (!rl_write_locked) {
+                                        /*
+                                         * We need the runlist locked for
+                                         * writing, so if it is locked for
+                                         * reading relock it now and retry in
+                                         * case it changed whilst we dropped
+                                         * the lock.
+                                         */
+                                        up_read(&ni->runlist.lock);
+                                        down_write(&ni->runlist.lock);
+                                        rl_write_locked = TRUE;
+                                        goto retry_remap;
+                                }
+                                err = ntfs_map_runlist_nolock(ni, bh_cpos,
+                                                NULL);
+                                if (likely(!err)) {
+                                        is_retry = TRUE;
+                                        goto retry_remap;
+                                }
+                                /*
+                                 * If @vcn is out of bounds, pretend @lcn is
+                                 * LCN_ENOENT.  As long as the buffer is out
+                                 * of bounds this will work fine.
+                                 */
+                                if (err == -ENOENT) {
+                                        lcn = LCN_ENOENT;
+                                        err = 0;
+                                        goto rl_not_mapped_enoent;
+                                }
+                        } else
+                                err = -EIO;
+                        /* Failed to map the buffer, even after retrying. */
+                        bh->b_blocknr = -1;
+                        ntfs_error(vol->sb, "Failed to write to inode 0x%lx, "
+                                        "attribute type 0x%x, vcn 0x%llx, "
+                                        "vcn offset 0x%x, because its "
+                                        "location on disk could not be "
+                                        "determined%s (error code %i).",
+                                        ni->mft_no, ni->type,
+                                        (unsigned long long)bh_cpos,
+                                        (unsigned)bh_pos &
+                                        vol->cluster_size_mask,
+                                        is_retry ? " even after retrying" : "",
+                                        err);
+                        break;
+                }
+rl_not_mapped_enoent:
+                /*
+                 * The buffer is in a hole or out of bounds.  We need to fill
+                 * the hole, unless the buffer is in a cluster which is not
+                 * touched by the write, in which case we just leave the buffer
+                 * unmapped.  This can only happen when the cluster size is
+                 * less than the page cache size.
+                 */
+                if (unlikely(vol->cluster_size < PAGE_CACHE_SIZE)) {
+                        bh_cend = (bh_end + vol->cluster_size - 1) >>
+                                        vol->cluster_size_bits;
+                        if ((bh_cend <= cpos || bh_cpos >= cend)) {
+                                bh->b_blocknr = -1;
+                                /*
+                                 * If the buffer is uptodate we skip it.  If it
+                                 * is not but the page is uptodate, we can set
+                                 * the buffer uptodate.  If the page is not
+                                 * uptodate, we can clear the buffer and set it
+                                 * uptodate.  Whether this is worthwhile is
+                                 * debatable and this could be removed.
+                                 */
+                                if (PageUptodate(page)) {
+                                        if (!buffer_uptodate(bh))
+                                                set_buffer_uptodate(bh);
+                                } else if (!buffer_uptodate(bh)) {
+                                        u8 *kaddr = kmap_atomic(page, KM_USER0);
+                                        memset(kaddr + bh_offset(bh), 0,
+                                                        blocksize);
+                                        kunmap_atomic(kaddr, KM_USER0);
+                                        flush_dcache_page(page);
+                                        set_buffer_uptodate(bh);
+                                }
+                                continue;
+                        }
+                }
+                /*
+                 * Out of bounds buffer is invalid if it was not really out of
+                 * bounds.
+                 */
+                BUG_ON(lcn != LCN_HOLE);
+                /*
+                 * We need the runlist locked for writing, so if it is locked
+                 * for reading relock it now and retry in case it changed
+                 * whilst we dropped the lock.
+                 */
+                BUG_ON(!rl);
+                if (!rl_write_locked) {
+                        up_read(&ni->runlist.lock);
+                        down_write(&ni->runlist.lock);
+                        rl_write_locked = TRUE;
+                        goto retry_remap;
+                }
+                /* Find the previous last allocated cluster. */
+                BUG_ON(rl->lcn != LCN_HOLE);
+                lcn = -1;
+                rl2 = rl;
+                while (--rl2 >= ni->runlist.rl) {
+                        if (rl2->lcn >= 0) {
+                                lcn = rl2->lcn + rl2->length;
+                                break;
+                        }
+                }
+                rl2 = ntfs_cluster_alloc(vol, bh_cpos, 1, lcn, DATA_ZONE,
+                                FALSE);
+                if (IS_ERR(rl2)) {
+                        err = PTR_ERR(rl2);
+                        ntfs_debug("Failed to allocate cluster, error code %i.",
+                                        err);
+                        break;
+                }
+                lcn = rl2->lcn;
+                rl = ntfs_runlists_merge(ni->runlist.rl, rl2);
+                if (IS_ERR(rl)) {
+                        err = PTR_ERR(rl);
+                        if (err != -ENOMEM)
+                                err = -EIO;
+                        if (ntfs_cluster_free_from_rl(vol, rl2)) {
+                                ntfs_error(vol->sb, "Failed to release "
+                                                "allocated cluster in error "
+                                                "code path.  Run chkdsk to "
+                                                "recover the lost cluster.");
+                                NVolSetErrors(vol);
+                        }
+                        ntfs_free(rl2);
+                        break;
+                }
+                ni->runlist.rl = rl;
+                status.runlist_merged = 1;
+                ntfs_debug("Allocated cluster, lcn 0x%llx.", lcn);
+                /* Map and lock the mft record and get the attribute record. */
+                if (!NInoAttr(ni))
+                        base_ni = ni;
+                else
+                        base_ni = ni->ext.base_ntfs_ino;
+                m = map_mft_record(base_ni);
+                if (IS_ERR(m)) {
+                        err = PTR_ERR(m);
+                        break;
+                }
+                ctx = ntfs_attr_get_search_ctx(base_ni, m);
+                if (unlikely(!ctx)) {
+                        err = -ENOMEM;
+                        unmap_mft_record(base_ni);
+                        break;
+                }
+                status.mft_attr_mapped = 1;
+                err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                                CASE_SENSITIVE, bh_cpos, NULL, 0, ctx);
+                if (unlikely(err)) {
+                        if (err == -ENOENT)
+                                err = -EIO;
+                        break;
+                }
+                m = ctx->mrec;
+                a = ctx->attr;
+                /*
+                 * Find the runlist element with which the attribute extent
+                 * starts.  Note, we cannot use the _attr_ version because we
+                 * have mapped the mft record.  That is ok because we know the
+                 * runlist fragment must be mapped already to have ever gotten
+                 * here, so we can just use the _rl_ version.
+                 */
+                vcn = sle64_to_cpu(a->data.non_resident.lowest_vcn);
+                rl2 = ntfs_rl_find_vcn_nolock(rl, vcn);
+                BUG_ON(!rl2);
+                BUG_ON(!rl2->length);
+                BUG_ON(rl2->lcn < LCN_HOLE);
+                highest_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn);
+                /*
+                 * If @highest_vcn is zero, calculate the real highest_vcn
+                 * (which can really be zero).
+                 */
+                if (!highest_vcn)
+                        highest_vcn = (sle64_to_cpu(
+                                        a->data.non_resident.allocated_size) >>
+                                        vol->cluster_size_bits) - 1;
+                /*
+                 * Determine the size of the mapping pairs array for the new
+                 * extent, i.e. the old extent with the hole filled.
+                 */
+                mp_size = ntfs_get_size_for_mapping_pairs(vol, rl2, vcn,
+                                highest_vcn);
+                if (unlikely(mp_size <= 0)) {
+                        if (!(err = mp_size))
+                                err = -EIO;
+                        ntfs_debug("Failed to get size for mapping pairs "
+                                        "array, error code %i.", err);
+                        break;
+                }
+                /*
+                 * Resize the attribute record to fit the new mapping pairs
+                 * array.
+                 */
+                attr_rec_len = le32_to_cpu(a->length);
+                err = ntfs_attr_record_resize(m, a, mp_size + le16_to_cpu(
+                                a->data.non_resident.mapping_pairs_offset));
+                if (unlikely(err)) {
+                        BUG_ON(err != -ENOSPC);
+                        // TODO: Deal with this by using the current attribute
+                        // and fill it with as much of the mapping pairs
+                        // array as possible.  Then loop over each attribute
+                        // extent rewriting the mapping pairs arrays as we go
+                        // along and if when we reach the end we have not
+                        // enough space, try to resize the last attribute
+                        // extent and if even that fails, add a new attribute
+                        // extent.
+                        // We could also try to resize at each step in the hope
+                        // that we will not need to rewrite every single extent.
+                        // Note, we may need to decompress some extents to fill
+                        // the runlist as we are walking the extents...
+                        ntfs_error(vol->sb, "Not enough space in the mft "
+                                        "record for the extended attribute "
+                                        "record.  This case is not "
+                                        "implemented yet.");
+                        err = -EOPNOTSUPP;
+                        break ;
+                }
+                status.mp_rebuilt = 1;
+                /*
+                 * Generate the mapping pairs array directly into the attribute
+                 * record.
+                 */
+                err = ntfs_mapping_pairs_build(vol, (u8*)a + le16_to_cpu(
+                                a->data.non_resident.mapping_pairs_offset),
+                                mp_size, rl2, vcn, highest_vcn, NULL);
+                if (unlikely(err)) {
+                        ntfs_error(vol->sb, "Cannot fill hole in inode 0x%lx, "
+                                        "attribute type 0x%x, because building "
+                                        "the mapping pairs failed with error "
+                                        "code %i.", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                        err = -EIO;
+                        break;
+                }
+                /* Update the highest_vcn but only if it was not set. */
+                if (unlikely(!a->data.non_resident.highest_vcn))
+                        a->data.non_resident.highest_vcn =
+                                        cpu_to_sle64(highest_vcn);
+                /*
+                 * If the attribute is sparse/compressed, update the compressed
+                 * size in the ntfs_inode structure and the attribute record.
+                 */
+                if (likely(NInoSparse(ni) || NInoCompressed(ni))) {
+                        /*
+                         * If we are not in the first attribute extent, switch
+                         * to it, but first ensure the changes will make it to
+                         * disk later.
+                         */
+                        if (a->data.non_resident.lowest_vcn) {
+                                flush_dcache_mft_record_page(ctx->ntfs_ino);
+                                mark_mft_record_dirty(ctx->ntfs_ino);
+                                ntfs_attr_reinit_search_ctx(ctx);
+                                err = ntfs_attr_lookup(ni->type, ni->name,
+                                                ni->name_len, CASE_SENSITIVE,
+                                                0, NULL, 0, ctx);
+                                if (unlikely(err)) {
+                                        status.attr_switched = 1;
+                                        break;
+                                }
+                                /* @m is not used any more so do not set it. */
+                                a = ctx->attr;
+                        }
+                        write_lock_irqsave(&ni->size_lock, flags);
+                        ni->itype.compressed.size += vol->cluster_size;
+                        a->data.non_resident.compressed_size =
+                                        cpu_to_sle64(ni->itype.compressed.size);
+                        write_unlock_irqrestore(&ni->size_lock, flags);
+                }
+                /* Ensure the changes make it to disk. */
+                flush_dcache_mft_record_page(ctx->ntfs_ino);
+                mark_mft_record_dirty(ctx->ntfs_ino);
+                ntfs_attr_put_search_ctx(ctx);
+                unmap_mft_record(base_ni);
+                /* Successfully filled the hole. */
+                status.runlist_merged = 0;
+                status.mft_attr_mapped = 0;
+                status.mp_rebuilt = 0;
+                /* Setup the map cache and use that to deal with the buffer. */
+                was_hole = TRUE;
+                vcn = bh_cpos;
+                vcn_len = 1;
+                lcn_block = lcn << (vol->cluster_size_bits - blocksize_bits);
+                cdelta = 0;
+                /*
+                 * If the number of remaining clusters in the @pages is smaller
+                 * or equal to the number of cached clusters, unlock the
+                 * runlist as the map cache will be used from now on.
+                 */
+                if (likely(vcn + vcn_len >= cend)) {
+                        up_write(&ni->runlist.lock);
+                        rl_write_locked = FALSE;
+                        rl = NULL;
+                }
+                goto map_buffer_cached;
+        } while (bh_pos += blocksize, (bh = bh->b_this_page) != head);
+        /* If there are no errors, do the next page. */
+        if (likely(!err && ++u < nr_pages))
+                goto do_next_page;
+        /* If there are no errors, release the runlist lock if we took it. */
+        if (likely(!err)) {
+                if (unlikely(rl_write_locked)) {
+                        up_write(&ni->runlist.lock);
+                        rl_write_locked = FALSE;
+                } else if (unlikely(rl))
+                        up_read(&ni->runlist.lock);
+                rl = NULL;
+        }
+        /* If we issued read requests, let them complete. */
+        read_lock_irqsave(&ni->size_lock, flags);
+        initialized_size = ni->initialized_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        while (wait_bh > wait) {
+                bh = *--wait_bh;
+                wait_on_buffer(bh);
+                if (likely(buffer_uptodate(bh))) {
+                        page = bh->b_page;
+                        bh_pos = ((s64)page->index << PAGE_CACHE_SHIFT) +
+                                        bh_offset(bh);
+                        /*
+                         * If the buffer overflows the initialized size, need
+                         * to zero the overflowing region.
+                         */
+                        if (unlikely(bh_pos + blocksize > initialized_size)) {
+                                u8 *kaddr;
+                                int ofs = 0;
+
+                                if (likely(bh_pos < initialized_size))
+                                        ofs = initialized_size - bh_pos;
+                                kaddr = kmap_atomic(page, KM_USER0);
+                                memset(kaddr + bh_offset(bh) + ofs, 0,
+                                                blocksize - ofs);
+                                kunmap_atomic(kaddr, KM_USER0);
+                                flush_dcache_page(page);
+                        }
+                } else /* if (unlikely(!buffer_uptodate(bh))) */
+                        err = -EIO;
+        }
+        if (likely(!err)) {
+                /* Clear buffer_new on all buffers. */
+                u = 0;
+                do {
+                        bh = head = page_buffers(pages[u]);
+                        do {
+                                if (buffer_new(bh))
+                                        clear_buffer_new(bh);
+                        } while ((bh = bh->b_this_page) != head);
+                } while (++u < nr_pages);
+                ntfs_debug("Done.");
+                return err;
+        }
+        if (status.attr_switched) {
+                /* Get back to the attribute extent we modified. */
+                ntfs_attr_reinit_search_ctx(ctx);
+                if (ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                                CASE_SENSITIVE, bh_cpos, NULL, 0, ctx)) {
+                        ntfs_error(vol->sb, "Failed to find required "
+                                        "attribute extent of attribute in "
+                                        "error code path.  Run chkdsk to "
+                                        "recover.");
+                        write_lock_irqsave(&ni->size_lock, flags);
+                        ni->itype.compressed.size += vol->cluster_size;
+                        write_unlock_irqrestore(&ni->size_lock, flags);
+                        flush_dcache_mft_record_page(ctx->ntfs_ino);
+                        mark_mft_record_dirty(ctx->ntfs_ino);
+                        /*
+                         * The only thing that is now wrong is the compressed
+                         * size of the base attribute extent which chkdsk
+                         * should be able to fix.
+                         */
+                        NVolSetErrors(vol);
+                } else {
+                        m = ctx->mrec;
+                        a = ctx->attr;
+                        status.attr_switched = 0;
+                }
+        }
+        /*
+         * If the runlist has been modified, need to restore it by punching a
+         * hole into it and we then need to deallocate the on-disk cluster as
+         * well.  Note, we only modify the runlist if we are able to generate a
+         * new mapping pairs array, i.e. only when the mapped attribute extent
+         * is not switched.
+         */
+        if (status.runlist_merged && !status.attr_switched) {
+                BUG_ON(!rl_write_locked);
+                /* Make the file cluster we allocated sparse in the runlist. */
+                if (ntfs_rl_punch_nolock(vol, &ni->runlist, bh_cpos, 1)) {
+                        ntfs_error(vol->sb, "Failed to punch hole into "
+                                        "attribute runlist in error code "
+                                        "path.  Run chkdsk to recover the "
+                                        "lost cluster.");
+                        make_bad_inode(vi);
+                        make_bad_inode(VFS_I(base_ni));
+                        NVolSetErrors(vol);
+                } else /* if (success) */ {
+                        status.runlist_merged = 0;
+                        /*
+                         * Deallocate the on-disk cluster we allocated but only
+                         * if we succeeded in punching its vcn out of the
+                         * runlist.
+                         */
+                        down_write(&vol->lcnbmp_lock);
+                        if (ntfs_bitmap_clear_bit(vol->lcnbmp_ino, lcn)) {
+                                ntfs_error(vol->sb, "Failed to release "
+                                                "allocated cluster in error "
+                                                "code path.  Run chkdsk to "
+                                                "recover the lost cluster.");
+                                NVolSetErrors(vol);
+                        }
+                        up_write(&vol->lcnbmp_lock);
+                }
+        }
+        /*
+         * Resize the attribute record to its old size and rebuild the mapping
+         * pairs array.  Note, we only can do this if the runlist has been
+         * restored to its old state which also implies that the mapped
+         * attribute extent is not switched.
+         */
+        if (status.mp_rebuilt && !status.runlist_merged) {
+                if (ntfs_attr_record_resize(m, a, attr_rec_len)) {
+                        ntfs_error(vol->sb, "Failed to restore attribute "
+                                        "record in error code path.  Run "
+                                        "chkdsk to recover.");
+                        make_bad_inode(vi);
+                        make_bad_inode(VFS_I(base_ni));
+                        NVolSetErrors(vol);
+                } else /* if (success) */ {
+                        if (ntfs_mapping_pairs_build(vol, (u8*)a +
+                                        le16_to_cpu(a->data.non_resident.
+                                        mapping_pairs_offset), attr_rec_len -
+                                        le16_to_cpu(a->data.non_resident.
+                                        mapping_pairs_offset), ni->runlist.rl,
+                                        vcn, highest_vcn, NULL)) {
+                                ntfs_error(vol->sb, "Failed to restore "
+                                                "mapping pairs array in error "
+                                                "code path.  Run chkdsk to "
+                                                "recover.");
+                                make_bad_inode(vi);
+                                make_bad_inode(VFS_I(base_ni));
+                                NVolSetErrors(vol);
+                        }
+                        flush_dcache_mft_record_page(ctx->ntfs_ino);
+                        mark_mft_record_dirty(ctx->ntfs_ino);
+                }
+        }
+        /* Release the mft record and the attribute. */
+        if (status.mft_attr_mapped) {
+                ntfs_attr_put_search_ctx(ctx);
+                unmap_mft_record(base_ni);
+        }
+        /* Release the runlist lock. */
+        if (rl_write_locked)
+                up_write(&ni->runlist.lock);
+        else if (rl)
+                up_read(&ni->runlist.lock);
+        /*
+         * Zero out any newly allocated blocks to avoid exposing stale data.
+         * If BH_New is set, we know that the block was newly allocated above
+         * and that it has not been fully zeroed and marked dirty yet.
+         */
+        nr_pages = u;
+        u = 0;
+        end = bh_cpos << vol->cluster_size_bits;
+        do {
+                page = pages[u];
+                bh = head = page_buffers(page);
+                do {
+                        if (u == nr_pages &&
+                                        ((s64)page->index << PAGE_CACHE_SHIFT) +
+                                        bh_offset(bh) >= end)
+                                break;
+                        if (!buffer_new(bh))
+                                continue;
+                        clear_buffer_new(bh);
+                        if (!buffer_uptodate(bh)) {
+                                if (PageUptodate(page))
+                                        set_buffer_uptodate(bh);
+                                else {
+                                        u8 *kaddr = kmap_atomic(page, KM_USER0);
+                                        memset(kaddr + bh_offset(bh), 0,
+                                                        blocksize);
+                                        kunmap_atomic(kaddr, KM_USER0);
+                                        flush_dcache_page(page);
+                                        set_buffer_uptodate(bh);
+                                }
+                        }
+                        mark_buffer_dirty(bh);
+                } while ((bh = bh->b_this_page) != head);
+        } while (++u <= nr_pages);
+        ntfs_error(vol->sb, "Failed.  Returning error code %i.", err);
+        return err;
+}
+
+/*
+ * Copy as much as we can into the pages and return the number of bytes which
+ * were sucessfully copied.  If a fault is encountered then clear the pages
+ * out to (ofs + bytes) and return the number of bytes which were copied.
+ */
+static inline size_t ntfs_copy_from_user(struct page **pages,
+                unsigned nr_pages, unsigned ofs, const char __user *buf,
+                size_t bytes)
+{
+        struct page **last_page = pages + nr_pages;
+        char *kaddr;
+        size_t total = 0;
+        unsigned len;
+        int left;
+
+        do {
+                len = PAGE_CACHE_SIZE - ofs;
+                if (len > bytes)
+                        len = bytes;
+                kaddr = kmap_atomic(*pages, KM_USER0);
+                left = __copy_from_user_inatomic(kaddr + ofs, buf, len);
+                kunmap_atomic(kaddr, KM_USER0);
+                if (unlikely(left)) {
+                        /* Do it the slow way. */
+                        kaddr = kmap(*pages);
+                        left = __copy_from_user(kaddr + ofs, buf, len);
+                        kunmap(*pages);
+                        if (unlikely(left))
+                                goto err_out;
+                }
+                total += len;
+                bytes -= len;
+                if (!bytes)
+                        break;
+                buf += len;
+                ofs = 0;
+        } while (++pages < last_page);
+out:
+        return total;
+err_out:
+        total += len - left;
+        /* Zero the rest of the target like __copy_from_user(). */
+        while (++pages < last_page) {
+                bytes -= len;
+                if (!bytes)
+                        break;
+                len = PAGE_CACHE_SIZE;
+                if (len > bytes)
+                        len = bytes;
+                kaddr = kmap_atomic(*pages, KM_USER0);
+                memset(kaddr, 0, len);
+                kunmap_atomic(kaddr, KM_USER0);
+        }
+        goto out;
+}
+
+static size_t __ntfs_copy_from_user_iovec(char *vaddr,
+                const struct iovec *iov, size_t iov_ofs, size_t bytes)
+{
+        size_t total = 0;
+
+        while (1) {
+                const char __user *buf = iov->iov_base + iov_ofs;
+                unsigned len;
+                size_t left;
+
+                len = iov->iov_len - iov_ofs;
+                if (len > bytes)
+                        len = bytes;
+                left = __copy_from_user_inatomic(vaddr, buf, len);
+                total += len;
+                bytes -= len;
+                vaddr += len;
+                if (unlikely(left)) {
+                        /*
+                         * Zero the rest of the target like __copy_from_user().
+                         */
+                        memset(vaddr, 0, bytes);
+                        total -= left;
+                        break;
+                }
+                if (!bytes)
+                        break;
+                iov++;
+                iov_ofs = 0;
+        }
+        return total;
+}
+
+static inline void ntfs_set_next_iovec(const struct iovec **iovp,
+                size_t *iov_ofsp, size_t bytes)
+{
+        const struct iovec *iov = *iovp;
+        size_t iov_ofs = *iov_ofsp;
+
+        while (bytes) {
+                unsigned len;
+
+                len = iov->iov_len - iov_ofs;
+                if (len > bytes)
+                        len = bytes;
+                bytes -= len;
+                iov_ofs += len;
+                if (iov->iov_len == iov_ofs) {
+                        iov++;
+                        iov_ofs = 0;
+                }
+        }
+        *iovp = iov;
+        *iov_ofsp = iov_ofs;
+}
+
+/*
+ * This has the same side-effects and return value as ntfs_copy_from_user().
+ * The difference is that on a fault we need to memset the remainder of the
+ * pages (out to offset + bytes), to emulate ntfs_copy_from_user()'s
+ * single-segment behaviour.
+ *
+ * We call the same helper (__ntfs_copy_from_user_iovec()) both when atomic and
+ * when not atomic.  This is ok because __ntfs_copy_from_user_iovec() calls
+ * __copy_from_user_inatomic() and it is ok to call this when non-atomic.  In
+ * fact, the only difference between __copy_from_user_inatomic() and
+ * __copy_from_user() is that the latter calls might_sleep().  And on many
+ * architectures __copy_from_user_inatomic() is just defined to
+ * __copy_from_user() so it makes no difference at all on those architectures.
+ */
+static inline size_t ntfs_copy_from_user_iovec(struct page **pages,
+                unsigned nr_pages, unsigned ofs, const struct iovec **iov,
+                size_t *iov_ofs, size_t bytes)
+{
+        struct page **last_page = pages + nr_pages;
+        char *kaddr;
+        size_t copied, len, total = 0;
+
+        do {
+                len = PAGE_CACHE_SIZE - ofs;
+                if (len > bytes)
+                        len = bytes;
+                kaddr = kmap_atomic(*pages, KM_USER0);
+                copied = __ntfs_copy_from_user_iovec(kaddr + ofs,
+                                *iov, *iov_ofs, len);
+                kunmap_atomic(kaddr, KM_USER0);
+                if (unlikely(copied != len)) {
+                        /* Do it the slow way. */
+                        kaddr = kmap(*pages);
+                        copied = __ntfs_copy_from_user_iovec(kaddr + ofs,
+                                        *iov, *iov_ofs, len);
+                        kunmap(*pages);
+                        if (unlikely(copied != len))
+                                goto err_out;
+                }
+                total += len;
+                bytes -= len;
+                if (!bytes)
+                        break;
+                ntfs_set_next_iovec(iov, iov_ofs, len);
+                ofs = 0;
+        } while (++pages < last_page);
+out:
+        return total;
+err_out:
+        total += copied;
+        /* Zero the rest of the target like __copy_from_user(). */
+        while (++pages < last_page) {
+                bytes -= len;
+                if (!bytes)
+                        break;
+                len = PAGE_CACHE_SIZE;
+                if (len > bytes)
+                        len = bytes;
+                kaddr = kmap_atomic(*pages, KM_USER0);
+                memset(kaddr, 0, len);
+                kunmap_atomic(kaddr, KM_USER0);
+        }
+        goto out;
+}
+
+static inline void ntfs_flush_dcache_pages(struct page **pages,
+                unsigned nr_pages)
+{
+        BUG_ON(!nr_pages);
+        do {
+                /*
+                 * Warning: Do not do the decrement at the same time as the
+                 * call because flush_dcache_page() is a NULL macro on i386
+                 * and hence the decrement never happens.
+                 */
+                flush_dcache_page(pages[nr_pages]);
+        } while (--nr_pages > 0);
+}
+
+/**
+ * ntfs_commit_pages_after_non_resident_write - commit the received data
+ * @pages:      array of destination pages
+ * @nr_pages:   number of pages in @pages
+ * @pos:        byte position in file at which the write begins
+ * @bytes:      number of bytes to be written
+ *
+ * See description of ntfs_commit_pages_after_write(), below.
+ */
+static inline int ntfs_commit_pages_after_non_resident_write(
+                struct page **pages, const unsigned nr_pages,
+                s64 pos, size_t bytes)
+{
+        s64 end, initialized_size;
+        struct inode *vi;
+        ntfs_inode *ni, *base_ni;
+        struct buffer_head *bh, *head;
+        ntfs_attr_search_ctx *ctx;
+        MFT_RECORD *m;
+        ATTR_RECORD *a;
+        unsigned long flags;
+        unsigned blocksize, u;
+        int err;
+
+        vi = pages[0]->mapping->host;
+        ni = NTFS_I(vi);
+        blocksize = 1 << vi->i_blkbits;
+        end = pos + bytes;
+        u = 0;
+        do {
+                s64 bh_pos;
+                struct page *page;
+                BOOL partial;
+
+                page = pages[u];
+                bh_pos = (s64)page->index << PAGE_CACHE_SHIFT;
+                bh = head = page_buffers(page);
+                partial = FALSE;
+                do {
+                        s64 bh_end;
+
+                        bh_end = bh_pos + blocksize;
+                        if (bh_end <= pos || bh_pos >= end) {
+                                if (!buffer_uptodate(bh))
+                                        partial = TRUE;
+                        } else {
+                                set_buffer_uptodate(bh);
+                                mark_buffer_dirty(bh);
+                        }
+                } while (bh_pos += blocksize, (bh = bh->b_this_page) != head);
+                /*
+                 * If all buffers are now uptodate but the page is not, set the
+                 * page uptodate.
+                 */
+                if (!partial && !PageUptodate(page))
+                        SetPageUptodate(page);
+        } while (++u < nr_pages);
+        /*
+         * Finally, if we do not need to update initialized_size or i_size we
+         * are finished.
+         */
+        read_lock_irqsave(&ni->size_lock, flags);
+        initialized_size = ni->initialized_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        if (end <= initialized_size) {
+                ntfs_debug("Done.");
+                return 0;
+        }
+        /*
+         * Update initialized_size/i_size as appropriate, both in the inode and
+         * the mft record.
+         */
+        if (!NInoAttr(ni))
+                base_ni = ni;
+        else
+                base_ni = ni->ext.base_ntfs_ino;
+        /* Map, pin, and lock the mft record. */
+        m = map_mft_record(base_ni);
+        if (IS_ERR(m)) {
+                err = PTR_ERR(m);
+                m = NULL;
+                ctx = NULL;
+                goto err_out;
+        }
+        BUG_ON(!NInoNonResident(ni));
+        ctx = ntfs_attr_get_search_ctx(base_ni, m);
+        if (unlikely(!ctx)) {
+                err = -ENOMEM;
+                goto err_out;
+        }
+        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                        CASE_SENSITIVE, 0, NULL, 0, ctx);
+        if (unlikely(err)) {
+                if (err == -ENOENT)
+                        err = -EIO;
+                goto err_out;
+        }
+        a = ctx->attr;
+        BUG_ON(!a->non_resident);
+        write_lock_irqsave(&ni->size_lock, flags);
+        BUG_ON(end > ni->allocated_size);
+        ni->initialized_size = end;
+        a->data.non_resident.initialized_size = cpu_to_sle64(end);
+        if (end > i_size_read(vi)) {
+                i_size_write(vi, end);
+                a->data.non_resident.data_size =
+                                a->data.non_resident.initialized_size;
+        }
+        write_unlock_irqrestore(&ni->size_lock, flags);
+        /* Mark the mft record dirty, so it gets written back. */
+        flush_dcache_mft_record_page(ctx->ntfs_ino);
+        mark_mft_record_dirty(ctx->ntfs_ino);
+        ntfs_attr_put_search_ctx(ctx);
+        unmap_mft_record(base_ni);
+        ntfs_debug("Done.");
+        return 0;
+err_out:
+        if (ctx)
+                ntfs_attr_put_search_ctx(ctx);
+        if (m)
+                unmap_mft_record(base_ni);
+        ntfs_error(vi->i_sb, "Failed to update initialized_size/i_size (error "
+                        "code %i).", err);
+        if (err != -ENOMEM) {
+                NVolSetErrors(ni->vol);
+                make_bad_inode(VFS_I(base_ni));
+                make_bad_inode(vi);
+        }
+        return err;
+}
+
+/**
+ * ntfs_commit_pages_after_write - commit the received data
+ * @pages:      array of destination pages
+ * @nr_pages:   number of pages in @pages
+ * @pos:        byte position in file at which the write begins
+ * @bytes:      number of bytes to be written
+ *
+ * This is called from ntfs_file_buffered_write() with i_sem held on the inode
+ * (@pages[0]->mapping->host).  There are @nr_pages pages in @pages which are
+ * locked but not kmap()ped.  The source data has already been copied into the
+ * @page.  ntfs_prepare_pages_for_non_resident_write() has been called before
+ * the data was copied (for non-resident attributes only) and it returned
+ * success.
+ *
+ * Need to set uptodate and mark dirty all buffers within the boundary of the
+ * write.  If all buffers in a page are uptodate we set the page uptodate, too.
+ *
+ * Setting the buffers dirty ensures that they get written out later when
+ * ntfs_writepage() is invoked by the VM.
+ *
+ * Finally, we need to update i_size and initialized_size as appropriate both
+ * in the inode and the mft record.
+ *
+ * This is modelled after fs/buffer.c::generic_commit_write(), which marks
+ * buffers uptodate and dirty, sets the page uptodate if all buffers in the
+ * page are uptodate, and updates i_size if the end of io is beyond i_size.  In
+ * that case, it also marks the inode dirty.
+ *
+ * If things have gone as outlined in
+ * ntfs_prepare_pages_for_non_resident_write(), we do not need to do any page
+ * content modifications here for non-resident attributes.  For resident
+ * attributes we need to do the uptodate bringing here which we combine with
+ * the copying into the mft record which means we save one atomic kmap.
+ *
+ * Return 0 on success or -errno on error.
+ */
+static int ntfs_commit_pages_after_write(struct page **pages,
+                const unsigned nr_pages, s64 pos, size_t bytes)
+{
+        s64 end, initialized_size;
+        loff_t i_size;
+        struct inode *vi;
+        ntfs_inode *ni, *base_ni;
+        struct page *page;
+        ntfs_attr_search_ctx *ctx;
+        MFT_RECORD *m;
+        ATTR_RECORD *a;
+        char *kattr, *kaddr;
+        unsigned long flags;
+        u32 attr_len;
+        int err;
+
+        BUG_ON(!nr_pages);
+        BUG_ON(!pages);
+        page = pages[0];
+        BUG_ON(!page);
+        vi = page->mapping->host;
+        ni = NTFS_I(vi);
+        ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, start page "
+                        "index 0x%lx, nr_pages 0x%x, pos 0x%llx, bytes 0x%zx.",
+                        vi->i_ino, ni->type, page->index, nr_pages,
+                        (long long)pos, bytes);
+        if (NInoNonResident(ni))
+                return ntfs_commit_pages_after_non_resident_write(pages,
+                                nr_pages, pos, bytes);
+        BUG_ON(nr_pages > 1);
+        /*
+         * Attribute is resident, implying it is not compressed, encrypted, or
+         * sparse.
+         */
+        if (!NInoAttr(ni))
+                base_ni = ni;
+        else
+                base_ni = ni->ext.base_ntfs_ino;
+        BUG_ON(NInoNonResident(ni));
+        /* Map, pin, and lock the mft record. */
+        m = map_mft_record(base_ni);
+        if (IS_ERR(m)) {
+                err = PTR_ERR(m);
+                m = NULL;
+                ctx = NULL;
+                goto err_out;
+        }
+        ctx = ntfs_attr_get_search_ctx(base_ni, m);
+        if (unlikely(!ctx)) {
+                err = -ENOMEM;
+                goto err_out;
+        }
+        err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
+                        CASE_SENSITIVE, 0, NULL, 0, ctx);
+        if (unlikely(err)) {
+                if (err == -ENOENT)
+                        err = -EIO;
+                goto err_out;
+        }
+        a = ctx->attr;
+        BUG_ON(a->non_resident);
+        /* The total length of the attribute value. */
+        attr_len = le32_to_cpu(a->data.resident.value_length);
+        i_size = i_size_read(vi);
+        BUG_ON(attr_len != i_size);
+        BUG_ON(pos > attr_len);
+        end = pos + bytes;
+        BUG_ON(end > le32_to_cpu(a->length) -
+                        le16_to_cpu(a->data.resident.value_offset));
+        kattr = (u8*)a + le16_to_cpu(a->data.resident.value_offset);
+        kaddr = kmap_atomic(page, KM_USER0);
+        /* Copy the received data from the page to the mft record. */
+        memcpy(kattr + pos, kaddr + pos, bytes);
+        /* Update the attribute length if necessary. */
+        if (end > attr_len) {
+                attr_len = end;
+                a->data.resident.value_length = cpu_to_le32(attr_len);
+        }
+        /*
+         * If the page is not uptodate, bring the out of bounds area(s)
+         * uptodate by copying data from the mft record to the page.
+         */
+        if (!PageUptodate(page)) {
+                if (pos > 0)
+                        memcpy(kaddr, kattr, pos);
+                if (end < attr_len)
+                        memcpy(kaddr + end, kattr + end, attr_len - end);
+                /* Zero the region outside the end of the attribute value. */
+                memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len);
+                flush_dcache_page(page);
+                SetPageUptodate(page);
+        }
+        kunmap_atomic(kaddr, KM_USER0);
+        /* Update initialized_size/i_size if necessary. */
+        read_lock_irqsave(&ni->size_lock, flags);
+        initialized_size = ni->initialized_size;
+        BUG_ON(end > ni->allocated_size);
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        BUG_ON(initialized_size != i_size);
+        if (end > initialized_size) {
+                unsigned long flags;
+
+                write_lock_irqsave(&ni->size_lock, flags);
+                ni->initialized_size = end;
+                i_size_write(vi, end);
+                write_unlock_irqrestore(&ni->size_lock, flags);
+        }
+        /* Mark the mft record dirty, so it gets written back. */
+        flush_dcache_mft_record_page(ctx->ntfs_ino);
+        mark_mft_record_dirty(ctx->ntfs_ino);
+        ntfs_attr_put_search_ctx(ctx);
+        unmap_mft_record(base_ni);
+        ntfs_debug("Done.");
+        return 0;
+err_out:
+        if (err == -ENOMEM) {
+                ntfs_warning(vi->i_sb, "Error allocating memory required to "
+                                "commit the write.");
+                if (PageUptodate(page)) {
+                        ntfs_warning(vi->i_sb, "Page is uptodate, setting "
+                                        "dirty so the write will be retried "
+                                        "later on by the VM.");
+                        /*
+                         * Put the page on mapping->dirty_pages, but leave its
+                         * buffers' dirty state as-is.
+                         */
+                        __set_page_dirty_nobuffers(page);
+                        err = 0;
+                } else
+                        ntfs_error(vi->i_sb, "Page is not uptodate.  Written "
+                                        "data has been lost.");
+        } else {
+                ntfs_error(vi->i_sb, "Resident attribute commit write failed "
+                                "with error %i.", err);
+                NVolSetErrors(ni->vol);
+                make_bad_inode(VFS_I(base_ni));
+                make_bad_inode(vi);
+        }
+        if (ctx)
+                ntfs_attr_put_search_ctx(ctx);
+        if (m)
+                unmap_mft_record(base_ni);
+        return err;
+}
+
+/**
+ * ntfs_file_buffered_write -
+ *
+ * Locking: The vfs is holding ->i_sem on the inode.
+ */
+static ssize_t ntfs_file_buffered_write(struct kiocb *iocb,
+                const struct iovec *iov, unsigned long nr_segs,
+                loff_t pos, loff_t *ppos, size_t count)
+{
+        struct file *file = iocb->ki_filp;
+        struct address_space *mapping = file->f_mapping;
+        struct inode *vi = mapping->host;
+        ntfs_inode *ni = NTFS_I(vi);
+        ntfs_volume *vol = ni->vol;
+        struct page *pages[NTFS_MAX_PAGES_PER_CLUSTER];
+        struct page *cached_page = NULL;
+        char __user *buf = NULL;
+        s64 end, ll;
+        VCN last_vcn;
+        LCN lcn;
+        unsigned long flags;
+        size_t bytes, iov_ofs = 0;      /* Offset in the current iovec. */
+        ssize_t status, written;
+        unsigned nr_pages;
+        int err;
+        struct pagevec lru_pvec;
+
+        ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, "
+                        "pos 0x%llx, count 0x%lx.",
+                        vi->i_ino, (unsigned)le32_to_cpu(ni->type),
+                        (unsigned long long)pos, (unsigned long)count);
+        if (unlikely(!count))
+                return 0;
+        BUG_ON(NInoMstProtected(ni));
+        /*
+         * If the attribute is not an index root and it is encrypted or
+         * compressed, we cannot write to it yet.  Note we need to check for
+         * AT_INDEX_ALLOCATION since this is the type of both directory and
+         * index inodes.
+         */
+        if (ni->type != AT_INDEX_ALLOCATION) {
+                /* If file is encrypted, deny access, just like NT4. */
+                if (NInoEncrypted(ni)) {
+                        /*
+                         * Reminder for later: Encrypted files are _always_
+                         * non-resident so that the content can always be
+                         * encrypted.
+                         */
+                        ntfs_debug("Denying write access to encrypted file.");
+                        return -EACCES;
+                }
+                if (NInoCompressed(ni)) {
+                        /* Only unnamed $DATA attribute can be compressed. */
+                        BUG_ON(ni->type != AT_DATA);
+                        BUG_ON(ni->name_len);
+                        /*
+                         * Reminder for later: If resident, the data is not
+                         * actually compressed.  Only on the switch to non-
+                         * resident does compression kick in.  This is in
+                         * contrast to encrypted files (see above).
+                         */
+                        ntfs_error(vi->i_sb, "Writing to compressed files is "
+                                        "not implemented yet.  Sorry.");
+                        return -EOPNOTSUPP;
+                }
+        }
+        /*
+         * If a previous ntfs_truncate() failed, repeat it and abort if it
+         * fails again.
+         */
+        if (unlikely(NInoTruncateFailed(ni))) {
+                down_write(&vi->i_alloc_sem);
+                err = ntfs_truncate(vi);
+                up_write(&vi->i_alloc_sem);
+                if (err || NInoTruncateFailed(ni)) {
+                        if (!err)
+                                err = -EIO;
+                        ntfs_error(vol->sb, "Cannot perform write to inode "
+                                        "0x%lx, attribute type 0x%x, because "
+                                        "ntfs_truncate() failed (error code "
+                                        "%i).", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                        return err;
+                }
+        }
+        /* The first byte after the write. */
+        end = pos + count;
+        /*
+         * If the write goes beyond the allocated size, extend the allocation
+         * to cover the whole of the write, rounded up to the nearest cluster.
+         */
+        read_lock_irqsave(&ni->size_lock, flags);
+        ll = ni->allocated_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        if (end > ll) {
+                /* Extend the allocation without changing the data size. */
+                ll = ntfs_attr_extend_allocation(ni, end, -1, pos);
+                if (likely(ll >= 0)) {
+                        BUG_ON(pos >= ll);
+                        /* If the extension was partial truncate the write. */
+                        if (end > ll) {
+                                ntfs_debug("Truncating write to inode 0x%lx, "
+                                                "attribute type 0x%x, because "
+                                                "the allocation was only "
+                                                "partially extended.",
+                                                vi->i_ino, (unsigned)
+                                                le32_to_cpu(ni->type));
+                                end = ll;
+                                count = ll - pos;
+                        }
+                } else {
+                        err = ll;
+                        read_lock_irqsave(&ni->size_lock, flags);
+                        ll = ni->allocated_size;
+                        read_unlock_irqrestore(&ni->size_lock, flags);
+                        /* Perform a partial write if possible or fail. */
+                        if (pos < ll) {
+                                ntfs_debug("Truncating write to inode 0x%lx, "
+                                                "attribute type 0x%x, because "
+                                                "extending the allocation "
+                                                "failed (error code %i).",
+                                                vi->i_ino, (unsigned)
+                                                le32_to_cpu(ni->type), err);
+                                end = ll;
+                                count = ll - pos;
+                        } else {
+                                ntfs_error(vol->sb, "Cannot perform write to "
+                                                "inode 0x%lx, attribute type "
+                                                "0x%x, because extending the "
+                                                "allocation failed (error "
+                                                "code %i).", vi->i_ino,
+                                                (unsigned)
+                                                le32_to_cpu(ni->type), err);
+                                return err;
+                        }
+                }
+        }
+        pagevec_init(&lru_pvec, 0);
+        written = 0;
+        /*
+         * If the write starts beyond the initialized size, extend it up to the
+         * beginning of the write and initialize all non-sparse space between
+         * the old initialized size and the new one.  This automatically also
+         * increments the vfs inode->i_size to keep it above or equal to the
+         * initialized_size.
+         */
+        read_lock_irqsave(&ni->size_lock, flags);
+        ll = ni->initialized_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        if (pos > ll) {
+                err = ntfs_attr_extend_initialized(ni, pos, &cached_page,
+                                &lru_pvec);
+                if (err < 0) {
+                        ntfs_error(vol->sb, "Cannot perform write to inode "
+                                        "0x%lx, attribute type 0x%x, because "
+                                        "extending the initialized size "
+                                        "failed (error code %i).", vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type), err);
+                        status = err;
+                        goto err_out;
+                }
+        }
+        /*
+         * Determine the number of pages per cluster for non-resident
+         * attributes.
+         */
+        nr_pages = 1;
+        if (vol->cluster_size > PAGE_CACHE_SIZE && NInoNonResident(ni))
+                nr_pages = vol->cluster_size >> PAGE_CACHE_SHIFT;
+        /* Finally, perform the actual write. */
+        last_vcn = -1;
+        if (likely(nr_segs == 1))
+                buf = iov->iov_base;
+        do {
+                VCN vcn;
+                pgoff_t idx, start_idx;
+                unsigned ofs, do_pages, u;
+                size_t copied;
+
+                start_idx = idx = pos >> PAGE_CACHE_SHIFT;
+                ofs = pos & ~PAGE_CACHE_MASK;
+                bytes = PAGE_CACHE_SIZE - ofs;
+                do_pages = 1;
+                if (nr_pages > 1) {
+                        vcn = pos >> vol->cluster_size_bits;
+                        if (vcn != last_vcn) {
+                                last_vcn = vcn;
+                                /*
+                                 * Get the lcn of the vcn the write is in.  If
+                                 * it is a hole, need to lock down all pages in
+                                 * the cluster.
+                                 */
+                                down_read(&ni->runlist.lock);
+                                lcn = ntfs_attr_vcn_to_lcn_nolock(ni, pos >>
+                                                vol->cluster_size_bits, FALSE);
+                                up_read(&ni->runlist.lock);
+                                if (unlikely(lcn < LCN_HOLE)) {
+                                        status = -EIO;
+                                        if (lcn == LCN_ENOMEM)
+                                                status = -ENOMEM;
+                                        else
+                                                ntfs_error(vol->sb, "Cannot "
+                                                        "perform write to "
+                                                        "inode 0x%lx, "
+                                                        "attribute type 0x%x, "
+                                                        "because the attribute "
+                                                        "is corrupt.",
+                                                        vi->i_ino, (unsigned)
+                                                        le32_to_cpu(ni->type));
+                                        break;
+                                }
+                                if (lcn == LCN_HOLE) {
+                                        start_idx = (pos & ~(s64)
+                                                        vol->cluster_size_mask)
+                                                        >> PAGE_CACHE_SHIFT;
+                                        bytes = vol->cluster_size - (pos &
+                                                        vol->cluster_size_mask);
+                                        do_pages = nr_pages;
+                                }
+                        }
+                }
+                if (bytes > count)
+                        bytes = count;
+                /*
+                 * Bring in the user page(s) that we will copy from _first_.
+                 * Otherwise there is a nasty deadlock on copying from the same
+                 * page(s) as we are writing to, without it/them being marked
+                 * up-to-date.  Note, at present there is nothing to stop the
+                 * pages being swapped out between us bringing them into memory
+                 * and doing the actual copying.
+                 */
+                if (likely(nr_segs == 1))
+                        ntfs_fault_in_pages_readable(buf, bytes);
+                else
+                        ntfs_fault_in_pages_readable_iovec(iov, iov_ofs, bytes);
+                /* Get and lock @do_pages starting at index @start_idx. */
+                status = __ntfs_grab_cache_pages(mapping, start_idx, do_pages,
+                                pages, &cached_page, &lru_pvec);
+                if (unlikely(status))
+                        break;
+                /*
+                 * For non-resident attributes, we need to fill any holes with
+                 * actual clusters and ensure all bufferes are mapped.  We also
+                 * need to bring uptodate any buffers that are only partially
+                 * being written to.
+                 */
+                if (NInoNonResident(ni)) {
+                        status = ntfs_prepare_pages_for_non_resident_write(
+                                        pages, do_pages, pos, bytes);
+                        if (unlikely(status)) {
+                                loff_t i_size;
+
+                                do {
+                                        unlock_page(pages[--do_pages]);
+                                        page_cache_release(pages[do_pages]);
+                                } while (do_pages);
+                                /*
+                                 * The write preparation may have instantiated
+                                 * allocated space outside i_size.  Trim this
+                                 * off again.  We can ignore any errors in this
+                                 * case as we will just be waisting a bit of
+                                 * allocated space, which is not a disaster.
+                                 */
+                                i_size = i_size_read(vi);
+                                if (pos + bytes > i_size)
+                                        vmtruncate(vi, i_size);
+                                break;
+                        }
+                }
+                u = (pos >> PAGE_CACHE_SHIFT) - pages[0]->index;
+                if (likely(nr_segs == 1)) {
+                        copied = ntfs_copy_from_user(pages + u, do_pages - u,
+                                        ofs, buf, bytes);
+                        buf += copied;
+                } else
+                        copied = ntfs_copy_from_user_iovec(pages + u,
+                                        do_pages - u, ofs, &iov, &iov_ofs,
+                                        bytes);
+                ntfs_flush_dcache_pages(pages + u, do_pages - u);
+                status = ntfs_commit_pages_after_write(pages, do_pages, pos,
+                                bytes);
+                if (likely(!status)) {
+                        written += copied;
+                        count -= copied;
+                        pos += copied;
+                        if (unlikely(copied != bytes))
+                                status = -EFAULT;
+                }
+                do {
+                        unlock_page(pages[--do_pages]);
+                        mark_page_accessed(pages[do_pages]);
+                        page_cache_release(pages[do_pages]);
+                } while (do_pages);
+                if (unlikely(status))
+                        break;
+                balance_dirty_pages_ratelimited(mapping);
+                cond_resched();
+        } while (count);
+err_out:
+        *ppos = pos;
+        if (cached_page)
+                page_cache_release(cached_page);
+        /* For now, when the user asks for O_SYNC, we actually give O_DSYNC. */
+        if (likely(!status)) {
+                if (unlikely((file->f_flags & O_SYNC) || IS_SYNC(vi))) {
+                        if (!mapping->a_ops->writepage || !is_sync_kiocb(iocb))
+                                status = generic_osync_inode(vi, mapping,
+                                                OSYNC_METADATA|OSYNC_DATA);
+                }
+        }
+        pagevec_lru_add(&lru_pvec);
+        ntfs_debug("Done.  Returning %s (written 0x%lx, status %li).",
+                        written ? "written" : "status", (unsigned long)written,
+                        (long)status);
+        return written ? written : status;
+}
+
+/**
+ * ntfs_file_aio_write_nolock -
+ */
+static ssize_t ntfs_file_aio_write_nolock(struct kiocb *iocb,
+                const struct iovec *iov, unsigned long nr_segs, loff_t *ppos)
+{
+        struct file *file = iocb->ki_filp;
+        struct address_space *mapping = file->f_mapping;
+        struct inode *inode = mapping->host;
+        loff_t pos;
+        unsigned long seg;
+        size_t count;           /* after file limit checks */
+        ssize_t written, err;
+
+        count = 0;
+        for (seg = 0; seg < nr_segs; seg++) {
+                const struct iovec *iv = &iov[seg];
+                /*
+                 * If any segment has a negative length, or the cumulative
+                 * length ever wraps negative then return -EINVAL.
+                 */
+                count += iv->iov_len;
+                if (unlikely((ssize_t)(count|iv->iov_len) < 0))
+                        return -EINVAL;
+                if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len))
+                        continue;
+                if (!seg)
+                        return -EFAULT;
+                nr_segs = seg;
+                count -= iv->iov_len;   /* This segment is no good */
+                break;
+        }
+        pos = *ppos;
+        vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
+        /* We can write back this queue in page reclaim. */
+        current->backing_dev_info = mapping->backing_dev_info;
+        written = 0;
+        err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
+        if (err)
+                goto out;
+        if (!count)
+                goto out;
+        err = remove_suid(file->f_dentry);
+        if (err)
+                goto out;
+        inode_update_time(inode, 1);
+        written = ntfs_file_buffered_write(iocb, iov, nr_segs, pos, ppos,
+                        count);
+out:
+        current->backing_dev_info = NULL;
+        return written ? written : err;
+}
+
+/**
+ * ntfs_file_aio_write -
+ */
+static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const char __user *buf,
+                size_t count, loff_t pos)
+{
+        struct file *file = iocb->ki_filp;
+        struct address_space *mapping = file->f_mapping;
+        struct inode *inode = mapping->host;
+        ssize_t ret;
+        struct iovec local_iov = { .iov_base = (void __user *)buf,
+                                   .iov_len = count };
+
+        BUG_ON(iocb->ki_pos != pos);
+
+        down(&inode->i_sem);
+        ret = ntfs_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
+        up(&inode->i_sem);
+        if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
+                int err = sync_page_range(inode, mapping, pos, ret);
+                if (err < 0)
+                        ret = err;
+        }
+        return ret;
+}
+
+/**
+ * ntfs_file_writev -
+ *
+ * Basically the same as generic_file_writev() except that it ends up calling
+ * ntfs_file_aio_write_nolock() instead of __generic_file_aio_write_nolock().
+ */
+static ssize_t ntfs_file_writev(struct file *file, const struct iovec *iov,
+                unsigned long nr_segs, loff_t *ppos)
+{
+        struct address_space *mapping = file->f_mapping;
+        struct inode *inode = mapping->host;
+        struct kiocb kiocb;
+        ssize_t ret;
+
+        down(&inode->i_sem);
+        init_sync_kiocb(&kiocb, file);
+        ret = ntfs_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos);
+        if (ret == -EIOCBQUEUED)
+                ret = wait_on_sync_kiocb(&kiocb);
+        up(&inode->i_sem);
+        if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
+                int err = sync_page_range(inode, mapping, *ppos - ret, ret);
+                if (err < 0)
+                        ret = err;
+        }
+        return ret;
+}
+
+/**
+ * ntfs_file_write - simple wrapper for ntfs_file_writev()
+ */
+static ssize_t ntfs_file_write(struct file *file, const char __user *buf,
+                size_t count, loff_t *ppos)
+{
+        struct iovec local_iov = { .iov_base = (void __user *)buf,
+                                   .iov_len = count };
+
+        return ntfs_file_writev(file, &local_iov, 1, ppos);
+}
+
+/**
  * ntfs_file_fsync - sync a file to disk
  * @filp:       file to be synced
  * @dentry:     dentry describing the file to sync
@@ -113,39 +2304,39 @@ static int ntfs_file_fsync(struct file *filp, struct dentry *dentry,
 #endif /* NTFS_RW */
 
 struct file_operations ntfs_file_ops = {
-        .llseek         = generic_file_llseek,    /* Seek inside file. */
-        .read           = generic_file_read,      /* Read from file. */
-        .aio_read       = generic_file_aio_read,  /* Async read from file. */
-        .readv          = generic_file_readv,     /* Read from file. */
+        .llseek         = generic_file_llseek,   /* Seek inside file. */
+        .read           = generic_file_read,     /* Read from file. */
+        .aio_read       = generic_file_aio_read, /* Async read from file. */
+        .readv          = generic_file_readv,    /* Read from file. */
 #ifdef NTFS_RW
-        .write          = generic_file_write,     /* Write to file. */
-        .aio_write      = generic_file_aio_write, /* Async write to file. */
-        .writev         = generic_file_writev,    /* Write to file. */
-        /*.release      = ,*/                     /* Last file is closed.  See
-                                                     fs/ext2/file.c::
-                                                     ext2_release_file() for
-                                                     how to use this to discard
-                                                     preallocated space for
-                                                     write opened files. */
-        .fsync          = ntfs_file_fsync,        /* Sync a file to disk. */
-        /*.aio_fsync    = ,*/                     /* Sync all outstanding async
-                                                     i/o operations on a
-                                                     kiocb. */
+        .write          = ntfs_file_write,       /* Write to file. */
+        .aio_write      = ntfs_file_aio_write,   /* Async write to file. */
+        .writev         = ntfs_file_writev,      /* Write to file. */
+        /*.release      = ,*/                    /* Last file is closed.  See
+                                                    fs/ext2/file.c::
+                                                    ext2_release_file() for
+                                                    how to use this to discard
+                                                    preallocated space for
+                                                    write opened files. */
+        .fsync          = ntfs_file_fsync,       /* Sync a file to disk. */
+        /*.aio_fsync    = ,*/                    /* Sync all outstanding async
+                                                    i/o operations on a
+                                                    kiocb. */
 #endif /* NTFS_RW */
-        /*.ioctl        = ,*/                     /* Perform function on the
-                                                     mounted filesystem. */
-        .mmap           = generic_file_mmap,      /* Mmap file. */
-        .open           = ntfs_file_open,         /* Open file. */
-        .sendfile       = generic_file_sendfile,  /* Zero-copy data send with
-                                                     the data source being on
-                                                     the ntfs partition.  We
-                                                     do not need to care about
-                                                     the data destination. */
-        /*.sendpage     = ,*/                     /* Zero-copy data send with
-                                                     the data destination being
-                                                     on the ntfs partition.  We
-                                                     do not need to care about
-                                                     the data source. */
+        /*.ioctl        = ,*/                    /* Perform function on the
+                                                    mounted filesystem. */
+        .mmap           = generic_file_mmap,     /* Mmap file. */
+        .open           = ntfs_file_open,        /* Open file. */
+        .sendfile       = generic_file_sendfile, /* Zero-copy data send with
+                                                    the data source being on
+                                                    the ntfs partition.  We do
+                                                    not need to care about the
+                                                    data destination. */
+        /*.sendpage     = ,*/                    /* Zero-copy data send with
+                                                    the data destination being
+                                                    on the ntfs partition.  We
+                                                    do not need to care about
+                                                    the data source. */
 };
 
 struct inode_operations ntfs_file_inode_ops = {
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 7ec045131808..b24f4c4b2c5c 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -30,6 +30,7 @@
 #include "debug.h"
 #include "inode.h"
 #include "attrib.h"
+#include "lcnalloc.h"
 #include "malloc.h"
 #include "mft.h"
 #include "time.h"
@@ -2291,11 +2292,16 @@ int ntfs_show_options(struct seq_file *sf, struct vfsmount *mnt)
 
 #ifdef NTFS_RW
 
+static const char *es = "  Leaving inconsistent metadata.  Unmount and run "
+                "chkdsk.";
+
 /**
  * ntfs_truncate - called when the i_size of an ntfs inode is changed
  * @vi:         inode for which the i_size was changed
  *
- * We do not support i_size changes yet.
+ * We only support i_size changes for normal files at present, i.e. not
+ * compressed and not encrypted.  This is enforced in ntfs_setattr(), see
+ * below.
  *
  * The kernel guarantees that @vi is a regular file (S_ISREG() is true) and
  * that the change is allowed.
@@ -2306,80 +2312,499 @@ int ntfs_show_options(struct seq_file *sf, struct vfsmount *mnt)
  * Returns 0 on success or -errno on error.
  *
  * Called with ->i_sem held.  In all but one case ->i_alloc_sem is held for
- * writing.  The only case where ->i_alloc_sem is not held is
+ * writing.  The only case in the kernel where ->i_alloc_sem is not held is
  * mm/filemap.c::generic_file_buffered_write() where vmtruncate() is called
- * with the current i_size as the offset which means that it is a noop as far
- * as ntfs_truncate() is concerned.
+ * with the current i_size as the offset.  The analogous place in NTFS is in
+ * fs/ntfs/file.c::ntfs_file_buffered_write() where we call vmtruncate() again
+ * without holding ->i_alloc_sem.
  */
 int ntfs_truncate(struct inode *vi)
 {
-        ntfs_inode *ni = NTFS_I(vi);
+        s64 new_size, old_size, nr_freed, new_alloc_size, old_alloc_size;
+        VCN highest_vcn;
+        unsigned long flags;
+        ntfs_inode *base_ni, *ni = NTFS_I(vi);
         ntfs_volume *vol = ni->vol;
         ntfs_attr_search_ctx *ctx;
         MFT_RECORD *m;
         ATTR_RECORD *a;
         const char *te = "  Leaving file length out of sync with i_size.";
-        int err;
+        int err, mp_size, size_change, alloc_change;
+        u32 attr_len;
 
         ntfs_debug("Entering for inode 0x%lx.", vi->i_ino);
         BUG_ON(NInoAttr(ni));
+        BUG_ON(S_ISDIR(vi->i_mode));
+        BUG_ON(NInoMstProtected(ni));
         BUG_ON(ni->nr_extents < 0);
-        m = map_mft_record(ni);
+retry_truncate:
+        /*
+         * Lock the runlist for writing and map the mft record to ensure it is
+         * safe to mess with the attribute runlist and sizes.
+         */
+        down_write(&ni->runlist.lock);
+        if (!NInoAttr(ni))
+                base_ni = ni;
+        else
+                base_ni = ni->ext.base_ntfs_ino;
+        m = map_mft_record(base_ni);
         if (IS_ERR(m)) {
                 err = PTR_ERR(m);
                 ntfs_error(vi->i_sb, "Failed to map mft record for inode 0x%lx "
                                 "(error code %d).%s", vi->i_ino, err, te);
                 ctx = NULL;
                 m = NULL;
-                goto err_out;
+                goto old_bad_out;
         }
-        ctx = ntfs_attr_get_search_ctx(ni, m);
+        ctx = ntfs_attr_get_search_ctx(base_ni, m);
         if (unlikely(!ctx)) {
                 ntfs_error(vi->i_sb, "Failed to allocate a search context for "
                                 "inode 0x%lx (not enough memory).%s",
                                 vi->i_ino, te);
                 err = -ENOMEM;
-                goto err_out;
+                goto old_bad_out;
         }
         err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len,
                         CASE_SENSITIVE, 0, NULL, 0, ctx);
         if (unlikely(err)) {
-                if (err == -ENOENT)
+                if (err == -ENOENT) {
                         ntfs_error(vi->i_sb, "Open attribute is missing from "
                                         "mft record.  Inode 0x%lx is corrupt.  "
-                                        "Run chkdsk.", vi->i_ino);
-                else
+                                        "Run chkdsk.%s", vi->i_ino, te);
+                        err = -EIO;
+                } else
                         ntfs_error(vi->i_sb, "Failed to lookup attribute in "
-                                        "inode 0x%lx (error code %d).",
-                                        vi->i_ino, err);
-                goto err_out;
+                                        "inode 0x%lx (error code %d).%s",
+                                        vi->i_ino, err, te);
+                goto old_bad_out;
         }
+        m = ctx->mrec;
         a = ctx->attr;
-        /* If the size has not changed there is nothing to do. */
-        if (ntfs_attr_size(a) == i_size_read(vi))
-                goto done;
-        // TODO: Implement the truncate...
-        ntfs_error(vi->i_sb, "Inode size has changed but this is not "
-                        "implemented yet.  Resetting inode size to old value. "
-                        " This is most likely a bug in the ntfs driver!");
-        i_size_write(vi, ntfs_attr_size(a)); 
-done:
+        /*
+         * The i_size of the vfs inode is the new size for the attribute value.
+         */
+        new_size = i_size_read(vi);
+        /* The current size of the attribute value is the old size. */
+        old_size = ntfs_attr_size(a);
+        /* Calculate the new allocated size. */
+        if (NInoNonResident(ni))
+                new_alloc_size = (new_size + vol->cluster_size - 1) &
+                                ~(s64)vol->cluster_size_mask;
+        else
+                new_alloc_size = (new_size + 7) & ~7;
+        /* The current allocated size is the old allocated size. */
+        read_lock_irqsave(&ni->size_lock, flags);
+        old_alloc_size = ni->allocated_size;
+        read_unlock_irqrestore(&ni->size_lock, flags);
+        /*
+         * The change in the file size.  This will be 0 if no change, >0 if the
+         * size is growing, and <0 if the size is shrinking.
+         */
+        size_change = -1;
+        if (new_size - old_size >= 0) {
+                size_change = 1;
+                if (new_size == old_size)
+                        size_change = 0;
+        }
+        /* As above for the allocated size. */
+        alloc_change = -1;
+        if (new_alloc_size - old_alloc_size >= 0) {
+                alloc_change = 1;
+                if (new_alloc_size == old_alloc_size)
+                        alloc_change = 0;
+        }
+        /*
+         * If neither the size nor the allocation are being changed there is
+         * nothing to do.
+         */
+        if (!size_change && !alloc_change)
+                goto unm_done;
+        /* If the size is changing, check if new size is allowed in $AttrDef. */
+        if (size_change) {
+                err = ntfs_attr_size_bounds_check(vol, ni->type, new_size);
+                if (unlikely(err)) {
+                        if (err == -ERANGE) {
+                                ntfs_error(vol->sb, "Truncate would cause the "
+                                                "inode 0x%lx to %simum size "
+                                                "for its attribute type "
+                                                "(0x%x).  Aborting truncate.",
+                                                vi->i_ino,
+                                                new_size > old_size ? "exceed "
+                                                "the max" : "go under the min",
+                                                le32_to_cpu(ni->type));
+                                err = -EFBIG;
+                        } else {
+                                ntfs_error(vol->sb, "Inode 0x%lx has unknown "
+                                                "attribute type 0x%x.  "
+                                                "Aborting truncate.",
+                                                vi->i_ino,
+                                                le32_to_cpu(ni->type));
+                                err = -EIO;
+                        }
+                        /* Reset the vfs inode size to the old size. */
+                        i_size_write(vi, old_size);
+                        goto err_out;
+                }
+        }
+        if (NInoCompressed(ni) || NInoEncrypted(ni)) {
+                ntfs_warning(vi->i_sb, "Changes in inode size are not "
+                                "supported yet for %s files, ignoring.",
+                                NInoCompressed(ni) ? "compressed" :
+                                "encrypted");
+                err = -EOPNOTSUPP;
+                goto bad_out;
+        }
+        if (a->non_resident)
+                goto do_non_resident_truncate;
+        BUG_ON(NInoNonResident(ni));
+        /* Resize the attribute record to best fit the new attribute size. */
+        if (new_size < vol->mft_record_size &&
+                        !ntfs_resident_attr_value_resize(m, a, new_size)) {
+                unsigned long flags;
+
+                /* The resize succeeded! */
+                flush_dcache_mft_record_page(ctx->ntfs_ino);
+                mark_mft_record_dirty(ctx->ntfs_ino);
+                write_lock_irqsave(&ni->size_lock, flags);
+                /* Update the sizes in the ntfs inode and all is done. */
+                ni->allocated_size = le32_to_cpu(a->length) -
+                                le16_to_cpu(a->data.resident.value_offset);
+                /*
+                 * Note ntfs_resident_attr_value_resize() has already done any
+                 * necessary data clearing in the attribute record.  When the
+                 * file is being shrunk vmtruncate() will already have cleared
+                 * the top part of the last partial page, i.e. since this is
+                 * the resident case this is the page with index 0.  However,
+                 * when the file is being expanded, the page cache page data
+                 * between the old data_size, i.e. old_size, and the new_size
+                 * has not been zeroed.  Fortunately, we do not need to zero it
+                 * either since on one hand it will either already be zero due
+                 * to both readpage and writepage clearing partial page data
+                 * beyond i_size in which case there is nothing to do or in the
+                 * case of the file being mmap()ped at the same time, POSIX
+                 * specifies that the behaviour is unspecified thus we do not
+                 * have to do anything.  This means that in our implementation
+                 * in the rare case that the file is mmap()ped and a write
+                 * occured into the mmap()ped region just beyond the file size
+                 * and writepage has not yet been called to write out the page
+                 * (which would clear the area beyond the file size) and we now
+                 * extend the file size to incorporate this dirty region
+                 * outside the file size, a write of the page would result in
+                 * this data being written to disk instead of being cleared.
+                 * Given both POSIX and the Linux mmap(2) man page specify that
+                 * this corner case is undefined, we choose to leave it like
+                 * that as this is much simpler for us as we cannot lock the
+                 * relevant page now since we are holding too many ntfs locks
+                 * which would result in a lock reversal deadlock.
+                 */
+                ni->initialized_size = new_size;
+                write_unlock_irqrestore(&ni->size_lock, flags);
+                goto unm_done;
+        }
+        /* If the above resize failed, this must be an attribute extension. */
+        BUG_ON(size_change < 0);
+        /*
+         * We have to drop all the locks so we can call
+         * ntfs_attr_make_non_resident().  This could be optimised by try-
+         * locking the first page cache page and only if that fails dropping
+         * the locks, locking the page, and redoing all the locking and
+         * lookups.  While this would be a huge optimisation, it is not worth
+         * it as this is definitely a slow code path as it only ever can happen
+         * once for any given file.
+         */
         ntfs_attr_put_search_ctx(ctx);
-        unmap_mft_record(ni);
-        NInoClearTruncateFailed(ni);
-        ntfs_debug("Done.");
-        return 0;
-err_out:
-        if (err != -ENOMEM) {
+        unmap_mft_record(base_ni);
+        up_write(&ni->runlist.lock);
+        /*
+         * Not enough space in the mft record, try to make the attribute
+         * non-resident and if successful restart the truncation process.
+         */
+        err = ntfs_attr_make_non_resident(ni, old_size);
+        if (likely(!err))
+                goto retry_truncate;
+        /*
+         * Could not make non-resident.  If this is due to this not being
+         * permitted for this attribute type or there not being enough space,
+         * try to make other attributes non-resident.  Otherwise fail.
+         */
+        if (unlikely(err != -EPERM && err != -ENOSPC)) {
+                ntfs_error(vol->sb, "Cannot truncate inode 0x%lx, attribute "
+                                "type 0x%x, because the conversion from "
+                                "resident to non-resident attribute failed "
+                                "with error code %i.", vi->i_ino,
+                                (unsigned)le32_to_cpu(ni->type), err);
+                if (err != -ENOMEM)
+                        err = -EIO;
+                goto conv_err_out;
+        }
+        /* TODO: Not implemented from here, abort. */
+        if (err == -ENOSPC)
+                ntfs_error(vol->sb, "Not enough space in the mft record/on "
+                                "disk for the non-resident attribute value.  "
+                                "This case is not implemented yet.");
+        else /* if (err == -EPERM) */
+                ntfs_error(vol->sb, "This attribute type may not be "
+                                "non-resident.  This case is not implemented "
+                                "yet.");
+        err = -EOPNOTSUPP;
+        goto conv_err_out;
+#if 0
+        // TODO: Attempt to make other attributes non-resident.
+        if (!err)
+                goto do_resident_extend;
+        /*
+         * Both the attribute list attribute and the standard information
+         * attribute must remain in the base inode.  Thus, if this is one of
+         * these attributes, we have to try to move other attributes out into
+         * extent mft records instead.
+         */
+        if (ni->type == AT_ATTRIBUTE_LIST ||
+                        ni->type == AT_STANDARD_INFORMATION) {
+                // TODO: Attempt to move other attributes into extent mft
+                // records.
+                err = -EOPNOTSUPP;
+                if (!err)
+                        goto do_resident_extend;
+                goto err_out;
+        }
+        // TODO: Attempt to move this attribute to an extent mft record, but
+        // only if it is not already the only attribute in an mft record in
+        // which case there would be nothing to gain.
+        err = -EOPNOTSUPP;
+        if (!err)
+                goto do_resident_extend;
+        /* There is nothing we can do to make enough space. )-: */
+        goto err_out;
+#endif
+do_non_resident_truncate:
+        BUG_ON(!NInoNonResident(ni));
+        if (alloc_change < 0) {
+                highest_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn);
+                if (highest_vcn > 0 &&
+                                old_alloc_size >> vol->cluster_size_bits >
+                                highest_vcn + 1) {
+                        /*
+                         * This attribute has multiple extents.  Not yet
+                         * supported.
+                         */
+                        ntfs_error(vol->sb, "Cannot truncate inode 0x%lx, "
+                                        "attribute type 0x%x, because the "
+                                        "attribute is highly fragmented (it "
+                                        "consists of multiple extents) and "
+                                        "this case is not implemented yet.",
+                                        vi->i_ino,
+                                        (unsigned)le32_to_cpu(ni->type));
+                        err = -EOPNOTSUPP;
+                        goto bad_out;
+                }
+        }
+        /*
+         * If the size is shrinking, need to reduce the initialized_size and
+         * the data_size before reducing the allocation.
+         */
+        if (size_change < 0) {
+                /*
+                 * Make the valid size smaller (i_size is already up-to-date).
+                 */
+                write_lock_irqsave(&ni->size_lock, flags);
+                if (new_size < ni->initialized_size) {
+                        ni->initialized_size = new_size;
+                        a->data.non_resident.initialized_size =
+                                        cpu_to_sle64(new_size);
+                }
+                a->data.non_resident.data_size = cpu_to_sle64(new_size);
+                write_unlock_irqrestore(&ni->size_lock, flags);
+                flush_dcache_mft_record_page(ctx->ntfs_ino);
+                mark_mft_record_dirty(ctx->ntfs_ino);
+                /* If the allocated size is not changing, we are done. */
+                if (!alloc_change)
+                        goto unm_done;
+                /*
+                 * If the size is shrinking it makes no sense for the
+                 * allocation to be growing.
+                 */
+                BUG_ON(alloc_change > 0);
+        } else /* if (size_change >= 0) */ {
+                /*
+                 * The file size is growing or staying the same but the
+                 * allocation can be shrinking, growing or staying the same.
+                 */
+                if (alloc_change > 0) {
+                        /*
+                         * We need to extend the allocation and possibly update
+                         * the data size.  If we are updating the data size,
+                         * since we are not touching the initialized_size we do
+                         * not need to worry about the actual data on disk.
+                         * And as far as the page cache is concerned, there
+                         * will be no pages beyond the old data size and any
+                         * partial region in the last page between the old and
+                         * new data size (or the end of the page if the new
+                         * data size is outside the page) does not need to be
+                         * modified as explained above for the resident
+                         * attribute truncate case.  To do this, we simply drop
+                         * the locks we hold and leave all the work to our
+                         * friendly helper ntfs_attr_extend_allocation().
+                         */
+                        ntfs_attr_put_search_ctx(ctx);
+                        unmap_mft_record(base_ni);
+                        up_write(&ni->runlist.lock);
+                        err = ntfs_attr_extend_allocation(ni, new_size,
+                                        size_change > 0 ? new_size : -1, -1);
+                        /*
+                         * ntfs_attr_extend_allocation() will have done error
+                         * output already.
+                         */
+                        goto done;
+                }
+                if (!alloc_change)
+                        goto alloc_done;
+        }
+        /* alloc_change < 0 */
+        /* Free the clusters. */
+        nr_freed = ntfs_cluster_free(ni, new_alloc_size >>
+                        vol->cluster_size_bits, -1, ctx);
+        m = ctx->mrec;
+        a = ctx->attr;
+        if (unlikely(nr_freed < 0)) {
+                ntfs_error(vol->sb, "Failed to release cluster(s) (error code "
+                                "%lli).  Unmount and run chkdsk to recover "
+                                "the lost cluster(s).", (long long)nr_freed);
                 NVolSetErrors(vol);
+                nr_freed = 0;
+        }
+        /* Truncate the runlist. */
+        err = ntfs_rl_truncate_nolock(vol, &ni->runlist,
+                        new_alloc_size >> vol->cluster_size_bits);
+        /*
+         * If the runlist truncation failed and/or the search context is no
+         * longer valid, we cannot resize the attribute record or build the
+         * mapping pairs array thus we mark the inode bad so that no access to
+         * the freed clusters can happen.
+         */
+        if (unlikely(err || IS_ERR(m))) {
+                ntfs_error(vol->sb, "Failed to %s (error code %li).%s",
+                                IS_ERR(m) ?
+                                "restore attribute search context" :
+                                "truncate attribute runlist",
+                                IS_ERR(m) ? PTR_ERR(m) : err, es);
+                err = -EIO;
+                goto bad_out;
+        }
+        /* Get the size for the shrunk mapping pairs array for the runlist. */
+        mp_size = ntfs_get_size_for_mapping_pairs(vol, ni->runlist.rl, 0, -1);
+        if (unlikely(mp_size <= 0)) {
+                ntfs_error(vol->sb, "Cannot shrink allocation of inode 0x%lx, "
+                                "attribute type 0x%x, because determining the "
+                                "size for the mapping pairs failed with error "
+                                "code %i.%s", vi->i_ino,
+                                (unsigned)le32_to_cpu(ni->type), mp_size, es);
+                err = -EIO;
+                goto bad_out;
+        }
+        /*
+         * Shrink the attribute record for the new mapping pairs array.  Note,
+         * this cannot fail since we are making the attribute smaller thus by
+         * definition there is enough space to do so.
+         */
+        attr_len = le32_to_cpu(a->length);
+        err = ntfs_attr_record_resize(m, a, mp_size +
+                        le16_to_cpu(a->data.non_resident.mapping_pairs_offset));
+        BUG_ON(err);
+        /*
+         * Generate the mapping pairs array directly into the attribute record.
+         */
+        err = ntfs_mapping_pairs_build(vol, (u8*)a +
+                        le16_to_cpu(a->data.non_resident.mapping_pairs_offset),
+                        mp_size, ni->runlist.rl, 0, -1, NULL);
+        if (unlikely(err)) {
+                ntfs_error(vol->sb, "Cannot shrink allocation of inode 0x%lx, "
+                                "attribute type 0x%x, because building the "
+                                "mapping pairs failed with error code %i.%s",
+                                vi->i_ino, (unsigned)le32_to_cpu(ni->type),
+                                err, es);
+                err = -EIO;
+                goto bad_out;
+        }
+        /* Update the allocated/compressed size as well as the highest vcn. */
+        a->data.non_resident.highest_vcn = cpu_to_sle64((new_alloc_size >>
+                        vol->cluster_size_bits) - 1);
+        write_lock_irqsave(&ni->size_lock, flags);
+        ni->allocated_size = new_alloc_size;
+        a->data.non_resident.allocated_size = cpu_to_sle64(new_alloc_size);
+        if (NInoSparse(ni) || NInoCompressed(ni)) {
+                if (nr_freed) {
+                        ni->itype.compressed.size -= nr_freed <<
+                                        vol->cluster_size_bits;
+                        BUG_ON(ni->itype.compressed.size < 0);
+                        a->data.non_resident.compressed_size = cpu_to_sle64(
+                                        ni->itype.compressed.size);
+                        vi->i_blocks = ni->itype.compressed.size >> 9;
+                }
+        } else
+                vi->i_blocks = new_alloc_size >> 9;
+        write_unlock_irqrestore(&ni->size_lock, flags);
+        /*
+         * We have shrunk the allocation.  If this is a shrinking truncate we
+         * have already dealt with the initialized_size and the data_size above
+         * and we are done.  If the truncate is only changing the allocation
+         * and not the data_size, we are also done.  If this is an extending
+         * truncate, need to extend the data_size now which is ensured by the
+         * fact that @size_change is positive.
+         */
+alloc_done:
+        /*
+         * If the size is growing, need to update it now.  If it is shrinking,
+         * we have already updated it above (before the allocation change).
+         */
+        if (size_change > 0)
+                a->data.non_resident.data_size = cpu_to_sle64(new_size);
+        /* Ensure the modified mft record is written out. */
+        flush_dcache_mft_record_page(ctx->ntfs_ino);
+        mark_mft_record_dirty(ctx->ntfs_ino);
+unm_done:
+        ntfs_attr_put_search_ctx(ctx);
+        unmap_mft_record(base_ni);
+        up_write(&ni->runlist.lock);
+done:
+        /* Update the mtime and ctime on the base inode. */
+        inode_update_time(VFS_I(base_ni), 1);
+        if (likely(!err)) {
+                NInoClearTruncateFailed(ni);
+                ntfs_debug("Done.");
+        }
+        return err;
+old_bad_out:
+        old_size = -1;
+bad_out:
+        if (err != -ENOMEM && err != -EOPNOTSUPP) {
                 make_bad_inode(vi);
+                make_bad_inode(VFS_I(base_ni));
+                NVolSetErrors(vol);
         }
+        if (err != -EOPNOTSUPP)
+                NInoSetTruncateFailed(ni);
+        else if (old_size >= 0)
+                i_size_write(vi, old_size);
+err_out:
         if (ctx)
                 ntfs_attr_put_search_ctx(ctx);
         if (m)
-                unmap_mft_record(ni);
-        NInoSetTruncateFailed(ni);
+                unmap_mft_record(base_ni);
+        up_write(&ni->runlist.lock);
+out:
+        ntfs_debug("Failed.  Returning error code %i.", err);
         return err;
+conv_err_out:
+        if (err != -ENOMEM && err != -EOPNOTSUPP) {
+                make_bad_inode(vi);
+                make_bad_inode(VFS_I(base_ni));
+                NVolSetErrors(vol);
+        }
+        if (err != -EOPNOTSUPP)
+                NInoSetTruncateFailed(ni);
+        else
+                i_size_write(vi, old_size);
+        goto out;
 }
 
 /**
@@ -2420,8 +2845,7 @@ int ntfs_setattr(struct dentry *dentry, struct iattr *attr)
 
         err = inode_change_ok(vi, attr);
         if (err)
-                return err;
-
+                goto out;
         /* We do not support NTFS ACLs yet. */
         if (ia_valid & (ATTR_UID | ATTR_GID | ATTR_MODE)) {
                 ntfs_warning(vi->i_sb, "Changes in user/group/mode are not "
@@ -2429,14 +2853,22 @@ int ntfs_setattr(struct dentry *dentry, struct iattr *attr)
                 err = -EOPNOTSUPP;
                 goto out;
         }
-
         if (ia_valid & ATTR_SIZE) {
                 if (attr->ia_size != i_size_read(vi)) {
-                        ntfs_warning(vi->i_sb, "Changes in inode size are not "
-                                        "supported yet, ignoring.");
-                        err = -EOPNOTSUPP;
-                        // TODO: Implement...
-                        // err = vmtruncate(vi, attr->ia_size);
+                        ntfs_inode *ni = NTFS_I(vi);
+                        /*
+                         * FIXME: For now we do not support resizing of
+                         * compressed or encrypted files yet.
+                         */
+                        if (NInoCompressed(ni) || NInoEncrypted(ni)) {
+                                ntfs_warning(vi->i_sb, "Changes in inode size "
+                                                "are not supported yet for "
+                                                "%s files, ignoring.",
+                                                NInoCompressed(ni) ?
+                                                "compressed" : "encrypted");
+                                err = -EOPNOTSUPP;
+                        } else
+                                err = vmtruncate(vi, attr->ia_size);
                         if (err || ia_valid == ATTR_SIZE)
                                 goto out;
                 } else {
diff --git a/fs/ntfs/layout.h b/fs/ntfs/layout.h
index 5c248d404f05..f5678d5d7919 100644
--- a/fs/ntfs/layout.h
+++ b/fs/ntfs/layout.h
@@ -1021,10 +1021,17 @@ enum {
         FILE_NAME_POSIX         = 0x00,
         /* This is the largest namespace. It is case sensitive and allows all
            Unicode characters except for: '\0' and '/'.  Beware that in
-           WinNT/2k files which eg have the same name except for their case
-           will not be distinguished by the standard utilities and thus a "del
-           filename" will delete both "filename" and "fileName" without
-           warning. */
+           WinNT/2k/2003 by default files which eg have the same name except
+           for their case will not be distinguished by the standard utilities
+           and thus a "del filename" will delete both "filename" and "fileName"
+           without warning.  However if for example Services For Unix (SFU) are
+           installed and the case sensitive option was enabled at installation
+           time, then you can create/access/delete such files.
+           Note that even SFU places restrictions on the filenames beyond the
+           '\0' and '/' and in particular the following set of characters is
+           not allowed: '"', '/', '<', '>', '\'.  All other characters,
+           including the ones no allowed in WIN32 namespace are allowed.
+           Tested with SFU 3.5 (this is now free) running on Windows XP. */
         FILE_NAME_WIN32         = 0x01,
         /* The standard WinNT/2k NTFS long filenames. Case insensitive.  All
            Unicode chars except: '\0', '"', '*', '/', ':', '<', '>', '?', '\',
@@ -2367,7 +2374,9 @@ typedef struct {
  * Extended attribute flags (8-bit).
  */
 enum {
-        NEED_EA = 0x80
+        NEED_EA = 0x80          /* If set the file to which the EA belongs
+                                   cannot be interpreted without understanding
+                                   the associates extended attributes. */
 } __attribute__ ((__packed__));
 
 typedef u8 EA_FLAGS;
@@ -2375,20 +2384,20 @@ typedef u8 EA_FLAGS;
 /*
  * Attribute: Extended attribute (EA) (0xe0).
  *
- * NOTE: Always non-resident. (Is this true?)
+ * NOTE: Can be resident or non-resident.
  *
  * Like the attribute list and the index buffer list, the EA attribute value is
  * a sequence of EA_ATTR variable length records.
- *
- * FIXME: It appears weird that the EA name is not unicode. Is it true?
  */
 typedef struct {
         le32 next_entry_offset; /* Offset to the next EA_ATTR. */
         EA_FLAGS flags;         /* Flags describing the EA. */
-        u8 ea_name_length;      /* Length of the name of the EA in bytes. */
+        u8 ea_name_length;      /* Length of the name of the EA in bytes
+                                   excluding the '\0' byte terminator. */
         le16 ea_value_length;   /* Byte size of the EA's value. */
-        u8 ea_name[0];          /* Name of the EA. */
-        u8 ea_value[0];         /* The value of the EA. Immediately follows
+        u8 ea_name[0];          /* Name of the EA.  Note this is ASCII, not
+                                   Unicode and it is zero terminated. */
+        u8 ea_value[0];         /* The value of the EA.  Immediately follows
                                    the name. */
 } __attribute__ ((__packed__)) EA_ATTR;
 
diff --git a/fs/ntfs/lcnalloc.c b/fs/ntfs/lcnalloc.c
index 5af3bf0b7eee..29cabf93d2d2 100644
--- a/fs/ntfs/lcnalloc.c
+++ b/fs/ntfs/lcnalloc.c
@@ -76,6 +76,7 @@ int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
  * @count:      number of clusters to allocate
  * @start_lcn:  starting lcn at which to allocate the clusters (or -1 if none)
  * @zone:       zone from which to allocate the clusters
+ * @is_extension:       if TRUE, this is an attribute extension
  *
  * Allocate @count clusters preferably starting at cluster @start_lcn or at the
  * current allocator position if @start_lcn is -1, on the mounted ntfs volume
@@ -86,6 +87,13 @@ int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
  * @start_vcn specifies the vcn of the first allocated cluster.  This makes
  * merging the resulting runlist with the old runlist easier.
  *
+ * If @is_extension is TRUE, the caller is allocating clusters to extend an
+ * attribute and if it is FALSE, the caller is allocating clusters to fill a
+ * hole in an attribute.  Practically the difference is that if @is_extension
+ * is TRUE the returned runlist will be terminated with LCN_ENOENT and if
+ * @is_extension is FALSE the runlist will be terminated with
+ * LCN_RL_NOT_MAPPED.
+ *
  * You need to check the return value with IS_ERR().  If this is false, the
  * function was successful and the return value is a runlist describing the
  * allocated cluster(s).  If IS_ERR() is true, the function failed and
@@ -137,7 +145,8 @@ int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
  */
 runlist_element *ntfs_cluster_alloc(ntfs_volume *vol, const VCN start_vcn,
                 const s64 count, const LCN start_lcn,
-                const NTFS_CLUSTER_ALLOCATION_ZONES zone)
+                const NTFS_CLUSTER_ALLOCATION_ZONES zone,
+                const BOOL is_extension)
 {
         LCN zone_start, zone_end, bmp_pos, bmp_initial_pos, last_read_pos, lcn;
         LCN prev_lcn = 0, prev_run_len = 0, mft_zone_size;
@@ -310,7 +319,7 @@ runlist_element *ntfs_cluster_alloc(ntfs_volume *vol, const VCN start_vcn,
                                 continue;
                         }
                         bit = 1 << (lcn & 7);
-                        ntfs_debug("bit %i.", bit);
+                        ntfs_debug("bit 0x%x.", bit);
                         /* If the bit is already set, go onto the next one. */
                         if (*byte & bit) {
                                 lcn++;
@@ -729,7 +738,7 @@ out:
         /* Add runlist terminator element. */
         if (likely(rl)) {
                 rl[rlpos].vcn = rl[rlpos - 1].vcn + rl[rlpos - 1].length;
-                rl[rlpos].lcn = LCN_RL_NOT_MAPPED;
+                rl[rlpos].lcn = is_extension ? LCN_ENOENT : LCN_RL_NOT_MAPPED;
                 rl[rlpos].length = 0;
         }
         if (likely(page && !IS_ERR(page))) {
@@ -782,6 +791,7 @@ out:
  * @ni:         ntfs inode whose runlist describes the clusters to free
  * @start_vcn:  vcn in the runlist of @ni at which to start freeing clusters
  * @count:      number of clusters to free or -1 for all clusters
+ * @ctx:        active attribute search context if present or NULL if not
  * @is_rollback:        true if this is a rollback operation
  *
  * Free @count clusters starting at the cluster @start_vcn in the runlist
@@ -791,15 +801,39 @@ out:
  * deallocated.  Thus, to completely free all clusters in a runlist, use
  * @start_vcn = 0 and @count = -1.
  *
+ * If @ctx is specified, it is an active search context of @ni and its base mft
+ * record.  This is needed when __ntfs_cluster_free() encounters unmapped
+ * runlist fragments and allows their mapping.  If you do not have the mft
+ * record mapped, you can specify @ctx as NULL and __ntfs_cluster_free() will
+ * perform the necessary mapping and unmapping.
+ *
+ * Note, __ntfs_cluster_free() saves the state of @ctx on entry and restores it
+ * before returning.  Thus, @ctx will be left pointing to the same attribute on
+ * return as on entry.  However, the actual pointers in @ctx may point to
+ * different memory locations on return, so you must remember to reset any
+ * cached pointers from the @ctx, i.e. after the call to __ntfs_cluster_free(),
+ * you will probably want to do:
+ *      m = ctx->mrec;
+ *      a = ctx->attr;
+ * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that
+ * you cache ctx->mrec in a variable @m of type MFT_RECORD *.
+ *
  * @is_rollback should always be FALSE, it is for internal use to rollback
  * errors.  You probably want to use ntfs_cluster_free() instead.
  *
- * Note, ntfs_cluster_free() does not modify the runlist at all, so the caller
- * has to deal with it later.
+ * Note, __ntfs_cluster_free() does not modify the runlist, so you have to
+ * remove from the runlist or mark sparse the freed runs later.
  *
  * Return the number of deallocated clusters (not counting sparse ones) on
  * success and -errno on error.
  *
+ * WARNING: If @ctx is supplied, regardless of whether success or failure is
+ *          returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx
+ *          is no longer valid, i.e. you need to either call
+ *          ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it.
+ *          In that case PTR_ERR(@ctx->mrec) will give you the error code for
+ *          why the mapping of the old inode failed.
+ *
  * Locking: - The runlist described by @ni must be locked for writing on entry
  *            and is locked on return.  Note the runlist may be modified when
  *            needed runlist fragments need to be mapped.
@@ -807,9 +841,13 @@ out:
  *            on return.
  *          - This function takes the volume lcn bitmap lock for writing and
  *            modifies the bitmap contents.
+ *          - If @ctx is NULL, the base mft record of @ni must not be mapped on
+ *            entry and it will be left unmapped on return.
+ *          - If @ctx is not NULL, the base mft record must be mapped on entry
+ *            and it will be left mapped on return.
  */
 s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn, s64 count,
-                const BOOL is_rollback)
+                ntfs_attr_search_ctx *ctx, const BOOL is_rollback)
 {
         s64 delta, to_free, total_freed, real_freed;
         ntfs_volume *vol;
@@ -839,7 +877,7 @@ s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn, s64 count,
 
         total_freed = real_freed = 0;
 
-        rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, TRUE);
+        rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, ctx);
         if (IS_ERR(rl)) {
                 if (!is_rollback)
                         ntfs_error(vol->sb, "Failed to find first runlist "
@@ -893,7 +931,7 @@ s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn, s64 count,
 
                         /* Attempt to map runlist. */
                         vcn = rl->vcn;
-                        rl = ntfs_attr_find_vcn_nolock(ni, vcn, TRUE);
+                        rl = ntfs_attr_find_vcn_nolock(ni, vcn, ctx);
                         if (IS_ERR(rl)) {
                                 err = PTR_ERR(rl);
                                 if (!is_rollback)
@@ -961,7 +999,7 @@ err_out:
          * If rollback fails, set the volume errors flag, emit an error
          * message, and return the error code.
          */
-        delta = __ntfs_cluster_free(ni, start_vcn, total_freed, TRUE);
+        delta = __ntfs_cluster_free(ni, start_vcn, total_freed, ctx, TRUE);
         if (delta < 0) {
                 ntfs_error(vol->sb, "Failed to rollback (error %i).  Leaving "
                                 "inconsistent metadata!  Unmount and run "
diff --git a/fs/ntfs/lcnalloc.h b/fs/ntfs/lcnalloc.h
index a6a8827882e7..72cbca7003b2 100644
--- a/fs/ntfs/lcnalloc.h
+++ b/fs/ntfs/lcnalloc.h
@@ -27,6 +27,7 @@
 
 #include <linux/fs.h>
 
+#include "attrib.h"
 #include "types.h"
 #include "inode.h"
 #include "runlist.h"
@@ -41,16 +42,18 @@ typedef enum {
 
 extern runlist_element *ntfs_cluster_alloc(ntfs_volume *vol,
                 const VCN start_vcn, const s64 count, const LCN start_lcn,
-                const NTFS_CLUSTER_ALLOCATION_ZONES zone);
+                const NTFS_CLUSTER_ALLOCATION_ZONES zone,
+                const BOOL is_extension);
 
 extern s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn,
-                s64 count, const BOOL is_rollback);
+                s64 count, ntfs_attr_search_ctx *ctx, const BOOL is_rollback);
 
 /**
  * ntfs_cluster_free - free clusters on an ntfs volume
  * @ni:         ntfs inode whose runlist describes the clusters to free
  * @start_vcn:  vcn in the runlist of @ni at which to start freeing clusters
  * @count:      number of clusters to free or -1 for all clusters
+ * @ctx:        active attribute search context if present or NULL if not
  *
  * Free @count clusters starting at the cluster @start_vcn in the runlist
  * described by the ntfs inode @ni.
@@ -59,12 +62,36 @@ extern s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn,
  * deallocated.  Thus, to completely free all clusters in a runlist, use
  * @start_vcn = 0 and @count = -1.
  *
- * Note, ntfs_cluster_free() does not modify the runlist at all, so the caller
- * has to deal with it later.
+ * If @ctx is specified, it is an active search context of @ni and its base mft
+ * record.  This is needed when ntfs_cluster_free() encounters unmapped runlist
+ * fragments and allows their mapping.  If you do not have the mft record
+ * mapped, you can specify @ctx as NULL and ntfs_cluster_free() will perform
+ * the necessary mapping and unmapping.
+ *
+ * Note, ntfs_cluster_free() saves the state of @ctx on entry and restores it
+ * before returning.  Thus, @ctx will be left pointing to the same attribute on
+ * return as on entry.  However, the actual pointers in @ctx may point to
+ * different memory locations on return, so you must remember to reset any
+ * cached pointers from the @ctx, i.e. after the call to ntfs_cluster_free(),
+ * you will probably want to do:
+ *      m = ctx->mrec;
+ *      a = ctx->attr;
+ * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that
+ * you cache ctx->mrec in a variable @m of type MFT_RECORD *.
+ *
+ * Note, ntfs_cluster_free() does not modify the runlist, so you have to remove
+ * from the runlist or mark sparse the freed runs later.
  *
  * Return the number of deallocated clusters (not counting sparse ones) on
  * success and -errno on error.
  *
+ * WARNING: If @ctx is supplied, regardless of whether success or failure is
+ *          returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx
+ *          is no longer valid, i.e. you need to either call
+ *          ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it.
+ *          In that case PTR_ERR(@ctx->mrec) will give you the error code for
+ *          why the mapping of the old inode failed.
+ *
  * Locking: - The runlist described by @ni must be locked for writing on entry
  *            and is locked on return.  Note the runlist may be modified when
  *            needed runlist fragments need to be mapped.
@@ -72,11 +99,15 @@ extern s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn,
  *            on return.
  *          - This function takes the volume lcn bitmap lock for writing and
  *            modifies the bitmap contents.
+ *          - If @ctx is NULL, the base mft record of @ni must not be mapped on
+ *            entry and it will be left unmapped on return.
+ *          - If @ctx is not NULL, the base mft record must be mapped on entry
+ *            and it will be left mapped on return.
  */
 static inline s64 ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn,
-                s64 count)
+                s64 count, ntfs_attr_search_ctx *ctx)
 {
-        return __ntfs_cluster_free(ni, start_vcn, count, FALSE);
+        return __ntfs_cluster_free(ni, start_vcn, count, ctx, FALSE);
 }
 
 extern int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
diff --git a/fs/ntfs/malloc.h b/fs/ntfs/malloc.h
index 590887b943f5..e38e402e4103 100644
--- a/fs/ntfs/malloc.h
+++ b/fs/ntfs/malloc.h
@@ -39,8 +39,7 @@
  * If there was insufficient memory to complete the request, return NULL.
  * Depending on @gfp_mask the allocation may be guaranteed to succeed.
  */
-static inline void *__ntfs_malloc(unsigned long size,
-                gfp_t gfp_mask)
+static inline void *__ntfs_malloc(unsigned long size, gfp_t gfp_mask)
 {
         if (likely(size <= PAGE_SIZE)) {
                 BUG_ON(!size);
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index b011369b5956..0c65cbb8c5cf 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -49,7 +49,8 @@ static inline MFT_RECORD *map_mft_record_page(ntfs_inode *ni)
         ntfs_volume *vol = ni->vol;
         struct inode *mft_vi = vol->mft_ino;
         struct page *page;
-        unsigned long index, ofs, end_index;
+        unsigned long index, end_index;
+        unsigned ofs;
 
         BUG_ON(ni->page);
         /*
@@ -1308,7 +1309,7 @@ static int ntfs_mft_bitmap_extend_allocation_nolock(ntfs_volume *vol)
         ll = mftbmp_ni->allocated_size;
         read_unlock_irqrestore(&mftbmp_ni->size_lock, flags);
         rl = ntfs_attr_find_vcn_nolock(mftbmp_ni,
-                        (ll - 1) >> vol->cluster_size_bits, TRUE);
+                        (ll - 1) >> vol->cluster_size_bits, NULL);
         if (unlikely(IS_ERR(rl) || !rl->length || rl->lcn < 0)) {
                 up_write(&mftbmp_ni->runlist.lock);
                 ntfs_error(vol->sb, "Failed to determine last allocated "
@@ -1354,7 +1355,8 @@ static int ntfs_mft_bitmap_extend_allocation_nolock(ntfs_volume *vol)
                 up_write(&vol->lcnbmp_lock);
                 ntfs_unmap_page(page);
                 /* Allocate a cluster from the DATA_ZONE. */
-                rl2 = ntfs_cluster_alloc(vol, rl[1].vcn, 1, lcn, DATA_ZONE);
+                rl2 = ntfs_cluster_alloc(vol, rl[1].vcn, 1, lcn, DATA_ZONE,
+                                TRUE);
                 if (IS_ERR(rl2)) {
                         up_write(&mftbmp_ni->runlist.lock);
                         ntfs_error(vol->sb, "Failed to allocate a cluster for "
@@ -1738,7 +1740,7 @@ static int ntfs_mft_data_extend_allocation_nolock(ntfs_volume *vol)
         ll = mft_ni->allocated_size;
         read_unlock_irqrestore(&mft_ni->size_lock, flags);
         rl = ntfs_attr_find_vcn_nolock(mft_ni,
-                        (ll - 1) >> vol->cluster_size_bits, TRUE);
+                        (ll - 1) >> vol->cluster_size_bits, NULL);
         if (unlikely(IS_ERR(rl) || !rl->length || rl->lcn < 0)) {
                 up_write(&mft_ni->runlist.lock);
                 ntfs_error(vol->sb, "Failed to determine last allocated "
@@ -1779,7 +1781,8 @@ static int ntfs_mft_data_extend_allocation_nolock(ntfs_volume *vol)
                         nr > min_nr ? "default" : "minimal", (long long)nr);
         old_last_vcn = rl[1].vcn;
         do {
-                rl2 = ntfs_cluster_alloc(vol, old_last_vcn, nr, lcn, MFT_ZONE);
+                rl2 = ntfs_cluster_alloc(vol, old_last_vcn, nr, lcn, MFT_ZONE,
+                                TRUE);
                 if (likely(!IS_ERR(rl2)))
                         break;
                 if (PTR_ERR(rl2) != -ENOSPC || nr == min_nr) {
@@ -1951,20 +1954,21 @@ restore_undo_alloc:
                 NVolSetErrors(vol);
                 return ret;
         }
-        a = ctx->attr;
-        a->data.non_resident.highest_vcn = cpu_to_sle64(old_last_vcn - 1);
+        ctx->attr->data.non_resident.highest_vcn =
+                        cpu_to_sle64(old_last_vcn - 1);
 undo_alloc:
-        if (ntfs_cluster_free(mft_ni, old_last_vcn, -1) < 0) {
+        if (ntfs_cluster_free(mft_ni, old_last_vcn, -1, ctx) < 0) {
                 ntfs_error(vol->sb, "Failed to free clusters from mft data "
                                 "attribute.%s", es);
                 NVolSetErrors(vol);
         }
+        a = ctx->attr;
         if (ntfs_rl_truncate_nolock(vol, &mft_ni->runlist, old_last_vcn)) {
                 ntfs_error(vol->sb, "Failed to truncate mft data attribute "
                                 "runlist.%s", es);
                 NVolSetErrors(vol);
         }
-        if (mp_rebuilt) {
+        if (mp_rebuilt && !IS_ERR(ctx->mrec)) {
                 if (ntfs_mapping_pairs_build(vol, (u8*)a + le16_to_cpu(
                                 a->data.non_resident.mapping_pairs_offset),
                                 old_alen - le16_to_cpu(
@@ -1981,6 +1985,10 @@ undo_alloc:
                 }
                 flush_dcache_mft_record_page(ctx->ntfs_ino);
                 mark_mft_record_dirty(ctx->ntfs_ino);
+        } else if (IS_ERR(ctx->mrec)) {
+                ntfs_error(vol->sb, "Failed to restore attribute search "
+                                "context.%s", es);
+                NVolSetErrors(vol);
         }
         if (ctx)
                 ntfs_attr_put_search_ctx(ctx);
diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c
index 453d0d51ea4b..6c16db9e1a8a 100644
--- a/fs/ntfs/super.c
+++ b/fs/ntfs/super.c
@@ -1447,7 +1447,7 @@ not_enabled:
         if (unlikely(i_size_read(tmp_ino) < sizeof(USN_HEADER))) {
                 ntfs_error(vol->sb, "Found corrupt $UsnJrnl/$DATA/$Max "
                                 "attribute (size is 0x%llx but should be at "
-                                "least 0x%x bytes).", i_size_read(tmp_ino),
+                                "least 0x%zx bytes).", i_size_read(tmp_ino),
                                 sizeof(USN_HEADER));
                 return FALSE;
         }
diff --git a/fs/open.c b/fs/open.c
index f0d90cf0495c..8d06ec911fd9 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -739,7 +739,8 @@ asmlinkage long sys_fchown(unsigned int fd, uid_t user, gid_t group)
 }
 
 static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
-                                        int flags, struct file *f)
+                                        int flags, struct file *f,
+                                        int (*open)(struct inode *, struct file *))
 {
         struct inode *inode;
         int error;
@@ -761,11 +762,14 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
         f->f_op = fops_get(inode->i_fop);
         file_move(f, &inode->i_sb->s_files);
 
-        if (f->f_op && f->f_op->open) {
-                error = f->f_op->open(inode,f);
+        if (!open && f->f_op)
+                open = f->f_op->open;
+        if (open) {
+                error = open(inode, f);
                 if (error)
                         goto cleanup_all;
         }
+
         f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
 
         file_ra_state_init(&f->f_ra, f->f_mapping->host->i_mapping);
@@ -814,28 +818,75 @@ struct file *filp_open(const char * filename, int flags, int mode)
 {
         int namei_flags, error;
         struct nameidata nd;
-        struct file *f;
 
         namei_flags = flags;
         if ((namei_flags+1) & O_ACCMODE)
                 namei_flags++;
-        if (namei_flags & O_TRUNC)
-                namei_flags |= 2;
-
-        error = -ENFILE;
-        f = get_empty_filp();
-        if (f == NULL)
-                return ERR_PTR(error);
 
         error = open_namei(filename, namei_flags, mode, &nd);
         if (!error)
-                return __dentry_open(nd.dentry, nd.mnt, flags, f);
+                return nameidata_to_filp(&nd, flags);
 
-        put_filp(f);
         return ERR_PTR(error);
 }
 EXPORT_SYMBOL(filp_open);
 
+/**
+ * lookup_instantiate_filp - instantiates the open intent filp
+ * @nd: pointer to nameidata
+ * @dentry: pointer to dentry
+ * @open: open callback
+ *
+ * Helper for filesystems that want to use lookup open intents and pass back
+ * a fully instantiated struct file to the caller.
+ * This function is meant to be called from within a filesystem's
+ * lookup method.
+ * Note that in case of error, nd->intent.open.file is destroyed, but the
+ * path information remains valid.
+ * If the open callback is set to NULL, then the standard f_op->open()
+ * filesystem callback is substituted.
+ */
+struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry,
+                int (*open)(struct inode *, struct file *))
+{
+        if (IS_ERR(nd->intent.open.file))
+                goto out;
+        if (IS_ERR(dentry))
+                goto out_err;
+        nd->intent.open.file = __dentry_open(dget(dentry), mntget(nd->mnt),
+                                             nd->intent.open.flags - 1,
+                                             nd->intent.open.file,
+                                             open);
+out:
+        return nd->intent.open.file;
+out_err:
+        release_open_intent(nd);
+        nd->intent.open.file = (struct file *)dentry;
+        goto out;
+}
+EXPORT_SYMBOL_GPL(lookup_instantiate_filp);
+
+/**
+ * nameidata_to_filp - convert a nameidata to an open filp.
+ * @nd: pointer to nameidata
+ * @flags: open flags
+ *
+ * Note that this function destroys the original nameidata
+ */
+struct file *nameidata_to_filp(struct nameidata *nd, int flags)
+{
+        struct file *filp;
+
+        /* Pick up the filp from the open intent */
+        filp = nd->intent.open.file;
+        /* Has the filesystem initialised the file for us? */
+        if (filp->f_dentry == NULL)
+                filp = __dentry_open(nd->dentry, nd->mnt, flags, filp, NULL);
+        else
+                path_release(nd);
+        return filp;
+}
+
 struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags)
 {
         int error;
@@ -846,7 +897,7 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags)
         if (f == NULL)
                 return ERR_PTR(error);
 
-        return __dentry_open(dentry, mnt, flags, f);
+        return __dentry_open(dentry, mnt, flags, f, NULL);
 }
 EXPORT_SYMBOL(dentry_open);
 
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 77e178f13162..9c06c5434ec4 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -192,6 +192,7 @@ check_partition(struct gendisk *hd, struct block_device *bdev)
 struct part_attribute {
         struct attribute attr;
         ssize_t (*show)(struct hd_struct *,char *);
+        ssize_t (*store)(struct hd_struct *,const char *, size_t);
 };
 
 static ssize_t 
@@ -201,14 +202,33 @@ part_attr_show(struct kobject * kobj, struct attribute * attr, char * page)
         struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr);
         ssize_t ret = 0;
         if (part_attr->show)
-                ret = part_attr->show(p,page);
+                ret = part_attr->show(p, page);
+        return ret;
+}
+static ssize_t
+part_attr_store(struct kobject * kobj, struct attribute * attr,
+                const char *page, size_t count)
+{
+        struct hd_struct * p = container_of(kobj,struct hd_struct,kobj);
+        struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr);
+        ssize_t ret = 0;
+
+        if (part_attr->store)
+                ret = part_attr->store(p, page, count);
         return ret;
 }
 
 static struct sysfs_ops part_sysfs_ops = {
         .show   =       part_attr_show,
+        .store  =       part_attr_store,
 };
 
+static ssize_t part_uevent_store(struct hd_struct * p,
+                                 const char *page, size_t count)
+{
+        kobject_hotplug(&p->kobj, KOBJ_ADD);
+        return count;
+}
 static ssize_t part_dev_read(struct hd_struct * p, char *page)
 {
         struct gendisk *disk = container_of(p->kobj.parent,struct gendisk,kobj);
@@ -229,6 +249,10 @@ static ssize_t part_stat_read(struct hd_struct * p, char *page)
                        p->reads, (unsigned long long)p->read_sectors,
                        p->writes, (unsigned long long)p->write_sectors);
 }
+static struct part_attribute part_attr_uevent = {
+        .attr = {.name = "uevent", .mode = S_IWUSR },
+        .store  = part_uevent_store
+};
 static struct part_attribute part_attr_dev = {
         .attr = {.name = "dev", .mode = S_IRUGO },
         .show   = part_dev_read
@@ -247,6 +271,7 @@ static struct part_attribute part_attr_stat = {
 };
 
 static struct attribute * default_attrs[] = {
+        &part_attr_uevent.attr,
         &part_attr_dev.attr,
         &part_attr_start.attr,
         &part_attr_size.attr,
@@ -430,7 +455,7 @@ void del_gendisk(struct gendisk *disk)
         disk->flags &= ~GENHD_FL_UP;
         unlink_gendisk(disk);
         disk_stat_set_all(disk, 0);
-        disk->stamp = disk->stamp_idle = 0;
+        disk->stamp = 0;
 
         devfs_remove_disk(disk);
 
diff --git a/fs/proc/array.c b/fs/proc/array.c
index d84eecacbeaf..3e1239e4b303 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -438,7 +438,7 @@ static int do_task_stat(struct task_struct *task, char * buffer, int whole)
                 jiffies_to_clock_t(it_real_value),
                 start_time,
                 vsize,
-                mm ? get_mm_counter(mm, rss) : 0, /* you might want to shift this left 3 */
+                mm ? get_mm_rss(mm) : 0,
                 rsslim,
                 mm ? mm->start_code : 0,
                 mm ? mm->end_code : 0,
diff --git a/fs/proc/generic.c b/fs/proc/generic.c
index 8a8c34461d48..b638fb500743 100644
--- a/fs/proc/generic.c
+++ b/fs/proc/generic.c
@@ -533,7 +533,7 @@ static void proc_kill_inodes(struct proc_dir_entry *de)
          */
         file_list_lock();
         list_for_each(p, &sb->s_files) {
-                struct file * filp = list_entry(p, struct file, f_list);
+                struct file * filp = list_entry(p, struct file, f_u.fu_list);
                 struct dentry * dentry = filp->f_dentry;
                 struct inode * inode;
                 struct file_operations *fops;
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index effa6c0c467a..e6a818a93f3d 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -156,10 +156,13 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino,
 
         WARN_ON(de && de->deleted);
 
+        if (de != NULL && !try_module_get(de->owner))
+                goto out_mod;
+
         inode = iget(sb, ino);
         if (!inode)
-                goto out_fail;
-        
+                goto out_ino;
+
         PROC_I(inode)->pde = de;
         if (de) {
                 if (de->mode) {
@@ -171,20 +174,20 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino,
                         inode->i_size = de->size;
                 if (de->nlink)
                         inode->i_nlink = de->nlink;
-                if (!try_module_get(de->owner))
-                        goto out_fail;
                 if (de->proc_iops)
                         inode->i_op = de->proc_iops;
                 if (de->proc_fops)
                         inode->i_fop = de->proc_fops;
         }
 
-out:
         return inode;
 
-out_fail:
+out_ino:
+        if (de != NULL)
+                module_put(de->owner);
+out_mod:
         de_put(de);
-        goto out;
+        return NULL;
 }                       
 
 int proc_fill_super(struct super_block *s, void *data, int silent)
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index a3453555a94e..5b6b0b6038a7 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -629,12 +629,4 @@ void __init proc_misc_init(void)
         if (entry)
                 entry->proc_fops = &proc_sysrq_trigger_operations;
 #endif
-#ifdef CONFIG_PPC32
-        {
-                extern struct file_operations ppc_htab_operations;
-                entry = create_proc_entry("ppc_htab", S_IRUGO|S_IWUSR, NULL);
-                if (entry)
-                        entry->proc_fops = &ppc_htab_operations;
-        }
-#endif
 }
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index c7ef3e48e35b..d2fa42006d8f 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -14,22 +14,41 @@
 char *task_mem(struct mm_struct *mm, char *buffer)
 {
         unsigned long data, text, lib;
+        unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss;
+
+        /*
+         * Note: to minimize their overhead, mm maintains hiwater_vm and
+         * hiwater_rss only when about to *lower* total_vm or rss.  Any
+         * collector of these hiwater stats must therefore get total_vm
+         * and rss too, which will usually be the higher.  Barriers? not
+         * worth the effort, such snapshots can always be inconsistent.
+         */
+        hiwater_vm = total_vm = mm->total_vm;
+        if (hiwater_vm < mm->hiwater_vm)
+                hiwater_vm = mm->hiwater_vm;
+        hiwater_rss = total_rss = get_mm_rss(mm);
+        if (hiwater_rss < mm->hiwater_rss)
+                hiwater_rss = mm->hiwater_rss;
 
         data = mm->total_vm - mm->shared_vm - mm->stack_vm;
         text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10;
         lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text;
         buffer += sprintf(buffer,
+                "VmPeak:\t%8lu kB\n"
                 "VmSize:\t%8lu kB\n"
                 "VmLck:\t%8lu kB\n"
+                "VmHWM:\t%8lu kB\n"
                 "VmRSS:\t%8lu kB\n"
                 "VmData:\t%8lu kB\n"
                 "VmStk:\t%8lu kB\n"
                 "VmExe:\t%8lu kB\n"
                 "VmLib:\t%8lu kB\n"
                 "VmPTE:\t%8lu kB\n",
-                (mm->total_vm - mm->reserved_vm) << (PAGE_SHIFT-10),
+                hiwater_vm << (PAGE_SHIFT-10),
+                (total_vm - mm->reserved_vm) << (PAGE_SHIFT-10),
                 mm->locked_vm << (PAGE_SHIFT-10),
-                get_mm_counter(mm, rss) << (PAGE_SHIFT-10),
+                hiwater_rss << (PAGE_SHIFT-10),
+                total_rss << (PAGE_SHIFT-10),
                 data << (PAGE_SHIFT-10),
                 mm->stack_vm << (PAGE_SHIFT-10), text, lib,
                 (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10);
@@ -44,13 +63,11 @@ unsigned long task_vsize(struct mm_struct *mm)
 int task_statm(struct mm_struct *mm, int *shared, int *text,
                int *data, int *resident)
 {
-        int rss = get_mm_counter(mm, rss);
-
-        *shared = rss - get_mm_counter(mm, anon_rss);
+        *shared = get_mm_counter(mm, file_rss);
         *text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK))
                                                                 >> PAGE_SHIFT;
         *data = mm->total_vm - mm->shared_vm;
-        *resident = rss;
+        *resident = *shared + get_mm_counter(mm, anon_rss);
         return mm->total_vm;
 }
 
@@ -186,13 +203,14 @@ static void smaps_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
                                 struct mem_size_stats *mss)
 {
         pte_t *pte, ptent;
+        spinlock_t *ptl;
         unsigned long pfn;
         struct page *page;
 
-        pte = pte_offset_map(pmd, addr);
+        pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
         do {
                 ptent = *pte;
-                if (pte_none(ptent) || !pte_present(ptent))
+                if (!pte_present(ptent))
                         continue;
 
                 mss->resident += PAGE_SIZE;
@@ -213,8 +231,8 @@ static void smaps_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
                                 mss->private_clean += PAGE_SIZE;
                 }
         } while (pte++, addr += PAGE_SIZE, addr != end);
-        pte_unmap(pte - 1);
-        cond_resched_lock(&vma->vm_mm->page_table_lock);
+        pte_unmap_unlock(pte - 1, ptl);
+        cond_resched();
 }
 
 static inline void smaps_pmd_range(struct vm_area_struct *vma, pud_t *pud,
@@ -268,17 +286,11 @@ static inline void smaps_pgd_range(struct vm_area_struct *vma,
 static int show_smap(struct seq_file *m, void *v)
 {
         struct vm_area_struct *vma = v;
-        struct mm_struct *mm = vma->vm_mm;
         struct mem_size_stats mss;
 
         memset(&mss, 0, sizeof mss);
-
-        if (mm) {
-                spin_lock(&mm->page_table_lock);
+        if (vma->vm_mm)
                 smaps_pgd_range(vma, vma->vm_start, vma->vm_end, &mss);
-                spin_unlock(&mm->page_table_lock);
-        }
-
         return show_map_internal(m, v, &mss);
 }
 
@@ -407,7 +419,6 @@ static struct numa_maps *get_numa_maps(const struct vm_area_struct *vma)
         for_each_node(i)
                 md->node[i] =0;
 
-        spin_lock(&mm->page_table_lock);
         for (vaddr = vma->vm_start; vaddr < vma->vm_end; vaddr += PAGE_SIZE) {
                 page = follow_page(mm, vaddr, 0);
                 if (page) {
@@ -422,8 +433,8 @@ static struct numa_maps *get_numa_maps(const struct vm_area_struct *vma)
                                 md->anon++;
                         md->node[page_to_nid(page)]++;
                 }
+                cond_resched();
         }
-        spin_unlock(&mm->page_table_lock);
         return md;
 }
 
@@ -469,7 +480,7 @@ static int show_numa_map(struct seq_file *m, void *v)
                 seq_printf(m, " interleave={");
                 first = 1;
                 for_each_node(n) {
-                        if (test_bit(n, pol->v.nodes)) {
+                        if (node_isset(n, pol->v.nodes)) {
                                 if (!first)
                                         seq_putc(m,',');
                                 else
diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c
index 2706e2adffab..45829889dcdc 100644
--- a/fs/reiserfs/fix_node.c
+++ b/fs/reiserfs/fix_node.c
@@ -2022,7 +2022,7 @@ static int get_neighbors(struct tree_balance *p_s_tb, int n_h)
 }
 
 #ifdef CONFIG_REISERFS_CHECK
-void *reiserfs_kmalloc(size_t size, int flags, struct super_block *s)
+void *reiserfs_kmalloc(size_t size, gfp_t flags, struct super_block *s)
 {
         void *vp;
         static size_t malloced;
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index d76ee6c4f9b8..5f82352b97e1 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -2842,7 +2842,7 @@ static int reiserfs_set_page_dirty(struct page *page)
  * even in -o notail mode, we can't be sure an old mount without -o notail
  * didn't create files with tails.
  */
-static int reiserfs_releasepage(struct page *page, int unused_gfp_flags)
+static int reiserfs_releasepage(struct page *page, gfp_t unused_gfp_flags)
 {
         struct inode *inode = page->mapping->host;
         struct reiserfs_journal *j = SB_JOURNAL(inode->i_sb);
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 44b02fc02ebe..42afb5bef111 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -1024,12 +1024,8 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
                                 strcpy(REISERFS_SB(s)->s_qf_names[qtype], arg);
                                 *mount_options |= 1 << REISERFS_QUOTA;
                         } else {
-                                if (REISERFS_SB(s)->s_qf_names[qtype]) {
-                                        kfree(REISERFS_SB(s)->
-                                              s_qf_names[qtype]);
-                                        REISERFS_SB(s)->s_qf_names[qtype] =
-                                            NULL;
-                                }
+                                kfree(REISERFS_SB(s)->s_qf_names[qtype]);
+                                REISERFS_SB(s)->s_qf_names[qtype] = NULL;
                         }
                 }
                 if (c == 'f') {
@@ -1158,11 +1154,10 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
         if (!reiserfs_parse_options
             (s, arg, &mount_options, &blocks, NULL, &commit_max_age)) {
 #ifdef CONFIG_QUOTA
-                for (i = 0; i < MAXQUOTAS; i++)
-                        if (REISERFS_SB(s)->s_qf_names[i]) {
-                                kfree(REISERFS_SB(s)->s_qf_names[i]);
-                                REISERFS_SB(s)->s_qf_names[i] = NULL;
-                        }
+                for (i = 0; i < MAXQUOTAS; i++) {
+                        kfree(REISERFS_SB(s)->s_qf_names[i]);
+                        REISERFS_SB(s)->s_qf_names[i] = NULL;
+                }
 #endif
                 return -EINVAL;
         }
@@ -1940,13 +1935,11 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
                 brelse(SB_BUFFER_WITH_SB(s));
 #ifdef CONFIG_QUOTA
         for (j = 0; j < MAXQUOTAS; j++) {
-                if (sbi->s_qf_names[j])
-                        kfree(sbi->s_qf_names[j]);
+                kfree(sbi->s_qf_names[j]);
+                sbi->s_qf_names[j] = NULL;
         }
 #endif
-        if (sbi != NULL) {
-                kfree(sbi);
-        }
+        kfree(sbi);
 
         s->s_fs_info = NULL;
         return errval;
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 87ac9dc8b381..72e120798677 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -453,7 +453,7 @@ static struct page *reiserfs_get_page(struct inode *dir, unsigned long n)
         struct page *page;
         /* We can deadlock if we try to free dentries,
            and an unlink/rmdir has just occured - GFP_NOFS avoids this */
-        mapping->flags = (mapping->flags & ~__GFP_BITS_MASK) | GFP_NOFS;
+        mapping_set_gfp_mask(mapping, GFP_NOFS);
         page = read_cache_page(mapping, n,
                                (filler_t *) mapping->a_ops->readpage, NULL);
         if (!IS_ERR(page)) {
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index 6703efa3c430..a47ac9aac8b2 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -296,8 +296,7 @@ reiserfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
                 }
         }
 
-        if (value)
-                kfree(value);
+        kfree(value);
 
         if (!error) {
                 /* Release the old one */
diff --git a/fs/super.c b/fs/super.c
index 6e57ee252e14..f60155ec7780 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -513,7 +513,7 @@ static void mark_files_ro(struct super_block *sb)
         struct file *f;
 
         file_list_lock();
-        list_for_each_entry(f, &sb->s_files, f_list) {
+        list_for_each_entry(f, &sb->s_files, f_u.fu_list) {
                 if (S_ISREG(f->f_dentry->d_inode->i_mode) && file_count(f))
                         f->f_mode &= ~FMODE_WRITE;
         }
diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c
index 1c6f6b57ef1c..ef46939c0c1a 100644
--- a/fs/vfat/namei.c
+++ b/fs/vfat/namei.c
@@ -621,8 +621,7 @@ static int vfat_build_slots(struct inode *dir, const unsigned char *name,
         }
 
         /* build the entry of long file name */
-        for (cksum = i = 0; i < 11; i++)
-                cksum = (((cksum&1)<<7)|((cksum&0xfe)>>1)) + msdos_name[i];
+        cksum = fat_checksum(msdos_name);
 
         *nr_slots = usize / 13;
         for (ps = slots, i = *nr_slots; i > 0; i--, ps++) {
@@ -888,10 +887,10 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry,
 {
         struct buffer_head *dotdot_bh;
         struct msdos_dir_entry *dotdot_de;
-        loff_t dotdot_i_pos;
         struct inode *old_inode, *new_inode;
         struct fat_slot_info old_sinfo, sinfo;
         struct timespec ts;
+        loff_t dotdot_i_pos, new_i_pos;
         int err, is_dir, update_dotdot, corrupt = 0;
 
         old_sinfo.bh = sinfo.bh = dotdot_bh = NULL;
@@ -914,31 +913,24 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry,
 
         ts = CURRENT_TIME_SEC;
         if (new_inode) {
-                err = vfat_find(new_dir, &new_dentry->d_name, &sinfo);
-                if (err)
-                        goto out;
-                if (MSDOS_I(new_inode)->i_pos != sinfo.i_pos) {
-                        /* WTF??? Cry and fail. */
-                        printk(KERN_WARNING "vfat_rename: fs corrupted\n");
-                        goto out;
-                }
-
                 if (is_dir) {
                         err = fat_dir_empty(new_inode);
                         if (err)
                                 goto out;
                 }
+                new_i_pos = MSDOS_I(new_inode)->i_pos;
                 fat_detach(new_inode);
         } else {
                 err = vfat_add_entry(new_dir, &new_dentry->d_name, is_dir, 0,
                                      &ts, &sinfo);
                 if (err)
                         goto out;
+                new_i_pos = sinfo.i_pos;
         }
         new_dir->i_version++;
 
         fat_detach(old_inode);
-        fat_attach(old_inode, sinfo.i_pos);
+        fat_attach(old_inode, new_i_pos);
         if (IS_DIRSYNC(new_dir)) {
                 err = fat_sync_inode(old_inode);
                 if (err)
@@ -1002,7 +994,7 @@ error_inode:
         fat_detach(old_inode);
         fat_attach(old_inode, old_sinfo.i_pos);
         if (new_inode) {
-                fat_attach(new_inode, sinfo.i_pos);
+                fat_attach(new_inode, new_i_pos);
                 if (corrupt)
                         corrupt |= fat_sync_inode(new_inode);
         } else {
diff --git a/fs/xattr.c b/fs/xattr.c
index 3f9c64bea151..f6e00c0e114f 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -143,7 +143,7 @@ getxattr(struct dentry *d, char __user *name, void __user *value, size_t size)
         if (size) {
                 if (size > XATTR_SIZE_MAX)
                         size = XATTR_SIZE_MAX;
-                kvalue = kmalloc(size, GFP_KERNEL);
+                kvalue = kzalloc(size, GFP_KERNEL);
                 if (!kvalue)
                         return -ENOMEM;
         }
@@ -154,11 +154,15 @@ getxattr(struct dentry *d, char __user *name, void __user *value, size_t size)
         error = -EOPNOTSUPP;
         if (d->d_inode->i_op && d->d_inode->i_op->getxattr)
                 error = d->d_inode->i_op->getxattr(d, kname, kvalue, size);
-        else if (!strncmp(kname, XATTR_SECURITY_PREFIX,
-                          sizeof XATTR_SECURITY_PREFIX - 1)) {
+
+        if (!strncmp(kname, XATTR_SECURITY_PREFIX,
+                     sizeof XATTR_SECURITY_PREFIX - 1)) {
                 const char *suffix = kname + sizeof XATTR_SECURITY_PREFIX - 1;
-                error = security_inode_getsecurity(d->d_inode, suffix, kvalue,
-                                                   size);
+                int rv = security_inode_getsecurity(d->d_inode, suffix, kvalue,
+                                                    size, error);
+                /* Security module active: overwrite error value */
+                if (rv != -EOPNOTSUPP)
+                        error = rv;
         }
         if (error > 0) {
                 if (size && copy_to_user(value, kvalue, error))
diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c
index d2653b589b1c..3c92162dc728 100644
--- a/fs/xfs/linux-2.6/kmem.c
+++ b/fs/xfs/linux-2.6/kmem.c
@@ -45,11 +45,11 @@
 
 
 void *
-kmem_alloc(size_t size, gfp_t flags)
+kmem_alloc(size_t size, unsigned int __nocast flags)
 {
-        int             retries = 0;
-        unsigned int    lflags = kmem_flags_convert(flags);
-        void            *ptr;
+        int     retries = 0;
+        gfp_t   lflags = kmem_flags_convert(flags);
+        void    *ptr;
 
         do {
                 if (size < MAX_SLAB_SIZE || retries > MAX_VMALLOCS)
@@ -67,7 +67,7 @@ kmem_alloc(size_t size, gfp_t flags)
 }
 
 void *
-kmem_zalloc(size_t size, gfp_t flags)
+kmem_zalloc(size_t size, unsigned int __nocast flags)
 {
         void    *ptr;
 
@@ -90,7 +90,7 @@ kmem_free(void *ptr, size_t size)
 
 void *
 kmem_realloc(void *ptr, size_t newsize, size_t oldsize,
-             gfp_t flags)
+             unsigned int __nocast flags)
 {
         void    *new;
 
@@ -105,11 +105,11 @@ kmem_realloc(void *ptr, size_t newsize, size_t oldsize,
 }
 
 void *
-kmem_zone_alloc(kmem_zone_t *zone, gfp_t flags)
+kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags)
 {
-        int             retries = 0;
-        unsigned int    lflags = kmem_flags_convert(flags);
-        void            *ptr;
+        int     retries = 0;
+        gfp_t   lflags = kmem_flags_convert(flags);
+        void    *ptr;
 
         do {
                 ptr = kmem_cache_alloc(zone, lflags);
@@ -124,7 +124,7 @@ kmem_zone_alloc(kmem_zone_t *zone, gfp_t flags)
 }
 
 void *
-kmem_zone_zalloc(kmem_zone_t *zone, gfp_t flags)
+kmem_zone_zalloc(kmem_zone_t *zone, unsigned int __nocast flags)
 {
         void    *ptr;
 
diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h
index ee7010f085bc..f4bb78c268c0 100644
--- a/fs/xfs/linux-2.6/kmem.h
+++ b/fs/xfs/linux-2.6/kmem.h
@@ -81,9 +81,9 @@ typedef unsigned long xfs_pflags_t;
         *(NSTATEP) = *(OSTATEP);        \
 } while (0)
 
-static __inline unsigned int kmem_flags_convert(gfp_t flags)
+static __inline gfp_t kmem_flags_convert(unsigned int __nocast flags)
 {
-        unsigned int    lflags = __GFP_NOWARN;  /* we'll report problems, if need be */
+        gfp_t lflags = __GFP_NOWARN;    /* we'll report problems, if need be */
 
 #ifdef DEBUG
         if (unlikely(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL))) {
@@ -125,16 +125,16 @@ kmem_zone_destroy(kmem_zone_t *zone)
                 BUG();
 }
 
-extern void         *kmem_zone_zalloc(kmem_zone_t *, gfp_t);
-extern void         *kmem_zone_alloc(kmem_zone_t *, gfp_t);
+extern void         *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast);
+extern void         *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast);
 
-extern void         *kmem_alloc(size_t, gfp_t);
-extern void         *kmem_realloc(void *, size_t, size_t, gfp_t);
-extern void         *kmem_zalloc(size_t, gfp_t);
+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_free(void *, size_t);
 
 typedef struct shrinker *kmem_shaker_t;
-typedef int (*kmem_shake_func_t)(int, unsigned int);
+typedef int (*kmem_shake_func_t)(int, gfp_t);
 
 static __inline kmem_shaker_t
 kmem_shake_register(kmem_shake_func_t sfunc)
@@ -149,7 +149,7 @@ kmem_shake_deregister(kmem_shaker_t shrinker)
 }
 
 static __inline int
-kmem_shake_allow(unsigned int gfp_mask)
+kmem_shake_allow(gfp_t gfp_mask)
 {
         return (gfp_mask & __GFP_WAIT);
 }
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index c6c077978fe3..7aa398724706 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -1296,7 +1296,7 @@ linvfs_invalidate_page(
 STATIC int
 linvfs_release_page(
         struct page             *page,
-        int                     gfp_mask)
+        gfp_t                   gfp_mask)
 {
         struct inode            *inode = page->mapping->host;
         int                     dirty, delalloc, unmapped, unwritten;
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index e82cf72ac599..4cd46abe8434 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -64,7 +64,7 @@
 
 STATIC kmem_cache_t *pagebuf_zone;
 STATIC kmem_shaker_t pagebuf_shake;
-STATIC int xfsbufd_wakeup(int, unsigned int);
+STATIC int xfsbufd_wakeup(int, gfp_t);
 STATIC void pagebuf_delwri_queue(xfs_buf_t *, int);
 
 STATIC struct workqueue_struct *xfslogd_workqueue;
@@ -181,8 +181,9 @@ set_page_region(
         size_t          offset,
         size_t          length)
 {
-        page->private |= page_region_mask(offset, length);
-        if (page->private == ~0UL)
+        set_page_private(page,
+                page_private(page) | page_region_mask(offset, length));
+        if (page_private(page) == ~0UL)
                 SetPageUptodate(page);
 }
 
@@ -194,7 +195,7 @@ test_page_region(
 {
         unsigned long   mask = page_region_mask(offset, length);
 
-        return (mask && (page->private & mask) == mask);
+        return (mask && (page_private(page) & mask) == mask);
 }
 
 /*
@@ -383,7 +384,7 @@ _pagebuf_lookup_pages(
         size_t                  blocksize = bp->pb_target->pbr_bsize;
         size_t                  size = bp->pb_count_desired;
         size_t                  nbytes, offset;
-        int                     gfp_mask = pb_to_gfp(flags);
+        gfp_t                   gfp_mask = pb_to_gfp(flags);
         unsigned short          page_count, i;
         pgoff_t                 first;
         loff_t                  end;
@@ -1749,8 +1750,8 @@ STATIC int xfsbufd_force_sleep;
 
 STATIC int
 xfsbufd_wakeup(
-        int                     priority,
-        unsigned int            mask)
+        int             priority,
+        gfp_t           mask)
 {
         if (xfsbufd_force_sleep)
                 return 0;