Merge branch 'mpi-master' into wip-k-fmlpwip-k-fmlp

Conflicts:
	litmus/sched_cedf.c

31 files changed, 2257 insertions, 1404 deletions

diff --git a/fs/ubifs/Kconfig b/fs/ubifs/Kconfig
index 830e3f76f442..f8b0160da2da 100644
--- a/fs/ubifs/Kconfig
+++ b/fs/ubifs/Kconfig
@@ -44,29 +44,17 @@ config UBIFS_FS_ZLIB
 
 # Debugging-related stuff
 config UBIFS_FS_DEBUG
-        bool "Enable debugging"
+        bool "Enable debugging support"
         depends on UBIFS_FS
         select DEBUG_FS
-        select KALLSYMS_ALL
-        help
-          This option enables UBIFS debugging.
-
-config UBIFS_FS_DEBUG_MSG_LVL
-        int "Default message level (0 = no extra messages, 3 = lots)"
-        depends on UBIFS_FS_DEBUG
-        default "0"
-        help
-          This controls the amount of debugging messages produced by UBIFS.
-          If reporting bugs, please try to have available a full dump of the
-          messages at level 1 while the misbehaviour was occurring. Level 2
-          may become necessary if level 1 messages were not enough to find the
-          bug. Generally Level 3 should be avoided.
-
-config UBIFS_FS_DEBUG_CHKS
-        bool "Enable extra checks"
-        depends on UBIFS_FS_DEBUG
-        help
-          If extra checks are enabled UBIFS will check the consistency of its
-          internal data structures during operation. However, UBIFS performance
-          is dramatically slower when this option is selected especially if the
-          file system is large.
+        select KALLSYMS
+        help
+          This option enables UBIFS debugging support. It makes sure various
+          assertions, self-checks, debugging messages and test modes are compiled
+          in (this all is compiled out otherwise). Assertions are light-weight
+          and this option also enables them. Self-checks, debugging messages and
+          test modes are switched off by default. Thus, it is safe and actually
+          recommended to have debugging support enabled, and it should not slow
+          down UBIFS. You can then further enable / disable individual  debugging
+          features using UBIFS module parameters and the corresponding sysfs
+          interfaces.
diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c
index c8ff0d1ae5d3..315de66e52b2 100644
--- a/fs/ubifs/budget.c
+++ b/fs/ubifs/budget.c
@@ -106,7 +106,7 @@ static long long get_liability(struct ubifs_info *c)
         long long liab;
 
         spin_lock(&c->space_lock);
-        liab = c->budg_idx_growth + c->budg_data_growth + c->budg_dd_growth;
+        liab = c->bi.idx_growth + c->bi.data_growth + c->bi.dd_growth;
         spin_unlock(&c->space_lock);
         return liab;
 }
@@ -147,7 +147,7 @@ static int make_free_space(struct ubifs_info *c)
                 if (liab2 < liab1)
                         return -EAGAIN;
 
-                dbg_budg("new liability %lld (not shrinked)", liab2);
+                dbg_budg("new liability %lld (not shrunk)", liab2);
 
                 /* Liability did not shrink again, try GC */
                 dbg_budg("Run GC");
@@ -180,7 +180,7 @@ int ubifs_calc_min_idx_lebs(struct ubifs_info *c)
         int idx_lebs;
         long long idx_size;
 
-        idx_size = c->old_idx_sz + c->budg_idx_growth + c->budg_uncommitted_idx;
+        idx_size = c->bi.old_idx_sz + c->bi.idx_growth + c->bi.uncommitted_idx;
         /* And make sure we have thrice the index size of space reserved */
         idx_size += idx_size << 1;
         /*
@@ -292,13 +292,13 @@ static int can_use_rp(struct ubifs_info *c)
  * budgeted index space to the size of the current index, multiplies this by 3,
  * and makes sure this does not exceed the amount of free LEBs.
  *
- * Notes about @c->min_idx_lebs and @c->lst.idx_lebs variables:
+ * Notes about @c->bi.min_idx_lebs and @c->lst.idx_lebs variables:
  * o @c->lst.idx_lebs is the number of LEBs the index currently uses. It might
  *    be large, because UBIFS does not do any index consolidation as long as
  *    there is free space. IOW, the index may take a lot of LEBs, but the LEBs
  *    will contain a lot of dirt.
- * o @c->min_idx_lebs is the number of LEBS the index presumably takes. IOW,
- *    the index may be consolidated to take up to @c->min_idx_lebs LEBs.
+ * o @c->bi.min_idx_lebs is the number of LEBS the index presumably takes. IOW,
+ *    the index may be consolidated to take up to @c->bi.min_idx_lebs LEBs.
  *
  * This function returns zero in case of success, and %-ENOSPC in case of
  * failure.
@@ -343,13 +343,13 @@ static int do_budget_space(struct ubifs_info *c)
                c->lst.taken_empty_lebs;
         if (unlikely(rsvd_idx_lebs > lebs)) {
                 dbg_budg("out of indexing space: min_idx_lebs %d (old %d), "
-                         "rsvd_idx_lebs %d", min_idx_lebs, c->min_idx_lebs,
+                         "rsvd_idx_lebs %d", min_idx_lebs, c->bi.min_idx_lebs,
                          rsvd_idx_lebs);
                 return -ENOSPC;
         }
 
         available = ubifs_calc_available(c, min_idx_lebs);
-        outstanding = c->budg_data_growth + c->budg_dd_growth;
+        outstanding = c->bi.data_growth + c->bi.dd_growth;
 
         if (unlikely(available < outstanding)) {
                 dbg_budg("out of data space: available %lld, outstanding %lld",
@@ -360,7 +360,7 @@ static int do_budget_space(struct ubifs_info *c)
         if (available - outstanding <= c->rp_size && !can_use_rp(c))
                 return -ENOSPC;
 
-        c->min_idx_lebs = min_idx_lebs;
+        c->bi.min_idx_lebs = min_idx_lebs;
         return 0;
 }
 
@@ -393,11 +393,11 @@ static int calc_data_growth(const struct ubifs_info *c,
 {
         int data_growth;
 
-        data_growth = req->new_ino  ? c->inode_budget : 0;
+        data_growth = req->new_ino  ? c->bi.inode_budget : 0;
         if (req->new_page)
-                data_growth += c->page_budget;
+                data_growth += c->bi.page_budget;
         if (req->new_dent)
-                data_growth += c->dent_budget;
+                data_growth += c->bi.dent_budget;
         data_growth += req->new_ino_d;
         return data_growth;
 }
@@ -413,12 +413,12 @@ static int calc_dd_growth(const struct ubifs_info *c,
 {
         int dd_growth;
 
-        dd_growth = req->dirtied_page ? c->page_budget : 0;
+        dd_growth = req->dirtied_page ? c->bi.page_budget : 0;
 
         if (req->dirtied_ino)
-                dd_growth += c->inode_budget << (req->dirtied_ino - 1);
+                dd_growth += c->bi.inode_budget << (req->dirtied_ino - 1);
         if (req->mod_dent)
-                dd_growth += c->dent_budget;
+                dd_growth += c->bi.dent_budget;
         dd_growth += req->dirtied_ino_d;
         return dd_growth;
 }
@@ -460,19 +460,19 @@ int ubifs_budget_space(struct ubifs_info *c, struct ubifs_budget_req *req)
 
 again:
         spin_lock(&c->space_lock);
-        ubifs_assert(c->budg_idx_growth >= 0);
-        ubifs_assert(c->budg_data_growth >= 0);
-        ubifs_assert(c->budg_dd_growth >= 0);
+        ubifs_assert(c->bi.idx_growth >= 0);
+        ubifs_assert(c->bi.data_growth >= 0);
+        ubifs_assert(c->bi.dd_growth >= 0);
 
-        if (unlikely(c->nospace) && (c->nospace_rp || !can_use_rp(c))) {
+        if (unlikely(c->bi.nospace) && (c->bi.nospace_rp || !can_use_rp(c))) {
                 dbg_budg("no space");
                 spin_unlock(&c->space_lock);
                 return -ENOSPC;
         }
 
-        c->budg_idx_growth += idx_growth;
-        c->budg_data_growth += data_growth;
-        c->budg_dd_growth += dd_growth;
+        c->bi.idx_growth += idx_growth;
+        c->bi.data_growth += data_growth;
+        c->bi.dd_growth += dd_growth;
 
         err = do_budget_space(c);
         if (likely(!err)) {
@@ -484,9 +484,9 @@ again:
         }
 
         /* Restore the old values */
-        c->budg_idx_growth -= idx_growth;
-        c->budg_data_growth -= data_growth;
-        c->budg_dd_growth -= dd_growth;
+        c->bi.idx_growth -= idx_growth;
+        c->bi.data_growth -= data_growth;
+        c->bi.dd_growth -= dd_growth;
         spin_unlock(&c->space_lock);
 
         if (req->fast) {
@@ -506,9 +506,9 @@ again:
                         goto again;
                 }
                 dbg_budg("FS is full, -ENOSPC");
-                c->nospace = 1;
+                c->bi.nospace = 1;
                 if (can_use_rp(c) || c->rp_size == 0)
-                        c->nospace_rp = 1;
+                        c->bi.nospace_rp = 1;
                 smp_wmb();
         } else
                 ubifs_err("cannot budget space, error %d", err);
@@ -523,8 +523,8 @@ again:
  * This function releases the space budgeted by 'ubifs_budget_space()'. Note,
  * since the index changes (which were budgeted for in @req->idx_growth) will
  * only be written to the media on commit, this function moves the index budget
- * from @c->budg_idx_growth to @c->budg_uncommitted_idx. The latter will be
- * zeroed by the commit operation.
+ * from @c->bi.idx_growth to @c->bi.uncommitted_idx. The latter will be zeroed
+ * by the commit operation.
  */
 void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req)
 {
@@ -553,23 +553,23 @@ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req)
         if (!req->data_growth && !req->dd_growth)
                 return;
 
-        c->nospace = c->nospace_rp = 0;
+        c->bi.nospace = c->bi.nospace_rp = 0;
         smp_wmb();
 
         spin_lock(&c->space_lock);
-        c->budg_idx_growth -= req->idx_growth;
-        c->budg_uncommitted_idx += req->idx_growth;
-        c->budg_data_growth -= req->data_growth;
-        c->budg_dd_growth -= req->dd_growth;
-        c->min_idx_lebs = ubifs_calc_min_idx_lebs(c);
-
-        ubifs_assert(c->budg_idx_growth >= 0);
-        ubifs_assert(c->budg_data_growth >= 0);
-        ubifs_assert(c->budg_dd_growth >= 0);
-        ubifs_assert(c->min_idx_lebs < c->main_lebs);
-        ubifs_assert(!(c->budg_idx_growth & 7));
-        ubifs_assert(!(c->budg_data_growth & 7));
-        ubifs_assert(!(c->budg_dd_growth & 7));
+        c->bi.idx_growth -= req->idx_growth;
+        c->bi.uncommitted_idx += req->idx_growth;
+        c->bi.data_growth -= req->data_growth;
+        c->bi.dd_growth -= req->dd_growth;
+        c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c);
+
+        ubifs_assert(c->bi.idx_growth >= 0);
+        ubifs_assert(c->bi.data_growth >= 0);
+        ubifs_assert(c->bi.dd_growth >= 0);
+        ubifs_assert(c->bi.min_idx_lebs < c->main_lebs);
+        ubifs_assert(!(c->bi.idx_growth & 7));
+        ubifs_assert(!(c->bi.data_growth & 7));
+        ubifs_assert(!(c->bi.dd_growth & 7));
         spin_unlock(&c->space_lock);
 }
 
@@ -586,13 +586,13 @@ void ubifs_convert_page_budget(struct ubifs_info *c)
 {
         spin_lock(&c->space_lock);
         /* Release the index growth reservation */
-        c->budg_idx_growth -= c->max_idx_node_sz << UBIFS_BLOCKS_PER_PAGE_SHIFT;
+        c->bi.idx_growth -= c->max_idx_node_sz << UBIFS_BLOCKS_PER_PAGE_SHIFT;
         /* Release the data growth reservation */
-        c->budg_data_growth -= c->page_budget;
+        c->bi.data_growth -= c->bi.page_budget;
         /* Increase the dirty data growth reservation instead */
-        c->budg_dd_growth += c->page_budget;
+        c->bi.dd_growth += c->bi.page_budget;
         /* And re-calculate the indexing space reservation */
-        c->min_idx_lebs = ubifs_calc_min_idx_lebs(c);
+        c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c);
         spin_unlock(&c->space_lock);
 }
 
@@ -612,7 +612,7 @@ void ubifs_release_dirty_inode_budget(struct ubifs_info *c,
 
         memset(&req, 0, sizeof(struct ubifs_budget_req));
         /* The "no space" flags will be cleared because dd_growth is > 0 */
-        req.dd_growth = c->inode_budget + ALIGN(ui->data_len, 8);
+        req.dd_growth = c->bi.inode_budget + ALIGN(ui->data_len, 8);
         ubifs_release_budget(c, &req);
 }
 
@@ -682,9 +682,9 @@ long long ubifs_get_free_space_nolock(struct ubifs_info *c)
         int rsvd_idx_lebs, lebs;
         long long available, outstanding, free;
 
-        ubifs_assert(c->min_idx_lebs == ubifs_calc_min_idx_lebs(c));
-        outstanding = c->budg_data_growth + c->budg_dd_growth;
-        available = ubifs_calc_available(c, c->min_idx_lebs);
+        ubifs_assert(c->bi.min_idx_lebs == ubifs_calc_min_idx_lebs(c));
+        outstanding = c->bi.data_growth + c->bi.dd_growth;
+        available = ubifs_calc_available(c, c->bi.min_idx_lebs);
 
         /*
          * When reporting free space to user-space, UBIFS guarantees that it is
@@ -697,8 +697,8 @@ long long ubifs_get_free_space_nolock(struct ubifs_info *c)
          * Note, the calculations below are similar to what we have in
          * 'do_budget_space()', so refer there for comments.
          */
-        if (c->min_idx_lebs > c->lst.idx_lebs)
-                rsvd_idx_lebs = c->min_idx_lebs - c->lst.idx_lebs;
+        if (c->bi.min_idx_lebs > c->lst.idx_lebs)
+                rsvd_idx_lebs = c->bi.min_idx_lebs - c->lst.idx_lebs;
         else
                 rsvd_idx_lebs = 0;
         lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt -
diff --git a/fs/ubifs/commit.c b/fs/ubifs/commit.c
index 37fa7ed062d8..87cd0ead8633 100644
--- a/fs/ubifs/commit.c
+++ b/fs/ubifs/commit.c
@@ -48,6 +48,56 @@
 #include <linux/slab.h>
 #include "ubifs.h"
 
+/*
+ * nothing_to_commit - check if there is nothing to commit.
+ * @c: UBIFS file-system description object
+ *
+ * This is a helper function which checks if there is anything to commit. It is
+ * used as an optimization to avoid starting the commit if it is not really
+ * necessary. Indeed, the commit operation always assumes flash I/O (e.g.,
+ * writing the commit start node to the log), and it is better to avoid doing
+ * this unnecessarily. E.g., 'ubifs_sync_fs()' runs the commit, but if there is
+ * nothing to commit, it is more optimal to avoid any flash I/O.
+ *
+ * This function has to be called with @c->commit_sem locked for writing -
+ * this function does not take LPT/TNC locks because the @c->commit_sem
+ * guarantees that we have exclusive access to the TNC and LPT data structures.
+ *
+ * This function returns %1 if there is nothing to commit and %0 otherwise.
+ */
+static int nothing_to_commit(struct ubifs_info *c)
+{
+        /*
+         * During mounting or remounting from R/O mode to R/W mode we may
+         * commit for various recovery-related reasons.
+         */
+        if (c->mounting || c->remounting_rw)
+                return 0;
+
+        /*
+         * If the root TNC node is dirty, we definitely have something to
+         * commit.
+         */
+        if (c->zroot.znode && test_bit(DIRTY_ZNODE, &c->zroot.znode->flags))
+                return 0;
+
+        /*
+         * Even though the TNC is clean, the LPT tree may have dirty nodes. For
+         * example, this may happen if the budgeting subsystem invoked GC to
+         * make some free space, and the GC found an LEB with only dirty and
+         * free space. In this case GC would just change the lprops of this
+         * LEB (by turning all space into free space) and unmap it.
+         */
+        if (c->nroot && test_bit(DIRTY_CNODE, &c->nroot->flags))
+                return 0;
+
+        ubifs_assert(atomic_long_read(&c->dirty_zn_cnt) == 0);
+        ubifs_assert(c->dirty_pn_cnt == 0);
+        ubifs_assert(c->dirty_nn_cnt == 0);
+
+        return 1;
+}
+
 /**
  * do_commit - commit the journal.
  * @c: UBIFS file-system description object
@@ -63,11 +113,19 @@ static int do_commit(struct ubifs_info *c)
         struct ubifs_lp_stats lst;
 
         dbg_cmt("start");
-        if (c->ro_media) {
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+
+        if (c->ro_error) {
                 err = -EROFS;
                 goto out_up;
         }
 
+        if (nothing_to_commit(c)) {
+                up_write(&c->commit_sem);
+                err = 0;
+                goto out_cancel;
+        }
+
         /* Sync all write buffers (necessary for recovery) */
         for (i = 0; i < c->jhead_cnt; i++) {
                 err = ubifs_wbuf_sync(&c->jheads[i].wbuf);
@@ -124,7 +182,7 @@ static int do_commit(struct ubifs_info *c)
         c->mst_node->root_len    = cpu_to_le32(zroot.len);
         c->mst_node->ihead_lnum  = cpu_to_le32(c->ihead_lnum);
         c->mst_node->ihead_offs  = cpu_to_le32(c->ihead_offs);
-        c->mst_node->index_size  = cpu_to_le64(c->old_idx_sz);
+        c->mst_node->index_size  = cpu_to_le64(c->bi.old_idx_sz);
         c->mst_node->lpt_lnum    = cpu_to_le32(c->lpt_lnum);
         c->mst_node->lpt_offs    = cpu_to_le32(c->lpt_offs);
         c->mst_node->nhead_lnum  = cpu_to_le32(c->nhead_lnum);
@@ -160,12 +218,12 @@ static int do_commit(struct ubifs_info *c)
         if (err)
                 goto out;
 
+out_cancel:
         spin_lock(&c->cs_lock);
         c->cmt_state = COMMIT_RESTING;
         wake_up(&c->cmt_wq);
         dbg_cmt("commit end");
         spin_unlock(&c->cs_lock);
-
         return 0;
 
 out_up:
@@ -519,7 +577,7 @@ int dbg_check_old_index(struct ubifs_info *c, struct ubifs_zbranch *zroot)
         size_t sz;
 
         if (!(ubifs_chk_flags & UBIFS_CHK_OLD_IDX))
-                goto out;
+                return 0;
 
         INIT_LIST_HEAD(&list);
 
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index c2a68baa782f..0bb2bcef0de9 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -34,7 +34,6 @@
 #include <linux/moduleparam.h>
 #include <linux/debugfs.h>
 #include <linux/math64.h>
-#include <linux/slab.h>
 
 #ifdef CONFIG_UBIFS_FS_DEBUG
 
@@ -43,15 +42,12 @@ DEFINE_SPINLOCK(dbg_lock);
 static char dbg_key_buf0[128];
 static char dbg_key_buf1[128];
 
-unsigned int ubifs_msg_flags = UBIFS_MSG_FLAGS_DEFAULT;
-unsigned int ubifs_chk_flags = UBIFS_CHK_FLAGS_DEFAULT;
+unsigned int ubifs_chk_flags;
 unsigned int ubifs_tst_flags;
 
-module_param_named(debug_msgs, ubifs_msg_flags, uint, S_IRUGO | S_IWUSR);
 module_param_named(debug_chks, ubifs_chk_flags, uint, S_IRUGO | S_IWUSR);
 module_param_named(debug_tsts, ubifs_tst_flags, uint, S_IRUGO | S_IWUSR);
 
-MODULE_PARM_DESC(debug_msgs, "Debug message type flags");
 MODULE_PARM_DESC(debug_chks, "Debug check flags");
 MODULE_PARM_DESC(debug_tsts, "Debug special test flags");
 
@@ -317,6 +313,8 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
                 printk(KERN_DEBUG "\tflags          %#x\n", sup_flags);
                 printk(KERN_DEBUG "\t  big_lpt      %u\n",
                        !!(sup_flags & UBIFS_FLG_BIGLPT));
+                printk(KERN_DEBUG "\t  space_fixup  %u\n",
+                       !!(sup_flags & UBIFS_FLG_SPACE_FIXUP));
                 printk(KERN_DEBUG "\tmin_io_size    %u\n",
                        le32_to_cpu(sup->min_io_size));
                 printk(KERN_DEBUG "\tleb_size       %u\n",
@@ -602,7 +600,7 @@ void dbg_dump_lstats(const struct ubifs_lp_stats *lst)
         spin_unlock(&dbg_lock);
 }
 
-void dbg_dump_budg(struct ubifs_info *c)
+void dbg_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi)
 {
         int i;
         struct rb_node *rb;
@@ -610,26 +608,42 @@ void dbg_dump_budg(struct ubifs_info *c)
         struct ubifs_gced_idx_leb *idx_gc;
         long long available, outstanding, free;
 
-        ubifs_assert(spin_is_locked(&c->space_lock));
+        spin_lock(&c->space_lock);
         spin_lock(&dbg_lock);
-        printk(KERN_DEBUG "(pid %d) Budgeting info: budg_data_growth %lld, "
-               "budg_dd_growth %lld, budg_idx_growth %lld\n", current->pid,
-               c->budg_data_growth, c->budg_dd_growth, c->budg_idx_growth);
-        printk(KERN_DEBUG "\tdata budget sum %lld, total budget sum %lld, "
-               "freeable_cnt %d\n", c->budg_data_growth + c->budg_dd_growth,
-               c->budg_data_growth + c->budg_dd_growth + c->budg_idx_growth,
-               c->freeable_cnt);
-        printk(KERN_DEBUG "\tmin_idx_lebs %d, old_idx_sz %lld, "
-               "calc_idx_sz %lld, idx_gc_cnt %d\n", c->min_idx_lebs,
-               c->old_idx_sz, c->calc_idx_sz, c->idx_gc_cnt);
+        printk(KERN_DEBUG "(pid %d) Budgeting info: data budget sum %lld, "
+               "total budget sum %lld\n", current->pid,
+               bi->data_growth + bi->dd_growth,
+               bi->data_growth + bi->dd_growth + bi->idx_growth);
+        printk(KERN_DEBUG "\tbudg_data_growth %lld, budg_dd_growth %lld, "
+               "budg_idx_growth %lld\n", bi->data_growth, bi->dd_growth,
+               bi->idx_growth);
+        printk(KERN_DEBUG "\tmin_idx_lebs %d, old_idx_sz %llu, "
+               "uncommitted_idx %lld\n", bi->min_idx_lebs, bi->old_idx_sz,
+               bi->uncommitted_idx);
+        printk(KERN_DEBUG "\tpage_budget %d, inode_budget %d, dent_budget %d\n",
+               bi->page_budget, bi->inode_budget, bi->dent_budget);
+        printk(KERN_DEBUG "\tnospace %u, nospace_rp %u\n",
+               bi->nospace, bi->nospace_rp);
+        printk(KERN_DEBUG "\tdark_wm %d, dead_wm %d, max_idx_node_sz %d\n",
+               c->dark_wm, c->dead_wm, c->max_idx_node_sz);
+
+        if (bi != &c->bi)
+                /*
+                 * If we are dumping saved budgeting data, do not print
+                 * additional information which is about the current state, not
+                 * the old one which corresponded to the saved budgeting data.
+                 */
+                goto out_unlock;
+
+        printk(KERN_DEBUG "\tfreeable_cnt %d, calc_idx_sz %lld, idx_gc_cnt %d\n",
+               c->freeable_cnt, c->calc_idx_sz, c->idx_gc_cnt);
         printk(KERN_DEBUG "\tdirty_pg_cnt %ld, dirty_zn_cnt %ld, "
                "clean_zn_cnt %ld\n", atomic_long_read(&c->dirty_pg_cnt),
                atomic_long_read(&c->dirty_zn_cnt),
                atomic_long_read(&c->clean_zn_cnt));
-        printk(KERN_DEBUG "\tdark_wm %d, dead_wm %d, max_idx_node_sz %d\n",
-               c->dark_wm, c->dead_wm, c->max_idx_node_sz);
         printk(KERN_DEBUG "\tgc_lnum %d, ihead_lnum %d\n",
                c->gc_lnum, c->ihead_lnum);
+
         /* If we are in R/O mode, journal heads do not exist */
         if (c->jheads)
                 for (i = 0; i < c->jhead_cnt; i++)
@@ -648,13 +662,15 @@ void dbg_dump_budg(struct ubifs_info *c)
         printk(KERN_DEBUG "\tcommit state %d\n", c->cmt_state);
 
         /* Print budgeting predictions */
-        available = ubifs_calc_available(c, c->min_idx_lebs);
-        outstanding = c->budg_data_growth + c->budg_dd_growth;
+        available = ubifs_calc_available(c, c->bi.min_idx_lebs);
+        outstanding = c->bi.data_growth + c->bi.dd_growth;
         free = ubifs_get_free_space_nolock(c);
         printk(KERN_DEBUG "Budgeting predictions:\n");
         printk(KERN_DEBUG "\tavailable: %lld, outstanding %lld, free %lld\n",
                available, outstanding, free);
+out_unlock:
         spin_unlock(&dbg_lock);
+        spin_unlock(&c->space_lock);
 }
 
 void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp)
@@ -729,7 +745,13 @@ void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp)
                 if (bud->lnum == lp->lnum) {
                         int head = 0;
                         for (i = 0; i < c->jhead_cnt; i++) {
-                                if (lp->lnum == c->jheads[i].wbuf.lnum) {
+                                /*
+                                 * Note, if we are in R/O mode or in the middle
+                                 * of mounting/re-mounting, the write-buffers do
+                                 * not exist.
+                                 */
+                                if (c->jheads &&
+                                    lp->lnum == c->jheads[i].wbuf.lnum) {
                                         printk(KERN_CONT ", jhead %s",
                                                dbg_jhead(i));
                                         head = 1;
@@ -810,16 +832,24 @@ void dbg_dump_leb(const struct ubifs_info *c, int lnum)
 {
         struct ubifs_scan_leb *sleb;
         struct ubifs_scan_node *snod;
+        void *buf;
 
         if (dbg_failure_mode)
                 return;
 
         printk(KERN_DEBUG "(pid %d) start dumping LEB %d\n",
                current->pid, lnum);
-        sleb = ubifs_scan(c, lnum, 0, c->dbg->buf, 0);
+
+        buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+        if (!buf) {
+                ubifs_err("cannot allocate memory for dumping LEB %d", lnum);
+                return;
+        }
+
+        sleb = ubifs_scan(c, lnum, 0, buf, 0);
         if (IS_ERR(sleb)) {
                 ubifs_err("scan error %d", (int)PTR_ERR(sleb));
-                return;
+                goto out;
         }
 
         printk(KERN_DEBUG "LEB %d has %d nodes ending at %d\n", lnum,
@@ -835,6 +865,9 @@ void dbg_dump_leb(const struct ubifs_info *c, int lnum)
         printk(KERN_DEBUG "(pid %d) finish dumping LEB %d\n",
                current->pid, lnum);
         ubifs_scan_destroy(sleb);
+
+out:
+        vfree(buf);
         return;
 }
 
@@ -961,11 +994,41 @@ void dbg_dump_index(struct ubifs_info *c)
 void dbg_save_space_info(struct ubifs_info *c)
 {
         struct ubifs_debug_info *d = c->dbg;
-
-        ubifs_get_lp_stats(c, &d->saved_lst);
+        int freeable_cnt;
 
         spin_lock(&c->space_lock);
+        memcpy(&d->saved_lst, &c->lst, sizeof(struct ubifs_lp_stats));
+        memcpy(&d->saved_bi, &c->bi, sizeof(struct ubifs_budg_info));
+        d->saved_idx_gc_cnt = c->idx_gc_cnt;
+
+        /*
+         * We use a dirty hack here and zero out @c->freeable_cnt, because it
+         * affects the free space calculations, and UBIFS might not know about
+         * all freeable eraseblocks. Indeed, we know about freeable eraseblocks
+         * only when we read their lprops, and we do this only lazily, upon the
+         * need. So at any given point of time @c->freeable_cnt might be not
+         * exactly accurate.
+         *
+         * Just one example about the issue we hit when we did not zero
+         * @c->freeable_cnt.
+         * 1. The file-system is mounted R/O, c->freeable_cnt is %0. We save the
+         *    amount of free space in @d->saved_free
+         * 2. We re-mount R/W, which makes UBIFS to read the "lsave"
+         *    information from flash, where we cache LEBs from various
+         *    categories ('ubifs_remount_fs()' -> 'ubifs_lpt_init()'
+         *    -> 'lpt_init_wr()' -> 'read_lsave()' -> 'ubifs_lpt_lookup()'
+         *    -> 'ubifs_get_pnode()' -> 'update_cats()'
+         *    -> 'ubifs_add_to_cat()').
+         * 3. Lsave contains a freeable eraseblock, and @c->freeable_cnt
+         *    becomes %1.
+         * 4. We calculate the amount of free space when the re-mount is
+         *    finished in 'dbg_check_space_info()' and it does not match
+         *    @d->saved_free.
+         */
+        freeable_cnt = c->freeable_cnt;
+        c->freeable_cnt = 0;
         d->saved_free = ubifs_get_free_space_nolock(c);
+        c->freeable_cnt = freeable_cnt;
         spin_unlock(&c->space_lock);
 }
 
@@ -982,12 +1045,15 @@ int dbg_check_space_info(struct ubifs_info *c)
 {
         struct ubifs_debug_info *d = c->dbg;
         struct ubifs_lp_stats lst;
-        long long avail, free;
+        long long free;
+        int freeable_cnt;
 
         spin_lock(&c->space_lock);
-        avail = ubifs_calc_available(c, c->min_idx_lebs);
+        freeable_cnt = c->freeable_cnt;
+        c->freeable_cnt = 0;
+        free = ubifs_get_free_space_nolock(c);
+        c->freeable_cnt = freeable_cnt;
         spin_unlock(&c->space_lock);
-        free = ubifs_get_free_space(c);
 
         if (free != d->saved_free) {
                 ubifs_err("free space changed from %lld to %lld",
@@ -1000,14 +1066,14 @@ int dbg_check_space_info(struct ubifs_info *c)
 out:
         ubifs_msg("saved lprops statistics dump");
         dbg_dump_lstats(&d->saved_lst);
-        ubifs_get_lp_stats(c, &lst);
-
+        ubifs_msg("saved budgeting info dump");
+        dbg_dump_budg(c, &d->saved_bi);
+        ubifs_msg("saved idx_gc_cnt %d", d->saved_idx_gc_cnt);
         ubifs_msg("current lprops statistics dump");
+        ubifs_get_lp_stats(c, &lst);
         dbg_dump_lstats(&lst);
-
-        spin_lock(&c->space_lock);
-        dbg_dump_budg(c);
-        spin_unlock(&c->space_lock);
+        ubifs_msg("current budgeting info dump");
+        dbg_dump_budg(c, &c->bi);
         dump_stack();
         return -EINVAL;
 }
@@ -1751,6 +1817,8 @@ static struct fsck_inode *add_inode(struct ubifs_info *c,
         struct rb_node **p, *parent = NULL;
         struct fsck_inode *fscki;
         ino_t inum = key_inum_flash(c, &ino->key);
+        struct inode *inode;
+        struct ubifs_inode *ui;
 
         p = &fsckd->inodes.rb_node;
         while (*p) {
@@ -1774,19 +1842,46 @@ static struct fsck_inode *add_inode(struct ubifs_info *c,
         if (!fscki)
                 return ERR_PTR(-ENOMEM);
 
+        inode = ilookup(c->vfs_sb, inum);
+
         fscki->inum = inum;
-        fscki->nlink = le32_to_cpu(ino->nlink);
-        fscki->size = le64_to_cpu(ino->size);
-        fscki->xattr_cnt = le32_to_cpu(ino->xattr_cnt);
-        fscki->xattr_sz = le32_to_cpu(ino->xattr_size);
-        fscki->xattr_nms = le32_to_cpu(ino->xattr_names);
-        fscki->mode = le32_to_cpu(ino->mode);
+        /*
+         * If the inode is present in the VFS inode cache, use it instead of
+         * the on-flash inode which might be out-of-date. E.g., the size might
+         * be out-of-date. If we do not do this, the following may happen, for
+         * example:
+         *   1. A power cut happens
+         *   2. We mount the file-system R/O, the replay process fixes up the
+         *      inode size in the VFS cache, but on on-flash.
+         *   3. 'check_leaf()' fails because it hits a data node beyond inode
+         *      size.
+         */
+        if (!inode) {
+                fscki->nlink = le32_to_cpu(ino->nlink);
+                fscki->size = le64_to_cpu(ino->size);
+                fscki->xattr_cnt = le32_to_cpu(ino->xattr_cnt);
+                fscki->xattr_sz = le32_to_cpu(ino->xattr_size);
+                fscki->xattr_nms = le32_to_cpu(ino->xattr_names);
+                fscki->mode = le32_to_cpu(ino->mode);
+        } else {
+                ui = ubifs_inode(inode);
+                fscki->nlink = inode->i_nlink;
+                fscki->size = inode->i_size;
+                fscki->xattr_cnt = ui->xattr_cnt;
+                fscki->xattr_sz = ui->xattr_size;
+                fscki->xattr_nms = ui->xattr_names;
+                fscki->mode = inode->i_mode;
+                iput(inode);
+        }
+
         if (S_ISDIR(fscki->mode)) {
                 fscki->calc_sz = UBIFS_INO_NODE_SZ;
                 fscki->calc_cnt = 2;
         }
+
         rb_link_node(&fscki->rb, parent, p);
         rb_insert_color(&fscki->rb, &fsckd->inodes);
+
         return fscki;
 }
 
@@ -2239,14 +2334,169 @@ out_free:
         return err;
 }
 
-static int invocation_cnt;
+/**
+ * dbg_check_data_nodes_order - check that list of data nodes is sorted.
+ * @c: UBIFS file-system description object
+ * @head: the list of nodes ('struct ubifs_scan_node' objects)
+ *
+ * This function returns zero if the list of data nodes is sorted correctly,
+ * and %-EINVAL if not.
+ */
+int dbg_check_data_nodes_order(struct ubifs_info *c, struct list_head *head)
+{
+        struct list_head *cur;
+        struct ubifs_scan_node *sa, *sb;
+
+        if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+                return 0;
+
+        for (cur = head->next; cur->next != head; cur = cur->next) {
+                ino_t inuma, inumb;
+                uint32_t blka, blkb;
+
+                cond_resched();
+                sa = container_of(cur, struct ubifs_scan_node, list);
+                sb = container_of(cur->next, struct ubifs_scan_node, list);
+
+                if (sa->type != UBIFS_DATA_NODE) {
+                        ubifs_err("bad node type %d", sa->type);
+                        dbg_dump_node(c, sa->node);
+                        return -EINVAL;
+                }
+                if (sb->type != UBIFS_DATA_NODE) {
+                        ubifs_err("bad node type %d", sb->type);
+                        dbg_dump_node(c, sb->node);
+                        return -EINVAL;
+                }
+
+                inuma = key_inum(c, &sa->key);
+                inumb = key_inum(c, &sb->key);
+
+                if (inuma < inumb)
+                        continue;
+                if (inuma > inumb) {
+                        ubifs_err("larger inum %lu goes before inum %lu",
+                                  (unsigned long)inuma, (unsigned long)inumb);
+                        goto error_dump;
+                }
+
+                blka = key_block(c, &sa->key);
+                blkb = key_block(c, &sb->key);
+
+                if (blka > blkb) {
+                        ubifs_err("larger block %u goes before %u", blka, blkb);
+                        goto error_dump;
+                }
+                if (blka == blkb) {
+                        ubifs_err("two data nodes for the same block");
+                        goto error_dump;
+                }
+        }
+
+        return 0;
+
+error_dump:
+        dbg_dump_node(c, sa->node);
+        dbg_dump_node(c, sb->node);
+        return -EINVAL;
+}
+
+/**
+ * dbg_check_nondata_nodes_order - check that list of data nodes is sorted.
+ * @c: UBIFS file-system description object
+ * @head: the list of nodes ('struct ubifs_scan_node' objects)
+ *
+ * This function returns zero if the list of non-data nodes is sorted correctly,
+ * and %-EINVAL if not.
+ */
+int dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head)
+{
+        struct list_head *cur;
+        struct ubifs_scan_node *sa, *sb;
+
+        if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+                return 0;
+
+        for (cur = head->next; cur->next != head; cur = cur->next) {
+                ino_t inuma, inumb;
+                uint32_t hasha, hashb;
+
+                cond_resched();
+                sa = container_of(cur, struct ubifs_scan_node, list);
+                sb = container_of(cur->next, struct ubifs_scan_node, list);
+
+                if (sa->type != UBIFS_INO_NODE && sa->type != UBIFS_DENT_NODE &&
+                    sa->type != UBIFS_XENT_NODE) {
+                        ubifs_err("bad node type %d", sa->type);
+                        dbg_dump_node(c, sa->node);
+                        return -EINVAL;
+                }
+                if (sa->type != UBIFS_INO_NODE && sa->type != UBIFS_DENT_NODE &&
+                    sa->type != UBIFS_XENT_NODE) {
+                        ubifs_err("bad node type %d", sb->type);
+                        dbg_dump_node(c, sb->node);
+                        return -EINVAL;
+                }
+
+                if (sa->type != UBIFS_INO_NODE && sb->type == UBIFS_INO_NODE) {
+                        ubifs_err("non-inode node goes before inode node");
+                        goto error_dump;
+                }
+
+                if (sa->type == UBIFS_INO_NODE && sb->type != UBIFS_INO_NODE)
+                        continue;
+
+                if (sa->type == UBIFS_INO_NODE && sb->type == UBIFS_INO_NODE) {
+                        /* Inode nodes are sorted in descending size order */
+                        if (sa->len < sb->len) {
+                                ubifs_err("smaller inode node goes first");
+                                goto error_dump;
+                        }
+                        continue;
+                }
+
+                /*
+                 * This is either a dentry or xentry, which should be sorted in
+                 * ascending (parent ino, hash) order.
+                 */
+                inuma = key_inum(c, &sa->key);
+                inumb = key_inum(c, &sb->key);
+
+                if (inuma < inumb)
+                        continue;
+                if (inuma > inumb) {
+                        ubifs_err("larger inum %lu goes before inum %lu",
+                                  (unsigned long)inuma, (unsigned long)inumb);
+                        goto error_dump;
+                }
+
+                hasha = key_block(c, &sa->key);
+                hashb = key_block(c, &sb->key);
+
+                if (hasha > hashb) {
+                        ubifs_err("larger hash %u goes before %u",
+                                  hasha, hashb);
+                        goto error_dump;
+                }
+        }
+
+        return 0;
+
+error_dump:
+        ubifs_msg("dumping first node");
+        dbg_dump_node(c, sa->node);
+        ubifs_msg("dumping second node");
+        dbg_dump_node(c, sb->node);
+        return -EINVAL;
+        return 0;
+}
 
 int dbg_force_in_the_gaps(void)
 {
-        if (!dbg_force_in_the_gaps_enabled)
+        if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
                 return 0;
-        /* Force in-the-gaps every 8th commit */
-        return !((invocation_cnt++) & 0x7);
+
+        return !(random32() & 7);
 }
 
 /* Failure mode for recovery testing */
@@ -2434,7 +2684,7 @@ int dbg_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset,
                  int len, int check)
 {
         if (in_failure_mode(desc))
-                return -EIO;
+                return -EROFS;
         return ubi_leb_read(desc, lnum, buf, offset, len, check);
 }
 
@@ -2444,7 +2694,7 @@ int dbg_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
         int err, failing;
 
         if (in_failure_mode(desc))
-                return -EIO;
+                return -EROFS;
         failing = do_fail(desc, lnum, 1);
         if (failing)
                 cut_data(buf, len);
@@ -2452,7 +2702,7 @@ int dbg_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
         if (err)
                 return err;
         if (failing)
-                return -EIO;
+                return -EROFS;
         return 0;
 }
 
@@ -2462,12 +2712,12 @@ int dbg_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf,
         int err;
 
         if (do_fail(desc, lnum, 1))
-                return -EIO;
+                return -EROFS;
         err = ubi_leb_change(desc, lnum, buf, len, dtype);
         if (err)
                 return err;
         if (do_fail(desc, lnum, 1))
-                return -EIO;
+                return -EROFS;
         return 0;
 }
 
@@ -2476,12 +2726,12 @@ int dbg_leb_erase(struct ubi_volume_desc *desc, int lnum)
         int err;
 
         if (do_fail(desc, lnum, 0))
-                return -EIO;
+                return -EROFS;
         err = ubi_leb_erase(desc, lnum);
         if (err)
                 return err;
         if (do_fail(desc, lnum, 0))
-                return -EIO;
+                return -EROFS;
         return 0;
 }
 
@@ -2490,19 +2740,19 @@ int dbg_leb_unmap(struct ubi_volume_desc *desc, int lnum)
         int err;
 
         if (do_fail(desc, lnum, 0))
-                return -EIO;
+                return -EROFS;
         err = ubi_leb_unmap(desc, lnum);
         if (err)
                 return err;
         if (do_fail(desc, lnum, 0))
-                return -EIO;
+                return -EROFS;
         return 0;
 }
 
 int dbg_is_mapped(struct ubi_volume_desc *desc, int lnum)
 {
         if (in_failure_mode(desc))
-                return -EIO;
+                return -EROFS;
         return ubi_is_mapped(desc, lnum);
 }
 
@@ -2511,12 +2761,12 @@ int dbg_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype)
         int err;
 
         if (do_fail(desc, lnum, 0))
-                return -EIO;
+                return -EROFS;
         err = ubi_leb_map(desc, lnum, dtype);
         if (err)
                 return err;
         if (do_fail(desc, lnum, 0))
-                return -EIO;
+                return -EROFS;
         return 0;
 }
 
@@ -2534,16 +2784,8 @@ int ubifs_debugging_init(struct ubifs_info *c)
         if (!c->dbg)
                 return -ENOMEM;
 
-        c->dbg->buf = vmalloc(c->leb_size);
-        if (!c->dbg->buf)
-                goto out;
-
         failure_mode_init(c);
         return 0;
-
-out:
-        kfree(c->dbg);
-        return -ENOMEM;
 }
 
 /**
@@ -2553,7 +2795,6 @@ out:
 void ubifs_debugging_exit(struct ubifs_info *c)
 {
         failure_mode_exit(c);
-        vfree(c->dbg->buf);
         kfree(c->dbg);
 }
 
@@ -2595,7 +2836,7 @@ void dbg_debugfs_exit(void)
 static int open_debugfs_file(struct inode *inode, struct file *file)
 {
         file->private_data = inode->i_private;
-        return 0;
+        return nonseekable_open(inode, file);
 }
 
 static ssize_t write_debugfs_file(struct file *file, const char __user *buf,
@@ -2606,18 +2847,15 @@ static ssize_t write_debugfs_file(struct file *file, const char __user *buf,
 
         if (file->f_path.dentry == d->dfs_dump_lprops)
                 dbg_dump_lprops(c);
-        else if (file->f_path.dentry == d->dfs_dump_budg) {
-                spin_lock(&c->space_lock);
-                dbg_dump_budg(c);
-                spin_unlock(&c->space_lock);
-        } else if (file->f_path.dentry == d->dfs_dump_tnc) {
+        else if (file->f_path.dentry == d->dfs_dump_budg)
+                dbg_dump_budg(c, &c->bi);
+        else if (file->f_path.dentry == d->dfs_dump_tnc) {
                 mutex_lock(&c->tnc_mutex);
                 dbg_dump_tnc(c);
                 mutex_unlock(&c->tnc_mutex);
         } else
                 return -EINVAL;
 
-        *ppos += count;
         return count;
 }
 
@@ -2625,6 +2863,7 @@ static const struct file_operations dfs_fops = {
         .open = open_debugfs_file,
         .write = write_debugfs_file,
         .owner = THIS_MODULE,
+        .llseek = no_llseek,
 };
 
 /**
@@ -2647,40 +2886,38 @@ int dbg_debugfs_init_fs(struct ubifs_info *c)
         struct ubifs_debug_info *d = c->dbg;
 
         sprintf(d->dfs_dir_name, "ubi%d_%d", c->vi.ubi_num, c->vi.vol_id);
-        d->dfs_dir = debugfs_create_dir(d->dfs_dir_name, dfs_rootdir);
-        if (IS_ERR(d->dfs_dir)) {
-                err = PTR_ERR(d->dfs_dir);
-                ubifs_err("cannot create \"%s\" debugfs directory, error %d\n",
-                          d->dfs_dir_name, err);
+        fname = d->dfs_dir_name;
+        dent = debugfs_create_dir(fname, dfs_rootdir);
+        if (IS_ERR_OR_NULL(dent))
                 goto out;
-        }
+        d->dfs_dir = dent;
 
         fname = "dump_lprops";
-        dent = debugfs_create_file(fname, S_IWUGO, d->dfs_dir, c, &dfs_fops);
-        if (IS_ERR(dent))
+        dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops);
+        if (IS_ERR_OR_NULL(dent))
                 goto out_remove;
         d->dfs_dump_lprops = dent;
 
         fname = "dump_budg";
-        dent = debugfs_create_file(fname, S_IWUGO, d->dfs_dir, c, &dfs_fops);
-        if (IS_ERR(dent))
+        dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops);
+        if (IS_ERR_OR_NULL(dent))
                 goto out_remove;
         d->dfs_dump_budg = dent;
 
         fname = "dump_tnc";
-        dent = debugfs_create_file(fname, S_IWUGO, d->dfs_dir, c, &dfs_fops);
-        if (IS_ERR(dent))
+        dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops);
+        if (IS_ERR_OR_NULL(dent))
                 goto out_remove;
         d->dfs_dump_tnc = dent;
 
         return 0;
 
 out_remove:
-        err = PTR_ERR(dent);
-        ubifs_err("cannot create \"%s\" debugfs directory, error %d\n",
-                  fname, err);
         debugfs_remove_recursive(d->dfs_dir);
 out:
+        err = dent ? PTR_ERR(dent) : -ENODEV;
+        ubifs_err("cannot create \"%s\" debugfs directory, error %d\n",
+                  fname, err);
         return err;
 }
 
diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h
index 29d960101ea6..a811ac4a26bb 100644
--- a/fs/ubifs/debug.h
+++ b/fs/ubifs/debug.h
@@ -23,11 +23,18 @@
 #ifndef __UBIFS_DEBUG_H__
 #define __UBIFS_DEBUG_H__
 
+/* Checking helper functions */
+typedef int (*dbg_leaf_callback)(struct ubifs_info *c,
+                                 struct ubifs_zbranch *zbr, void *priv);
+typedef int (*dbg_znode_callback)(struct ubifs_info *c,
+                                  struct ubifs_znode *znode, void *priv);
+
 #ifdef CONFIG_UBIFS_FS_DEBUG
 
+#include <linux/random.h>
+
 /**
  * ubifs_debug_info - per-FS debugging information.
- * @buf: a buffer of LEB size, used for various purposes
  * @old_zroot: old index root - used by 'dbg_check_old_index()'
  * @old_zroot_level: old index root level - used by 'dbg_check_old_index()'
  * @old_zroot_sqnum: old index root sqnum - used by 'dbg_check_old_index()'
@@ -45,16 +52,17 @@
  * @new_ihead_offs: used by debugging to check @c->ihead_offs
  *
  * @saved_lst: saved lprops statistics (used by 'dbg_save_space_info()')
- * @saved_free: saved free space (used by 'dbg_save_space_info()')
+ * @saved_bi: saved budgeting information
+ * @saved_free: saved amount of free space
+ * @saved_idx_gc_cnt: saved value of @c->idx_gc_cnt
  *
- * dfs_dir_name: name of debugfs directory containing this file-system's files
- * dfs_dir: direntry object of the file-system debugfs directory
- * dfs_dump_lprops: "dump lprops" debugfs knob
- * dfs_dump_budg: "dump budgeting information" debugfs knob
- * dfs_dump_tnc: "dump TNC" debugfs knob
+ * @dfs_dir_name: name of debugfs directory containing this file-system's files
+ * @dfs_dir: direntry object of the file-system debugfs directory
+ * @dfs_dump_lprops: "dump lprops" debugfs knob
+ * @dfs_dump_budg: "dump budgeting information" debugfs knob
+ * @dfs_dump_tnc: "dump TNC" debugfs knob
  */
 struct ubifs_debug_info {
-        void *buf;
         struct ubifs_zbranch old_zroot;
         int old_zroot_level;
         unsigned long long old_zroot_sqnum;
@@ -72,7 +80,9 @@ struct ubifs_debug_info {
         int new_ihead_offs;
 
         struct ubifs_lp_stats saved_lst;
+        struct ubifs_budg_info saved_bi;
         long long saved_free;
+        int saved_idx_gc_cnt;
 
         char dfs_dir_name[100];
         struct dentry *dfs_dir;
@@ -97,23 +107,7 @@ struct ubifs_debug_info {
         }                                                                      \
 } while (0)
 
-#define dbg_dump_stack() do {                                                  \
-        if (!dbg_failure_mode)                                                 \
-                dump_stack();                                                  \
-} while (0)
-
-/* Generic debugging messages */
-#define dbg_msg(fmt, ...) do {                                                 \
-        spin_lock(&dbg_lock);                                                  \
-        printk(KERN_DEBUG "UBIFS DBG (pid %d): %s: " fmt "\n", current->pid,   \
-               __func__, ##__VA_ARGS__);                                       \
-        spin_unlock(&dbg_lock);                                                \
-} while (0)
-
-#define dbg_do_msg(typ, fmt, ...) do {                                         \
-        if (ubifs_msg_flags & typ)                                             \
-                dbg_msg(fmt, ##__VA_ARGS__);                                   \
-} while (0)
+#define dbg_dump_stack() dump_stack()
 
 #define dbg_err(fmt, ...) do {                                                 \
         spin_lock(&dbg_lock);                                                  \
@@ -133,86 +127,43 @@ const char *dbg_key_str1(const struct ubifs_info *c,
 #define DBGKEY(key) dbg_key_str0(c, (key))
 #define DBGKEY1(key) dbg_key_str1(c, (key))
 
-/* General messages */
-#define dbg_gen(fmt, ...)   dbg_do_msg(UBIFS_MSG_GEN, fmt, ##__VA_ARGS__)
+#define ubifs_dbg_msg(type, fmt, ...) do {                        \
+        spin_lock(&dbg_lock);                                     \
+        pr_debug("UBIFS DBG " type ": " fmt "\n", ##__VA_ARGS__); \
+        spin_unlock(&dbg_lock);                                   \
+} while (0)
 
+/* Just a debugging messages not related to any specific UBIFS subsystem */
+#define dbg_msg(fmt, ...)   ubifs_dbg_msg("msg", fmt, ##__VA_ARGS__)
+/* General messages */
+#define dbg_gen(fmt, ...)   ubifs_dbg_msg("gen", fmt, ##__VA_ARGS__)
 /* Additional journal messages */
-#define dbg_jnl(fmt, ...)   dbg_do_msg(UBIFS_MSG_JNL, fmt, ##__VA_ARGS__)
-
+#define dbg_jnl(fmt, ...)   ubifs_dbg_msg("jnl", fmt, ##__VA_ARGS__)
 /* Additional TNC messages */
-#define dbg_tnc(fmt, ...)   dbg_do_msg(UBIFS_MSG_TNC, fmt, ##__VA_ARGS__)
-
+#define dbg_tnc(fmt, ...)   ubifs_dbg_msg("tnc", fmt, ##__VA_ARGS__)
 /* Additional lprops messages */
-#define dbg_lp(fmt, ...)    dbg_do_msg(UBIFS_MSG_LP, fmt, ##__VA_ARGS__)
-
+#define dbg_lp(fmt, ...)    ubifs_dbg_msg("lp", fmt, ##__VA_ARGS__)
 /* Additional LEB find messages */
-#define dbg_find(fmt, ...)  dbg_do_msg(UBIFS_MSG_FIND, fmt, ##__VA_ARGS__)
-
+#define dbg_find(fmt, ...)  ubifs_dbg_msg("find", fmt, ##__VA_ARGS__)
 /* Additional mount messages */
-#define dbg_mnt(fmt, ...)   dbg_do_msg(UBIFS_MSG_MNT, fmt, ##__VA_ARGS__)
-
+#define dbg_mnt(fmt, ...)   ubifs_dbg_msg("mnt", fmt, ##__VA_ARGS__)
 /* Additional I/O messages */
-#define dbg_io(fmt, ...)    dbg_do_msg(UBIFS_MSG_IO, fmt, ##__VA_ARGS__)
-
+#define dbg_io(fmt, ...)    ubifs_dbg_msg("io", fmt, ##__VA_ARGS__)
 /* Additional commit messages */
-#define dbg_cmt(fmt, ...)   dbg_do_msg(UBIFS_MSG_CMT, fmt, ##__VA_ARGS__)
-
+#define dbg_cmt(fmt, ...)   ubifs_dbg_msg("cmt", fmt, ##__VA_ARGS__)
 /* Additional budgeting messages */
-#define dbg_budg(fmt, ...)  dbg_do_msg(UBIFS_MSG_BUDG, fmt, ##__VA_ARGS__)
-
+#define dbg_budg(fmt, ...)  ubifs_dbg_msg("budg", fmt, ##__VA_ARGS__)
 /* Additional log messages */
-#define dbg_log(fmt, ...)   dbg_do_msg(UBIFS_MSG_LOG, fmt, ##__VA_ARGS__)
-
+#define dbg_log(fmt, ...)   ubifs_dbg_msg("log", fmt, ##__VA_ARGS__)
 /* Additional gc messages */
-#define dbg_gc(fmt, ...)    dbg_do_msg(UBIFS_MSG_GC, fmt, ##__VA_ARGS__)
-
+#define dbg_gc(fmt, ...)    ubifs_dbg_msg("gc", fmt, ##__VA_ARGS__)
 /* Additional scan messages */
-#define dbg_scan(fmt, ...)  dbg_do_msg(UBIFS_MSG_SCAN, fmt, ##__VA_ARGS__)
-
+#define dbg_scan(fmt, ...)  ubifs_dbg_msg("scan", fmt, ##__VA_ARGS__)
 /* Additional recovery messages */
-#define dbg_rcvry(fmt, ...) dbg_do_msg(UBIFS_MSG_RCVRY, fmt, ##__VA_ARGS__)
+#define dbg_rcvry(fmt, ...) ubifs_dbg_msg("rcvry", fmt, ##__VA_ARGS__)
 
 /*
- * Debugging message type flags (must match msg_type_names in debug.c).
- *
- * UBIFS_MSG_GEN: general messages
- * UBIFS_MSG_JNL: journal messages
- * UBIFS_MSG_MNT: mount messages
- * UBIFS_MSG_CMT: commit messages
- * UBIFS_MSG_FIND: LEB find messages
- * UBIFS_MSG_BUDG: budgeting messages
- * UBIFS_MSG_GC: garbage collection messages
- * UBIFS_MSG_TNC: TNC messages
- * UBIFS_MSG_LP: lprops messages
- * UBIFS_MSG_IO: I/O messages
- * UBIFS_MSG_LOG: log messages
- * UBIFS_MSG_SCAN: scan messages
- * UBIFS_MSG_RCVRY: recovery messages
- */
-enum {
-        UBIFS_MSG_GEN   = 0x1,
-        UBIFS_MSG_JNL   = 0x2,
-        UBIFS_MSG_MNT   = 0x4,
-        UBIFS_MSG_CMT   = 0x8,
-        UBIFS_MSG_FIND  = 0x10,
-        UBIFS_MSG_BUDG  = 0x20,
-        UBIFS_MSG_GC    = 0x40,
-        UBIFS_MSG_TNC   = 0x80,
-        UBIFS_MSG_LP    = 0x100,
-        UBIFS_MSG_IO    = 0x200,
-        UBIFS_MSG_LOG   = 0x400,
-        UBIFS_MSG_SCAN  = 0x800,
-        UBIFS_MSG_RCVRY = 0x1000,
-};
-
-/* Debugging message type flags for each default debug message level */
-#define UBIFS_MSG_LVL_0 0
-#define UBIFS_MSG_LVL_1 0x1
-#define UBIFS_MSG_LVL_2 0x7f
-#define UBIFS_MSG_LVL_3 0xffff
-
-/*
- * Debugging check flags (must match chk_names in debug.c).
+ * Debugging check flags.
  *
  * UBIFS_CHK_GEN: general checks
  * UBIFS_CHK_TNC: check TNC
@@ -233,32 +184,14 @@ enum {
 };
 
 /*
- * Special testing flags (must match tst_names in debug.c).
+ * Special testing flags.
  *
- * UBIFS_TST_FORCE_IN_THE_GAPS: force the use of in-the-gaps method
  * UBIFS_TST_RCVRY: failure mode for recovery testing
  */
 enum {
-        UBIFS_TST_FORCE_IN_THE_GAPS = 0x2,
         UBIFS_TST_RCVRY             = 0x4,
 };
 
-#if CONFIG_UBIFS_FS_DEBUG_MSG_LVL == 1
-#define UBIFS_MSG_FLAGS_DEFAULT UBIFS_MSG_LVL_1
-#elif CONFIG_UBIFS_FS_DEBUG_MSG_LVL == 2
-#define UBIFS_MSG_FLAGS_DEFAULT UBIFS_MSG_LVL_2
-#elif CONFIG_UBIFS_FS_DEBUG_MSG_LVL == 3
-#define UBIFS_MSG_FLAGS_DEFAULT UBIFS_MSG_LVL_3
-#else
-#define UBIFS_MSG_FLAGS_DEFAULT UBIFS_MSG_LVL_0
-#endif
-
-#ifdef CONFIG_UBIFS_FS_DEBUG_CHKS
-#define UBIFS_CHK_FLAGS_DEFAULT 0xffffffff
-#else
-#define UBIFS_CHK_FLAGS_DEFAULT 0
-#endif
-
 extern spinlock_t dbg_lock;
 
 extern unsigned int ubifs_msg_flags;
@@ -280,7 +213,7 @@ void dbg_dump_lpt_node(const struct ubifs_info *c, void *node, int lnum,
                        int offs);
 void dbg_dump_budget_req(const struct ubifs_budget_req *req);
 void dbg_dump_lstats(const struct ubifs_lp_stats *lst);
-void dbg_dump_budg(struct ubifs_info *c);
+void dbg_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi);
 void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp);
 void dbg_dump_lprops(struct ubifs_info *c);
 void dbg_dump_lpt_info(struct ubifs_info *c);
@@ -294,11 +227,6 @@ void dbg_dump_tnc(struct ubifs_info *c);
 void dbg_dump_index(struct ubifs_info *c);
 void dbg_dump_lpt_lebs(const struct ubifs_info *c);
 
-/* Checking helper functions */
-typedef int (*dbg_leaf_callback)(struct ubifs_info *c,
-                                 struct ubifs_zbranch *zbr, void *priv);
-typedef int (*dbg_znode_callback)(struct ubifs_info *c,
-                                  struct ubifs_znode *znode, void *priv);
 int dbg_walk_index(struct ubifs_info *c, dbg_leaf_callback leaf_cb,
                    dbg_znode_callback znode_cb, void *priv);
 
@@ -319,25 +247,24 @@ int dbg_check_idx_size(struct ubifs_info *c, long long idx_size);
 int dbg_check_filesystem(struct ubifs_info *c);
 void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat,
                     int add_pos);
-int dbg_check_lprops(struct ubifs_info *c);
 int dbg_check_lpt_nodes(struct ubifs_info *c, struct ubifs_cnode *cnode,
                         int row, int col);
 int dbg_check_inode_size(struct ubifs_info *c, const struct inode *inode,
                          loff_t size);
+int dbg_check_data_nodes_order(struct ubifs_info *c, struct list_head *head);
+int dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head);
 
 /* Force the use of in-the-gaps method for testing */
-
-#define dbg_force_in_the_gaps_enabled \
-        (ubifs_tst_flags & UBIFS_TST_FORCE_IN_THE_GAPS)
-
+static inline int dbg_force_in_the_gaps_enabled(void)
+{
+        return ubifs_chk_flags & UBIFS_CHK_GEN;
+}
 int dbg_force_in_the_gaps(void);
 
 /* Failure mode for recovery testing */
-
 #define dbg_failure_mode (ubifs_tst_flags & UBIFS_TST_RCVRY)
 
 #ifndef UBIFS_DBG_PRESERVE_UBI
-
 #define ubi_leb_read   dbg_leb_read
 #define ubi_leb_write  dbg_leb_write
 #define ubi_leb_change dbg_leb_change
@@ -345,7 +272,6 @@ int dbg_force_in_the_gaps(void);
 #define ubi_leb_unmap  dbg_leb_unmap
 #define ubi_is_mapped  dbg_is_mapped
 #define ubi_leb_map    dbg_leb_map
-
 #endif
 
 int dbg_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset,
@@ -392,87 +318,127 @@ void dbg_debugfs_exit_fs(struct ubifs_info *c);
                        __func__, __LINE__, current->pid);                      \
 } while (0)
 
-#define dbg_err(fmt, ...)   do {                                               \
-        if (0)                                                                 \
-                ubifs_err(fmt, ##__VA_ARGS__);                                 \
+#define dbg_err(fmt, ...)   do {                   \
+        if (0)                                     \
+                ubifs_err(fmt, ##__VA_ARGS__);     \
 } while (0)
 
-#define dbg_msg(fmt, ...) do {                                                 \
-        if (0)                                                                 \
-                printk(KERN_DEBUG "UBIFS DBG (pid %d): %s: " fmt "\n",         \
-                       current->pid, __func__, ##__VA_ARGS__);                 \
+#define ubifs_dbg_msg(fmt, ...) do {               \
+        if (0)                                     \
+                pr_debug(fmt "\n", ##__VA_ARGS__); \
 } while (0)
 
 #define dbg_dump_stack()
 #define ubifs_assert_cmt_locked(c)
 
-#define dbg_gen(fmt, ...)   dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_jnl(fmt, ...)   dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_tnc(fmt, ...)   dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_lp(fmt, ...)    dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_find(fmt, ...)  dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_mnt(fmt, ...)   dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_io(fmt, ...)    dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_cmt(fmt, ...)   dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_budg(fmt, ...)  dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_log(fmt, ...)   dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_gc(fmt, ...)    dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_scan(fmt, ...)  dbg_msg(fmt, ##__VA_ARGS__)
-#define dbg_rcvry(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_msg(fmt, ...)   ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_gen(fmt, ...)   ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_jnl(fmt, ...)   ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_tnc(fmt, ...)   ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_lp(fmt, ...)    ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_find(fmt, ...)  ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_mnt(fmt, ...)   ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_io(fmt, ...)    ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_cmt(fmt, ...)   ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_budg(fmt, ...)  ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_log(fmt, ...)   ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_gc(fmt, ...)    ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_scan(fmt, ...)  ubifs_dbg_msg(fmt, ##__VA_ARGS__)
+#define dbg_rcvry(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__)
 
 #define DBGKEY(key)  ((char *)(key))
 #define DBGKEY1(key) ((char *)(key))
 
-#define ubifs_debugging_init(c)                0
-#define ubifs_debugging_exit(c)                ({})
-
-#define dbg_ntype(type)                        ""
-#define dbg_cstate(cmt_state)                  ""
-#define dbg_jhead(jhead)                       ""
-#define dbg_get_key_dump(c, key)               ({})
-#define dbg_dump_inode(c, inode)               ({})
-#define dbg_dump_node(c, node)                 ({})
-#define dbg_dump_lpt_node(c, node, lnum, offs) ({})
-#define dbg_dump_budget_req(req)               ({})
-#define dbg_dump_lstats(lst)                   ({})
-#define dbg_dump_budg(c)                       ({})
-#define dbg_dump_lprop(c, lp)                  ({})
-#define dbg_dump_lprops(c)                     ({})
-#define dbg_dump_lpt_info(c)                   ({})
-#define dbg_dump_leb(c, lnum)                  ({})
-#define dbg_dump_znode(c, znode)               ({})
-#define dbg_dump_heap(c, heap, cat)            ({})
-#define dbg_dump_pnode(c, pnode, parent, iip)  ({})
-#define dbg_dump_tnc(c)                        ({})
-#define dbg_dump_index(c)                      ({})
-#define dbg_dump_lpt_lebs(c)                   ({})
-
-#define dbg_walk_index(c, leaf_cb, znode_cb, priv) 0
-#define dbg_old_index_check_init(c, zroot)         0
-#define dbg_save_space_info(c)                     ({})
-#define dbg_check_space_info(c)                    0
-#define dbg_check_old_index(c, zroot)              0
-#define dbg_check_cats(c)                          0
-#define dbg_check_ltab(c)                          0
-#define dbg_chk_lpt_free_spc(c)                    0
-#define dbg_chk_lpt_sz(c, action, len)             0
-#define dbg_check_synced_i_size(inode)             0
-#define dbg_check_dir_size(c, dir)                 0
-#define dbg_check_tnc(c, x)                        0
-#define dbg_check_idx_size(c, idx_size)            0
-#define dbg_check_filesystem(c)                    0
-#define dbg_check_heap(c, heap, cat, add_pos)      ({})
-#define dbg_check_lprops(c)                        0
-#define dbg_check_lpt_nodes(c, cnode, row, col)    0
-#define dbg_check_inode_size(c, inode, size)       0
-#define dbg_force_in_the_gaps_enabled              0
-#define dbg_force_in_the_gaps()                    0
-#define dbg_failure_mode                           0
-
-#define dbg_debugfs_init()                         0
-#define dbg_debugfs_exit()
-#define dbg_debugfs_init_fs(c)                     0
-#define dbg_debugfs_exit_fs(c)                     0
+static inline int ubifs_debugging_init(struct ubifs_info *c)      { return 0; }
+static inline void ubifs_debugging_exit(struct ubifs_info *c)     { return; }
+static inline const char *dbg_ntype(int type)                     { return ""; }
+static inline const char *dbg_cstate(int cmt_state)               { return ""; }
+static inline const char *dbg_jhead(int jhead)                    { return ""; }
+static inline const char *
+dbg_get_key_dump(const struct ubifs_info *c,
+                 const union ubifs_key *key)                      { return ""; }
+static inline void dbg_dump_inode(const struct ubifs_info *c,
+                                  const struct inode *inode)      { return; }
+static inline void dbg_dump_node(const struct ubifs_info *c,
+                                 const void *node)                { return; }
+static inline void dbg_dump_lpt_node(const struct ubifs_info *c,
+                                     void *node, int lnum,
+                                     int offs)                    { return; }
+static inline void
+dbg_dump_budget_req(const struct ubifs_budget_req *req)           { return; }
+static inline void
+dbg_dump_lstats(const struct ubifs_lp_stats *lst)                 { return; }
+static inline void
+dbg_dump_budg(struct ubifs_info *c,
+              const struct ubifs_budg_info *bi)                   { return; }
+static inline void dbg_dump_lprop(const struct ubifs_info *c,
+                                  const struct ubifs_lprops *lp)  { return; }
+static inline void dbg_dump_lprops(struct ubifs_info *c)          { return; }
+static inline void dbg_dump_lpt_info(struct ubifs_info *c)        { return; }
+static inline void dbg_dump_leb(const struct ubifs_info *c,
+                                int lnum)                         { return; }
+static inline void
+dbg_dump_znode(const struct ubifs_info *c,
+               const struct ubifs_znode *znode)                   { return; }
+static inline void dbg_dump_heap(struct ubifs_info *c,
+                                 struct ubifs_lpt_heap *heap,
+                                 int cat)                         { return; }
+static inline void dbg_dump_pnode(struct ubifs_info *c,
+                                  struct ubifs_pnode *pnode,
+                                  struct ubifs_nnode *parent,
+                                  int iip)                        { return; }
+static inline void dbg_dump_tnc(struct ubifs_info *c)             { return; }
+static inline void dbg_dump_index(struct ubifs_info *c)           { return; }
+static inline void dbg_dump_lpt_lebs(const struct ubifs_info *c)  { return; }
+
+static inline int dbg_walk_index(struct ubifs_info *c,
+                                 dbg_leaf_callback leaf_cb,
+                                 dbg_znode_callback znode_cb,
+                                 void *priv)                      { return 0; }
+static inline void dbg_save_space_info(struct ubifs_info *c)      { return; }
+static inline int dbg_check_space_info(struct ubifs_info *c)      { return 0; }
+static inline int dbg_check_lprops(struct ubifs_info *c)          { return 0; }
+static inline int
+dbg_old_index_check_init(struct ubifs_info *c,
+                         struct ubifs_zbranch *zroot)             { return 0; }
+static inline int
+dbg_check_old_index(struct ubifs_info *c,
+                    struct ubifs_zbranch *zroot)                  { return 0; }
+static inline int dbg_check_cats(struct ubifs_info *c)            { return 0; }
+static inline int dbg_check_ltab(struct ubifs_info *c)            { return 0; }
+static inline int dbg_chk_lpt_free_spc(struct ubifs_info *c)      { return 0; }
+static inline int dbg_chk_lpt_sz(struct ubifs_info *c,
+                                 int action, int len)             { return 0; }
+static inline int dbg_check_synced_i_size(struct inode *inode)    { return 0; }
+static inline int dbg_check_dir_size(struct ubifs_info *c,
+                                     const struct inode *dir)     { return 0; }
+static inline int dbg_check_tnc(struct ubifs_info *c, int extra)  { return 0; }
+static inline int dbg_check_idx_size(struct ubifs_info *c,
+                                     long long idx_size)          { return 0; }
+static inline int dbg_check_filesystem(struct ubifs_info *c)      { return 0; }
+static inline void dbg_check_heap(struct ubifs_info *c,
+                                  struct ubifs_lpt_heap *heap,
+                                  int cat, int add_pos)           { return; }
+static inline int dbg_check_lpt_nodes(struct ubifs_info *c,
+        struct ubifs_cnode *cnode, int row, int col)              { return 0; }
+static inline int dbg_check_inode_size(struct ubifs_info *c,
+                                       const struct inode *inode,
+                                       loff_t size)               { return 0; }
+static inline int
+dbg_check_data_nodes_order(struct ubifs_info *c,
+                           struct list_head *head)                { return 0; }
+static inline int
+dbg_check_nondata_nodes_order(struct ubifs_info *c,
+                              struct list_head *head)             { return 0; }
+
+static inline int dbg_force_in_the_gaps(void)                     { return 0; }
+#define dbg_force_in_the_gaps_enabled() 0
+#define dbg_failure_mode                0
+
+static inline int dbg_debugfs_init(void)                          { return 0; }
+static inline void dbg_debugfs_exit(void)                         { return; }
+static inline int dbg_debugfs_init_fs(struct ubifs_info *c)       { return 0; }
+static inline int dbg_debugfs_exit_fs(struct ubifs_info *c)       { return 0; }
 
 #endif /* !CONFIG_UBIFS_FS_DEBUG */
 #endif /* !__UBIFS_DEBUG_H__ */
diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c
index 87ebcce72213..ef5abd38f0bf 100644
--- a/fs/ubifs/dir.c
+++ b/fs/ubifs/dir.c
@@ -522,24 +522,6 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
         ubifs_assert(mutex_is_locked(&dir->i_mutex));
         ubifs_assert(mutex_is_locked(&inode->i_mutex));
 
-        /*
-         * Return -ENOENT if we've raced with unlink and i_nlink is 0.  Doing
-         * otherwise has the potential to corrupt the orphan inode list.
-         *
-         * Indeed, consider a scenario when 'vfs_link(dirA/fileA)' and
-         * 'vfs_unlink(dirA/fileA, dirB/fileB)' race. 'vfs_link()' does not
-         * lock 'dirA->i_mutex', so this is possible. Both of the functions
-         * lock 'fileA->i_mutex' though. Suppose 'vfs_unlink()' wins, and takes
-         * 'fileA->i_mutex' mutex first. Suppose 'fileA->i_nlink' is 1. In this
-         * case 'ubifs_unlink()' will drop the last reference, and put 'inodeA'
-         * to the list of orphans. After this, 'vfs_link()' will link
-         * 'dirB/fileB' to 'inodeA'. This is a problem because, for example,
-         * the subsequent 'vfs_unlink(dirB/fileB)' will add the same inode
-         * to the list of orphans.
-         */
-         if (inode->i_nlink == 0)
-                 return -ENOENT;
-
         err = dbg_check_synced_i_size(inode);
         if (err)
                 return err;
@@ -550,7 +532,7 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
 
         lock_2_inodes(dir, inode);
         inc_nlink(inode);
-        atomic_inc(&inode->i_count);
+        ihold(inode);
         inode->i_ctime = ubifs_current_time(inode);
         dir->i_size += sz_change;
         dir_ui->ui_size = dir->i_size;
@@ -621,7 +603,7 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry)
                 ubifs_release_budget(c, &req);
         else {
                 /* We've deleted something - clean the "no space" flags */
-                c->nospace = c->nospace_rp = 0;
+                c->bi.nospace = c->bi.nospace_rp = 0;
                 smp_wmb();
         }
         return 0;
@@ -711,7 +693,7 @@ static int ubifs_rmdir(struct inode *dir, struct dentry *dentry)
                 ubifs_release_budget(c, &req);
         else {
                 /* We've deleted something - clean the "no space" flags */
-                c->nospace = c->nospace_rp = 0;
+                c->bi.nospace = c->bi.nospace_rp = 0;
                 smp_wmb();
         }
         return 0;
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 03ae894c45de..5e7fccfc4b29 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -212,7 +212,7 @@ static void release_new_page_budget(struct ubifs_info *c)
  */
 static void release_existing_page_budget(struct ubifs_info *c)
 {
-        struct ubifs_budget_req req = { .dd_growth = c->page_budget};
+        struct ubifs_budget_req req = { .dd_growth = c->bi.page_budget};
 
         ubifs_release_budget(c, &req);
 }
@@ -433,8 +433,9 @@ static int ubifs_write_begin(struct file *file, struct address_space *mapping,
         struct page *page;
 
         ubifs_assert(ubifs_inode(inode)->ui_size == inode->i_size);
+        ubifs_assert(!c->ro_media && !c->ro_mount);
 
-        if (unlikely(c->ro_media))
+        if (unlikely(c->ro_error))
                 return -EROFS;
 
         /* Try out the fast-path part first */
@@ -447,10 +448,12 @@ static int ubifs_write_begin(struct file *file, struct address_space *mapping,
                 if (!(pos & ~PAGE_CACHE_MASK) && len == PAGE_CACHE_SIZE) {
                         /*
                          * We change whole page so no need to load it. But we
-                         * have to set the @PG_checked flag to make the further
-                         * code know that the page is new. This might be not
-                         * true, but it is better to budget more than to read
-                         * the page from the media.
+                         * do not know whether this page exists on the media or
+                         * not, so we assume the latter because it requires
+                         * larger budget. The assumption is that it is better
+                         * to budget a bit more than to read the page from the
+                         * media. Thus, we are setting the @PG_checked flag
+                         * here.
                          */
                         SetPageChecked(page);
                         skipped_read = 1;
@@ -558,6 +561,7 @@ static int ubifs_write_end(struct file *file, struct address_space *mapping,
                 dbg_gen("copied %d instead of %d, read page and repeat",
                         copied, len);
                 cancel_budget(c, page, ui, appending);
+                ClearPageChecked(page);
 
                 /*
                  * Return 0 to force VFS to repeat the whole operation, or the
@@ -967,11 +971,11 @@ static int do_writepage(struct page *page, int len)
  * the page locked, and it locks @ui_mutex. However, write-back does take inode
  * @i_mutex, which means other VFS operations may be run on this inode at the
  * same time. And the problematic one is truncation to smaller size, from where
- * we have to call 'truncate_setsize()', which first changes @inode->i_size, then
- * drops the truncated pages. And while dropping the pages, it takes the page
- * lock. This means that 'do_truncation()' cannot call 'truncate_setsize()' with
- * @ui_mutex locked, because it would deadlock with 'ubifs_writepage()'. This
- * means that @inode->i_size is changed while @ui_mutex is unlocked.
+ * we have to call 'truncate_setsize()', which first changes @inode->i_size,
+ * then drops the truncated pages. And while dropping the pages, it takes the
+ * page lock. This means that 'do_truncation()' cannot call 'truncate_setsize()'
+ * with @ui_mutex locked, because it would deadlock with 'ubifs_writepage()'.
+ * This means that @inode->i_size is changed while @ui_mutex is unlocked.
  *
  * XXX(truncate): with the new truncate sequence this is not true anymore,
  * and the calls to truncate_setsize can be move around freely.  They should
@@ -1185,7 +1189,7 @@ out_budg:
         if (budgeted)
                 ubifs_release_budget(c, &req);
         else {
-                c->nospace = c->nospace_rp = 0;
+                c->bi.nospace = c->bi.nospace_rp = 0;
                 smp_wmb();
         }
         return err;
@@ -1308,6 +1312,13 @@ int ubifs_fsync(struct file *file, int datasync)
 
         dbg_gen("syncing inode %lu", inode->i_ino);
 
+        if (c->ro_mount)
+                /*
+                 * For some really strange reasons VFS does not filter out
+                 * 'fsync()' for R/O mounted file-systems as per 2.6.39.
+                 */
+                return 0;
+
         /*
          * VFS has already synchronized dirty pages for this inode. Synchronize
          * the inode unless this is a 'datasync()' call.
@@ -1425,10 +1436,11 @@ static int ubifs_releasepage(struct page *page, gfp_t unused_gfp_flags)
 }
 
 /*
- * mmap()d file has taken write protection fault and is being made
- * writable. UBIFS must ensure page is budgeted for.
+ * mmap()d file has taken write protection fault and is being made writable.
+ * UBIFS must ensure page is budgeted for.
  */
-static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
+static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma,
+                                 struct vm_fault *vmf)
 {
         struct page *page = vmf->page;
         struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
@@ -1439,9 +1451,9 @@ static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vm
 
         dbg_gen("ino %lu, pg %lu, i_size %lld", inode->i_ino, page->index,
                 i_size_read(inode));
-        ubifs_assert(!(inode->i_sb->s_flags & MS_RDONLY));
+        ubifs_assert(!c->ro_media && !c->ro_mount);
 
-        if (unlikely(c->ro_media))
+        if (unlikely(c->ro_error))
                 return VM_FAULT_SIGBUS; /* -EROFS */
 
         /*
@@ -1529,7 +1541,6 @@ static int ubifs_file_mmap(struct file *file, struct vm_area_struct *vma)
 {
         int err;
 
-        /* 'generic_file_mmap()' takes care of NOMMU case */
         err = generic_file_mmap(file, vma);
         if (err)
                 return err;
diff --git a/fs/ubifs/find.c b/fs/ubifs/find.c
index 1d54383d1269..2559d174e004 100644
--- a/fs/ubifs/find.c
+++ b/fs/ubifs/find.c
@@ -252,8 +252,8 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp,
                  * But if the index takes fewer LEBs than it is reserved for it,
                  * this function must avoid picking those reserved LEBs.
                  */
-                if (c->min_idx_lebs >= c->lst.idx_lebs) {
-                        rsvd_idx_lebs = c->min_idx_lebs -  c->lst.idx_lebs;
+                if (c->bi.min_idx_lebs >= c->lst.idx_lebs) {
+                        rsvd_idx_lebs = c->bi.min_idx_lebs -  c->lst.idx_lebs;
                         exclude_index = 1;
                 }
                 spin_unlock(&c->space_lock);
@@ -276,7 +276,7 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp,
                         pick_free = 0;
         } else {
                 spin_lock(&c->space_lock);
-                exclude_index = (c->min_idx_lebs >= c->lst.idx_lebs);
+                exclude_index = (c->bi.min_idx_lebs >= c->lst.idx_lebs);
                 spin_unlock(&c->space_lock);
         }
 
@@ -501,8 +501,8 @@ int ubifs_find_free_space(struct ubifs_info *c, int min_space, int *offs,
 
         /* Check if there are enough empty LEBs for commit */
         spin_lock(&c->space_lock);
-        if (c->min_idx_lebs > c->lst.idx_lebs)
-                rsvd_idx_lebs = c->min_idx_lebs -  c->lst.idx_lebs;
+        if (c->bi.min_idx_lebs > c->lst.idx_lebs)
+                rsvd_idx_lebs = c->bi.min_idx_lebs -  c->lst.idx_lebs;
         else
                 rsvd_idx_lebs = 0;
         lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt -
diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c
index 918d1582ca05..ded29f6224c2 100644
--- a/fs/ubifs/gc.c
+++ b/fs/ubifs/gc.c
@@ -100,6 +100,10 @@ static int switch_gc_head(struct ubifs_info *c)
         if (err)
                 return err;
 
+        err = ubifs_wbuf_sync_nolock(wbuf);
+        if (err)
+                return err;
+
         err = ubifs_add_bud_to_log(c, GCHD, gc_lnum, 0);
         if (err)
                 return err;
@@ -118,17 +122,23 @@ static int switch_gc_head(struct ubifs_info *c)
  * This function compares data nodes @a and @b. Returns %1 if @a has greater
  * inode or block number, and %-1 otherwise.
  */
-int data_nodes_cmp(void *priv, struct list_head *a, struct list_head *b)
+static int data_nodes_cmp(void *priv, struct list_head *a, struct list_head *b)
 {
         ino_t inuma, inumb;
         struct ubifs_info *c = priv;
         struct ubifs_scan_node *sa, *sb;
 
         cond_resched();
+        if (a == b)
+                return 0;
+
         sa = list_entry(a, struct ubifs_scan_node, list);
         sb = list_entry(b, struct ubifs_scan_node, list);
+
         ubifs_assert(key_type(c, &sa->key) == UBIFS_DATA_KEY);
         ubifs_assert(key_type(c, &sb->key) == UBIFS_DATA_KEY);
+        ubifs_assert(sa->type == UBIFS_DATA_NODE);
+        ubifs_assert(sb->type == UBIFS_DATA_NODE);
 
         inuma = key_inum(c, &sa->key);
         inumb = key_inum(c, &sb->key);
@@ -155,30 +165,43 @@ int data_nodes_cmp(void *priv, struct list_head *a, struct list_head *b)
  * first and sorted by length in descending order. Directory entry nodes go
  * after inode nodes and are sorted in ascending hash valuer order.
  */
-int nondata_nodes_cmp(void *priv, struct list_head *a, struct list_head *b)
+static int nondata_nodes_cmp(void *priv, struct list_head *a,
+                             struct list_head *b)
 {
-        int typea, typeb;
         ino_t inuma, inumb;
         struct ubifs_info *c = priv;
         struct ubifs_scan_node *sa, *sb;
 
         cond_resched();
+        if (a == b)
+                return 0;
+
         sa = list_entry(a, struct ubifs_scan_node, list);
         sb = list_entry(b, struct ubifs_scan_node, list);
-        typea = key_type(c, &sa->key);
-        typeb = key_type(c, &sb->key);
-        ubifs_assert(typea != UBIFS_DATA_KEY && typeb != UBIFS_DATA_KEY);
+
+        ubifs_assert(key_type(c, &sa->key) != UBIFS_DATA_KEY &&
+                     key_type(c, &sb->key) != UBIFS_DATA_KEY);
+        ubifs_assert(sa->type != UBIFS_DATA_NODE &&
+                     sb->type != UBIFS_DATA_NODE);
 
         /* Inodes go before directory entries */
-        if (typea == UBIFS_INO_KEY) {
-                if (typeb == UBIFS_INO_KEY)
+        if (sa->type == UBIFS_INO_NODE) {
+                if (sb->type == UBIFS_INO_NODE)
                         return sb->len - sa->len;
                 return -1;
         }
-        if (typeb == UBIFS_INO_KEY)
+        if (sb->type == UBIFS_INO_NODE)
                 return 1;
 
-        ubifs_assert(typea == UBIFS_DENT_KEY && typeb == UBIFS_DENT_KEY);
+        ubifs_assert(key_type(c, &sa->key) == UBIFS_DENT_KEY ||
+                     key_type(c, &sa->key) == UBIFS_XENT_KEY);
+        ubifs_assert(key_type(c, &sb->key) == UBIFS_DENT_KEY ||
+                     key_type(c, &sb->key) == UBIFS_XENT_KEY);
+        ubifs_assert(sa->type == UBIFS_DENT_NODE ||
+                     sa->type == UBIFS_XENT_NODE);
+        ubifs_assert(sb->type == UBIFS_DENT_NODE ||
+                     sb->type == UBIFS_XENT_NODE);
+
         inuma = key_inum(c, &sa->key);
         inumb = key_inum(c, &sb->key);
 
@@ -224,17 +247,33 @@ int nondata_nodes_cmp(void *priv, struct list_head *a, struct list_head *b)
 static int sort_nodes(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
                       struct list_head *nondata, int *min)
 {
+        int err;
         struct ubifs_scan_node *snod, *tmp;
 
         *min = INT_MAX;
 
         /* Separate data nodes and non-data nodes */
         list_for_each_entry_safe(snod, tmp, &sleb->nodes, list) {
-                int err;
+                ubifs_assert(snod->type == UBIFS_INO_NODE  ||
+                             snod->type == UBIFS_DATA_NODE ||
+                             snod->type == UBIFS_DENT_NODE ||
+                             snod->type == UBIFS_XENT_NODE ||
+                             snod->type == UBIFS_TRUN_NODE);
+
+                if (snod->type != UBIFS_INO_NODE  &&
+                    snod->type != UBIFS_DATA_NODE &&
+                    snod->type != UBIFS_DENT_NODE &&
+                    snod->type != UBIFS_XENT_NODE) {
+                        /* Probably truncation node, zap it */
+                        list_del(&snod->list);
+                        kfree(snod);
+                        continue;
+                }
 
-                ubifs_assert(snod->type != UBIFS_IDX_NODE);
-                ubifs_assert(snod->type != UBIFS_REF_NODE);
-                ubifs_assert(snod->type != UBIFS_CS_NODE);
+                ubifs_assert(key_type(c, &snod->key) == UBIFS_DATA_KEY ||
+                             key_type(c, &snod->key) == UBIFS_INO_KEY  ||
+                             key_type(c, &snod->key) == UBIFS_DENT_KEY ||
+                             key_type(c, &snod->key) == UBIFS_XENT_KEY);
 
                 err = ubifs_tnc_has_node(c, &snod->key, 0, sleb->lnum,
                                          snod->offs, 0);
@@ -258,6 +297,13 @@ static int sort_nodes(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
         /* Sort data and non-data nodes */
         list_sort(c, &sleb->nodes, &data_nodes_cmp);
         list_sort(c, nondata, &nondata_nodes_cmp);
+
+        err = dbg_check_data_nodes_order(c, &sleb->nodes);
+        if (err)
+                return err;
+        err = dbg_check_nondata_nodes_order(c, nondata);
+        if (err)
+                return err;
         return 0;
 }
 
@@ -432,6 +478,37 @@ int ubifs_garbage_collect_leb(struct ubifs_info *c, struct ubifs_lprops *lp)
         ubifs_assert(c->gc_lnum != lnum);
         ubifs_assert(wbuf->lnum != lnum);
 
+        if (lp->free + lp->dirty == c->leb_size) {
+                /* Special case - a free LEB  */
+                dbg_gc("LEB %d is free, return it", lp->lnum);
+                ubifs_assert(!(lp->flags & LPROPS_INDEX));
+
+                if (lp->free != c->leb_size) {
+                        /*
+                         * Write buffers must be sync'd before unmapping
+                         * freeable LEBs, because one of them may contain data
+                         * which obsoletes something in 'lp->pnum'.
+                         */
+                        err = gc_sync_wbufs(c);
+                        if (err)
+                                return err;
+                        err = ubifs_change_one_lp(c, lp->lnum, c->leb_size,
+                                                  0, 0, 0, 0);
+                        if (err)
+                                return err;
+                }
+                err = ubifs_leb_unmap(c, lp->lnum);
+                if (err)
+                        return err;
+
+                if (c->gc_lnum == -1) {
+                        c->gc_lnum = lnum;
+                        return LEB_RETAINED;
+                }
+
+                return LEB_FREED;
+        }
+
         /*
          * We scan the entire LEB even though we only really need to scan up to
          * (c->leb_size - lp->free).
@@ -575,13 +652,14 @@ int ubifs_garbage_collect(struct ubifs_info *c, int anyway)
         struct ubifs_wbuf *wbuf = &c->jheads[GCHD].wbuf;
 
         ubifs_assert_cmt_locked(c);
+        ubifs_assert(!c->ro_media && !c->ro_mount);
 
         if (ubifs_gc_should_commit(c))
                 return -EAGAIN;
 
         mutex_lock_nested(&wbuf->io_mutex, wbuf->jhead);
 
-        if (c->ro_media) {
+        if (c->ro_error) {
                 ret = -EROFS;
                 goto out_unlock;
         }
@@ -640,51 +718,18 @@ int ubifs_garbage_collect(struct ubifs_info *c, int anyway)
                        "(min. space %d)", lp.lnum, lp.free, lp.dirty,
                        lp.free + lp.dirty, min_space);
 
-                if (lp.free + lp.dirty == c->leb_size) {
-                        /* An empty LEB was returned */
-                        dbg_gc("LEB %d is free, return it", lp.lnum);
-                        /*
-                         * ubifs_find_dirty_leb() doesn't return freeable index
-                         * LEBs.
-                         */
-                        ubifs_assert(!(lp.flags & LPROPS_INDEX));
-                        if (lp.free != c->leb_size) {
-                                /*
-                                 * Write buffers must be sync'd before
-                                 * unmapping freeable LEBs, because one of them
-                                 * may contain data which obsoletes something
-                                 * in 'lp.pnum'.
-                                 */
-                                ret = gc_sync_wbufs(c);
-                                if (ret)
-                                        goto out;
-                                ret = ubifs_change_one_lp(c, lp.lnum,
-                                                          c->leb_size, 0, 0, 0,
-                                                          0);
-                                if (ret)
-                                        goto out;
-                        }
-                        ret = ubifs_leb_unmap(c, lp.lnum);
-                        if (ret)
-                                goto out;
-                        ret = lp.lnum;
-                        break;
-                }
-
                 space_before = c->leb_size - wbuf->offs - wbuf->used;
                 if (wbuf->lnum == -1)
                         space_before = 0;
 
                 ret = ubifs_garbage_collect_leb(c, &lp);
                 if (ret < 0) {
-                        if (ret == -EAGAIN || ret == -ENOSPC) {
+                        if (ret == -EAGAIN) {
                                 /*
-                                 * These codes are not errors, so we have to
-                                 * return the LEB to lprops. But if the
-                                 * 'ubifs_return_leb()' function fails, its
-                                 * failure code is propagated to the caller
-                                 * instead of the original '-EAGAIN' or
-                                 * '-ENOSPC'.
+                                 * This is not error, so we have to return the
+                                 * LEB to lprops. But if 'ubifs_return_leb()'
+                                 * fails, its failure code is propagated to the
+                                 * caller instead of the original '-EAGAIN'.
                                  */
                                 err = ubifs_return_leb(c, lp.lnum);
                                 if (err)
@@ -774,8 +819,8 @@ out_unlock:
 out:
         ubifs_assert(ret < 0);
         ubifs_assert(ret != -ENOSPC && ret != -EAGAIN);
-        ubifs_ro_mode(c, ret);
         ubifs_wbuf_sync_nolock(wbuf);
+        ubifs_ro_mode(c, ret);
         mutex_unlock(&wbuf->io_mutex);
         ubifs_return_leb(c, lp.lnum);
         return ret;
diff --git a/fs/ubifs/io.c b/fs/ubifs/io.c
index bcf5a16f30bb..3be645e012c9 100644
--- a/fs/ubifs/io.c
+++ b/fs/ubifs/io.c
@@ -31,6 +31,26 @@
  * buffer is full or when it is not used for some time (by timer). This is
  * similar to the mechanism is used by JFFS2.
  *
+ * UBIFS distinguishes between minimum write size (@c->min_io_size) and maximum
+ * write size (@c->max_write_size). The latter is the maximum amount of bytes
+ * the underlying flash is able to program at a time, and writing in
+ * @c->max_write_size units should presumably be faster. Obviously,
+ * @c->min_io_size <= @c->max_write_size. Write-buffers are of
+ * @c->max_write_size bytes in size for maximum performance. However, when a
+ * write-buffer is flushed, only the portion of it (aligned to @c->min_io_size
+ * boundary) which contains data is written, not the whole write-buffer,
+ * because this is more space-efficient.
+ *
+ * This optimization adds few complications to the code. Indeed, on the one
+ * hand, we want to write in optimal @c->max_write_size bytes chunks, which
+ * also means aligning writes at the @c->max_write_size bytes offsets. On the
+ * other hand, we do not want to waste space when synchronizing the write
+ * buffer, so during synchronization we writes in smaller chunks. And this makes
+ * the next write offset to be not aligned to @c->max_write_size bytes. So the
+ * have to make sure that the write-buffer offset (@wbuf->offs) becomes aligned
+ * to @c->max_write_size bytes again. We do this by temporarily shrinking
+ * write-buffer size (@wbuf->size).
+ *
  * Write-buffers are defined by 'struct ubifs_wbuf' objects and protected by
  * mutexes defined inside these objects. Since sometimes upper-level code
  * has to lock the write-buffer (e.g. journal space reservation code), many
@@ -46,8 +66,8 @@
  * UBIFS uses padding when it pads to the next min. I/O unit. In this case it
  * uses padding nodes or padding bytes, if the padding node does not fit.
  *
- * All UBIFS nodes are protected by CRC checksums and UBIFS checks all nodes
- * every time they are read from the flash media.
+ * All UBIFS nodes are protected by CRC checksums and UBIFS checks CRC when
+ * they are read from the flash media.
  */
 
 #include <linux/crc32.h>
@@ -61,8 +81,8 @@
  */
 void ubifs_ro_mode(struct ubifs_info *c, int err)
 {
-        if (!c->ro_media) {
-                c->ro_media = 1;
+        if (!c->ro_error) {
+                c->ro_error = 1;
                 c->no_chk_data_crc = 0;
                 c->vfs_sb->s_flags |= MS_RDONLY;
                 ubifs_warn("switched to read-only mode, error %d", err);
@@ -88,8 +108,12 @@ void ubifs_ro_mode(struct ubifs_info *c, int err)
  * This function may skip data nodes CRC checking if @c->no_chk_data_crc is
  * true, which is controlled by corresponding UBIFS mount option. However, if
  * @must_chk_crc is true, then @c->no_chk_data_crc is ignored and CRC is
- * checked. Similarly, if @c->always_chk_crc is true, @c->no_chk_data_crc is
- * ignored and CRC is checked.
+ * checked. Similarly, if @c->mounting or @c->remounting_rw is true (we are
+ * mounting or re-mounting to R/W mode), @c->no_chk_data_crc is ignored and CRC
+ * is checked. This is because during mounting or re-mounting from R/O mode to
+ * R/W mode we may read journal nodes (when replying the journal or doing the
+ * recovery) and the journal nodes may potentially be corrupted, so checking is
+ * required.
  *
  * This function returns zero in case of success and %-EUCLEAN in case of bad
  * CRC or magic.
@@ -131,8 +155,8 @@ int ubifs_check_node(const struct ubifs_info *c, const void *buf, int lnum,
                    node_len > c->ranges[type].max_len)
                 goto out_len;
 
-        if (!must_chk_crc && type == UBIFS_DATA_NODE && !c->always_chk_crc &&
-             c->no_chk_data_crc)
+        if (!must_chk_crc && type == UBIFS_DATA_NODE && !c->mounting &&
+            !c->remounting_rw && c->no_chk_data_crc)
                 return 0;
 
         crc = crc32(UBIFS_CRC32_INIT, buf + 8, node_len - 8);
@@ -343,11 +367,17 @@ static void cancel_wbuf_timer_nolock(struct ubifs_wbuf *wbuf)
  *
  * This function synchronizes write-buffer @buf and returns zero in case of
  * success or a negative error code in case of failure.
+ *
+ * Note, although write-buffers are of @c->max_write_size, this function does
+ * not necessarily writes all @c->max_write_size bytes to the flash. Instead,
+ * if the write-buffer is only partially filled with data, only the used part
+ * of the write-buffer (aligned on @c->min_io_size boundary) is synchronized.
+ * This way we waste less space.
  */
 int ubifs_wbuf_sync_nolock(struct ubifs_wbuf *wbuf)
 {
         struct ubifs_info *c = wbuf->c;
-        int err, dirt;
+        int err, dirt, sync_len;
 
         cancel_wbuf_timer_nolock(wbuf);
         if (!wbuf->used || wbuf->lnum == -1)
@@ -356,28 +386,54 @@ int ubifs_wbuf_sync_nolock(struct ubifs_wbuf *wbuf)
 
         dbg_io("LEB %d:%d, %d bytes, jhead %s",
                wbuf->lnum, wbuf->offs, wbuf->used, dbg_jhead(wbuf->jhead));
-        ubifs_assert(!(c->vfs_sb->s_flags & MS_RDONLY));
         ubifs_assert(!(wbuf->avail & 7));
-        ubifs_assert(wbuf->offs + c->min_io_size <= c->leb_size);
-
-        if (c->ro_media)
+        ubifs_assert(wbuf->offs + wbuf->size <= c->leb_size);
+        ubifs_assert(wbuf->size >= c->min_io_size);
+        ubifs_assert(wbuf->size <= c->max_write_size);
+        ubifs_assert(wbuf->size % c->min_io_size == 0);
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        if (c->leb_size - wbuf->offs >= c->max_write_size)
+                ubifs_assert(!((wbuf->offs + wbuf->size) % c->max_write_size));
+
+        if (c->ro_error)
                 return -EROFS;
 
-        ubifs_pad(c, wbuf->buf + wbuf->used, wbuf->avail);
+        /*
+         * Do not write whole write buffer but write only the minimum necessary
+         * amount of min. I/O units.
+         */
+        sync_len = ALIGN(wbuf->used, c->min_io_size);
+        dirt = sync_len - wbuf->used;
+        if (dirt)
+                ubifs_pad(c, wbuf->buf + wbuf->used, dirt);
         err = ubi_leb_write(c->ubi, wbuf->lnum, wbuf->buf, wbuf->offs,
-                            c->min_io_size, wbuf->dtype);
+                            sync_len, wbuf->dtype);
         if (err) {
                 ubifs_err("cannot write %d bytes to LEB %d:%d",
-                          c->min_io_size, wbuf->lnum, wbuf->offs);
+                          sync_len, wbuf->lnum, wbuf->offs);
                 dbg_dump_stack();
                 return err;
         }
 
-        dirt = wbuf->avail;
-
         spin_lock(&wbuf->lock);
-        wbuf->offs += c->min_io_size;
-        wbuf->avail = c->min_io_size;
+        wbuf->offs += sync_len;
+        /*
+         * Now @wbuf->offs is not necessarily aligned to @c->max_write_size.
+         * But our goal is to optimize writes and make sure we write in
+         * @c->max_write_size chunks and to @c->max_write_size-aligned offset.
+         * Thus, if @wbuf->offs is not aligned to @c->max_write_size now, make
+         * sure that @wbuf->offs + @wbuf->size is aligned to
+         * @c->max_write_size. This way we make sure that after next
+         * write-buffer flush we are again at the optimal offset (aligned to
+         * @c->max_write_size).
+         */
+        if (c->leb_size - wbuf->offs < c->max_write_size)
+                wbuf->size = c->leb_size - wbuf->offs;
+        else if (wbuf->offs & (c->max_write_size - 1))
+                wbuf->size = ALIGN(wbuf->offs, c->max_write_size) - wbuf->offs;
+        else
+                wbuf->size = c->max_write_size;
+        wbuf->avail = wbuf->size;
         wbuf->used = 0;
         wbuf->next_ino = 0;
         spin_unlock(&wbuf->lock);
@@ -396,8 +452,8 @@ int ubifs_wbuf_sync_nolock(struct ubifs_wbuf *wbuf)
  * @dtype: data type
  *
  * This function targets the write-buffer to logical eraseblock @lnum:@offs.
- * The write-buffer is synchronized if it is not empty. Returns zero in case of
- * success and a negative error code in case of failure.
+ * The write-buffer has to be empty. Returns zero in case of success and a
+ * negative error code in case of failure.
  */
 int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs,
                            int dtype)
@@ -409,18 +465,18 @@ int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs,
         ubifs_assert(offs >= 0 && offs <= c->leb_size);
         ubifs_assert(offs % c->min_io_size == 0 && !(offs & 7));
         ubifs_assert(lnum != wbuf->lnum);
-
-        if (wbuf->used > 0) {
-                int err = ubifs_wbuf_sync_nolock(wbuf);
-
-                if (err)
-                        return err;
-        }
+        ubifs_assert(wbuf->used == 0);
 
         spin_lock(&wbuf->lock);
         wbuf->lnum = lnum;
         wbuf->offs = offs;
-        wbuf->avail = c->min_io_size;
+        if (c->leb_size - wbuf->offs < c->max_write_size)
+                wbuf->size = c->leb_size - wbuf->offs;
+        else if (wbuf->offs & (c->max_write_size - 1))
+                wbuf->size = ALIGN(wbuf->offs, c->max_write_size) - wbuf->offs;
+        else
+                wbuf->size = c->max_write_size;
+        wbuf->avail = wbuf->size;
         wbuf->used = 0;
         spin_unlock(&wbuf->lock);
         wbuf->dtype = dtype;
@@ -440,11 +496,12 @@ int ubifs_bg_wbufs_sync(struct ubifs_info *c)
 {
         int err, i;
 
+        ubifs_assert(!c->ro_media && !c->ro_mount);
         if (!c->need_wbuf_sync)
                 return 0;
         c->need_wbuf_sync = 0;
 
-        if (c->ro_media) {
+        if (c->ro_error) {
                 err = -EROFS;
                 goto out_timers;
         }
@@ -499,8 +556,9 @@ out_timers:
  *
  * This function writes data to flash via write-buffer @wbuf. This means that
  * the last piece of the node won't reach the flash media immediately if it
- * does not take whole minimal I/O unit. Instead, the node will sit in RAM
- * until the write-buffer is synchronized (e.g., by timer).
+ * does not take whole max. write unit (@c->max_write_size). Instead, the node
+ * will sit in RAM until the write-buffer is synchronized (e.g., by timer, or
+ * because more data are appended to the write-buffer).
  *
  * This function returns zero in case of success and a negative error code in
  * case of failure. If the node cannot be written because there is no more
@@ -509,7 +567,7 @@ out_timers:
 int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len)
 {
         struct ubifs_info *c = wbuf->c;
-        int err, written, n, aligned_len = ALIGN(len, 8), offs;
+        int err, written, n, aligned_len = ALIGN(len, 8);
 
         dbg_io("%d bytes (%s) to jhead %s wbuf at LEB %d:%d", len,
                dbg_ntype(((struct ubifs_ch *)buf)->node_type),
@@ -517,8 +575,15 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len)
         ubifs_assert(len > 0 && wbuf->lnum >= 0 && wbuf->lnum < c->leb_cnt);
         ubifs_assert(wbuf->offs >= 0 && wbuf->offs % c->min_io_size == 0);
         ubifs_assert(!(wbuf->offs & 7) && wbuf->offs <= c->leb_size);
-        ubifs_assert(wbuf->avail > 0 && wbuf->avail <= c->min_io_size);
+        ubifs_assert(wbuf->avail > 0 && wbuf->avail <= wbuf->size);
+        ubifs_assert(wbuf->size >= c->min_io_size);
+        ubifs_assert(wbuf->size <= c->max_write_size);
+        ubifs_assert(wbuf->size % c->min_io_size == 0);
         ubifs_assert(mutex_is_locked(&wbuf->io_mutex));
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        ubifs_assert(!c->space_fixup);
+        if (c->leb_size - wbuf->offs >= c->max_write_size)
+                ubifs_assert(!((wbuf->offs + wbuf->size) % c->max_write_size));
 
         if (c->leb_size - wbuf->offs - wbuf->used < aligned_len) {
                 err = -ENOSPC;
@@ -527,7 +592,7 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len)
 
         cancel_wbuf_timer_nolock(wbuf);
 
-        if (c->ro_media)
+        if (c->ro_error)
                 return -EROFS;
 
         if (aligned_len <= wbuf->avail) {
@@ -541,14 +606,18 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len)
                         dbg_io("flush jhead %s wbuf to LEB %d:%d",
                                dbg_jhead(wbuf->jhead), wbuf->lnum, wbuf->offs);
                         err = ubi_leb_write(c->ubi, wbuf->lnum, wbuf->buf,
-                                            wbuf->offs, c->min_io_size,
+                                            wbuf->offs, wbuf->size,
                                             wbuf->dtype);
                         if (err)
                                 goto out;
 
                         spin_lock(&wbuf->lock);
-                        wbuf->offs += c->min_io_size;
-                        wbuf->avail = c->min_io_size;
+                        wbuf->offs += wbuf->size;
+                        if (c->leb_size - wbuf->offs >= c->max_write_size)
+                                wbuf->size = c->max_write_size;
+                        else
+                                wbuf->size = c->leb_size - wbuf->offs;
+                        wbuf->avail = wbuf->size;
                         wbuf->used = 0;
                         wbuf->next_ino = 0;
                         spin_unlock(&wbuf->lock);
@@ -562,39 +631,63 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len)
                 goto exit;
         }
 
-        /*
-         * The node is large enough and does not fit entirely within current
-         * minimal I/O unit. We have to fill and flush write-buffer and switch
-         * to the next min. I/O unit.
-         */
-        dbg_io("flush jhead %s wbuf to LEB %d:%d",
-               dbg_jhead(wbuf->jhead), wbuf->lnum, wbuf->offs);
-        memcpy(wbuf->buf + wbuf->used, buf, wbuf->avail);
-        err = ubi_leb_write(c->ubi, wbuf->lnum, wbuf->buf, wbuf->offs,
-                            c->min_io_size, wbuf->dtype);
-        if (err)
-                goto out;
+        written = 0;
+
+        if (wbuf->used) {
+                /*
+                 * The node is large enough and does not fit entirely within
+                 * current available space. We have to fill and flush
+                 * write-buffer and switch to the next max. write unit.
+                 */
+                dbg_io("flush jhead %s wbuf to LEB %d:%d",
+                       dbg_jhead(wbuf->jhead), wbuf->lnum, wbuf->offs);
+                memcpy(wbuf->buf + wbuf->used, buf, wbuf->avail);
+                err = ubi_leb_write(c->ubi, wbuf->lnum, wbuf->buf, wbuf->offs,
+                                    wbuf->size, wbuf->dtype);
+                if (err)
+                        goto out;
+
+                wbuf->offs += wbuf->size;
+                len -= wbuf->avail;
+                aligned_len -= wbuf->avail;
+                written += wbuf->avail;
+        } else if (wbuf->offs & (c->max_write_size - 1)) {
+                /*
+                 * The write-buffer offset is not aligned to
+                 * @c->max_write_size and @wbuf->size is less than
+                 * @c->max_write_size. Write @wbuf->size bytes to make sure the
+                 * following writes are done in optimal @c->max_write_size
+                 * chunks.
+                 */
+                dbg_io("write %d bytes to LEB %d:%d",
+                       wbuf->size, wbuf->lnum, wbuf->offs);
+                err = ubi_leb_write(c->ubi, wbuf->lnum, buf, wbuf->offs,
+                                    wbuf->size, wbuf->dtype);
+                if (err)
+                        goto out;
 
-        offs = wbuf->offs + c->min_io_size;
-        len -= wbuf->avail;
-        aligned_len -= wbuf->avail;
-        written = wbuf->avail;
+                wbuf->offs += wbuf->size;
+                len -= wbuf->size;
+                aligned_len -= wbuf->size;
+                written += wbuf->size;
+        }
 
         /*
-         * The remaining data may take more whole min. I/O units, so write the
-         * remains multiple to min. I/O unit size directly to the flash media.
+         * The remaining data may take more whole max. write units, so write the
+         * remains multiple to max. write unit size directly to the flash media.
          * We align node length to 8-byte boundary because we anyway flash wbuf
          * if the remaining space is less than 8 bytes.
          */
-        n = aligned_len >> c->min_io_shift;
+        n = aligned_len >> c->max_write_shift;
         if (n) {
-                n <<= c->min_io_shift;
-                dbg_io("write %d bytes to LEB %d:%d", n, wbuf->lnum, offs);
-                err = ubi_leb_write(c->ubi, wbuf->lnum, buf + written, offs, n,
-                                    wbuf->dtype);
+                n <<= c->max_write_shift;
+                dbg_io("write %d bytes to LEB %d:%d", n, wbuf->lnum,
+                       wbuf->offs);
+                err = ubi_leb_write(c->ubi, wbuf->lnum, buf + written,
+                                    wbuf->offs, n, wbuf->dtype);
                 if (err)
                         goto out;
-                offs += n;
+                wbuf->offs += n;
                 aligned_len -= n;
                 len -= n;
                 written += n;
@@ -604,14 +697,17 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len)
         if (aligned_len)
                 /*
                  * And now we have what's left and what does not take whole
-                 * min. I/O unit, so write it to the write-buffer and we are
+                 * max. write unit, so write it to the write-buffer and we are
                  * done.
                  */
                 memcpy(wbuf->buf, buf + written, len);
 
-        wbuf->offs = offs;
+        if (c->leb_size - wbuf->offs >= c->max_write_size)
+                wbuf->size = c->max_write_size;
+        else
+                wbuf->size = c->leb_size - wbuf->offs;
+        wbuf->avail = wbuf->size - aligned_len;
         wbuf->used = aligned_len;
-        wbuf->avail = c->min_io_size - aligned_len;
         wbuf->next_ino = 0;
         spin_unlock(&wbuf->lock);
 
@@ -663,8 +759,10 @@ int ubifs_write_node(struct ubifs_info *c, void *buf, int len, int lnum,
                buf_len);
         ubifs_assert(lnum >= 0 && lnum < c->leb_cnt && offs >= 0);
         ubifs_assert(offs % c->min_io_size == 0 && offs < c->leb_size);
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        ubifs_assert(!c->space_fixup);
 
-        if (c->ro_media)
+        if (c->ro_error)
                 return -EROFS;
 
         ubifs_prepare_node(c, buf, len, 1);
@@ -815,7 +913,8 @@ int ubifs_read_node(const struct ubifs_info *c, void *buf, int type, int len,
         return 0;
 
 out:
-        ubifs_err("bad node at LEB %d:%d", lnum, offs);
+        ubifs_err("bad node at LEB %d:%d, LEB mapping status %d", lnum, offs,
+                  ubi_is_mapped(c->ubi, lnum));
         dbg_dump_node(c, buf);
         dbg_dump_stack();
         return -EINVAL;
@@ -833,11 +932,11 @@ int ubifs_wbuf_init(struct ubifs_info *c, struct ubifs_wbuf *wbuf)
 {
         size_t size;
 
-        wbuf->buf = kmalloc(c->min_io_size, GFP_KERNEL);
+        wbuf->buf = kmalloc(c->max_write_size, GFP_KERNEL);
         if (!wbuf->buf)
                 return -ENOMEM;
 
-        size = (c->min_io_size / UBIFS_CH_SZ + 1) * sizeof(ino_t);
+        size = (c->max_write_size / UBIFS_CH_SZ + 1) * sizeof(ino_t);
         wbuf->inodes = kmalloc(size, GFP_KERNEL);
         if (!wbuf->inodes) {
                 kfree(wbuf->buf);
@@ -847,7 +946,14 @@ int ubifs_wbuf_init(struct ubifs_info *c, struct ubifs_wbuf *wbuf)
 
         wbuf->used = 0;
         wbuf->lnum = wbuf->offs = -1;
-        wbuf->avail = c->min_io_size;
+        /*
+         * If the LEB starts at the max. write size aligned address, then
+         * write-buffer size has to be set to @c->max_write_size. Otherwise,
+         * set it to something smaller so that it ends at the closest max.
+         * write size boundary.
+         */
+        size = c->max_write_size - (c->leb_start % c->max_write_size);
+        wbuf->avail = wbuf->size = size;
         wbuf->dtype = UBI_UNKNOWN;
         wbuf->sync_callback = NULL;
         mutex_init(&wbuf->io_mutex);
diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c
index 8aacd64957a2..548acf494afd 100644
--- a/fs/ubifs/ioctl.c
+++ b/fs/ubifs/ioctl.c
@@ -160,7 +160,7 @@ long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
                 if (IS_RDONLY(inode))
                         return -EROFS;
 
-                if (!is_owner_or_cap(inode))
+                if (!inode_owner_or_capable(inode))
                         return -EACCES;
 
                 if (get_user(flags, (int __user *) arg))
diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c
index d321baeca68d..cef0460f4c54 100644
--- a/fs/ubifs/journal.c
+++ b/fs/ubifs/journal.c
@@ -122,11 +122,12 @@ static int reserve_space(struct ubifs_info *c, int jhead, int len)
          * better to try to allocate space at the ends of eraseblocks. This is
          * what the squeeze parameter does.
          */
+        ubifs_assert(!c->ro_media && !c->ro_mount);
         squeeze = (jhead == BASEHD);
 again:
         mutex_lock_nested(&wbuf->io_mutex, wbuf->jhead);
 
-        if (c->ro_media) {
+        if (c->ro_error) {
                 err = -EROFS;
                 goto out_unlock;
         }
@@ -140,14 +141,8 @@ again:
          * LEB with some empty space.
          */
         lnum = ubifs_find_free_space(c, len, &offs, squeeze);
-        if (lnum >= 0) {
-                /* Found an LEB, add it to the journal head */
-                err = ubifs_add_bud_to_log(c, jhead, lnum, offs);
-                if (err)
-                        goto out_return;
-                /* A new bud was successfully allocated and added to the log */
+        if (lnum >= 0)
                 goto out;
-        }
 
         err = lnum;
         if (err != -ENOSPC)
@@ -202,12 +197,23 @@ again:
                 return 0;
         }
 
-        err = ubifs_add_bud_to_log(c, jhead, lnum, 0);
-        if (err)
-                goto out_return;
         offs = 0;
 
 out:
+        /*
+         * Make sure we synchronize the write-buffer before we add the new bud
+         * to the log. Otherwise we may have a power cut after the log
+         * reference node for the last bud (@lnum) is written but before the
+         * write-buffer data are written to the next-to-last bud
+         * (@wbuf->lnum). And the effect would be that the recovery would see
+         * that there is corruption in the next-to-last bud.
+         */
+        err = ubifs_wbuf_sync_nolock(wbuf);
+        if (err)
+                goto out_return;
+        err = ubifs_add_bud_to_log(c, jhead, lnum, offs);
+        if (err)
+                goto out_return;
         err = ubifs_wbuf_seek_nolock(wbuf, lnum, offs, wbuf->dtype);
         if (err)
                 goto out_unlock;
@@ -379,10 +385,8 @@ out:
         if (err == -ENOSPC) {
                 /* This are some budgeting problems, print useful information */
                 down_write(&c->commit_sem);
-                spin_lock(&c->space_lock);
                 dbg_dump_stack();
-                dbg_dump_budg(c);
-                spin_unlock(&c->space_lock);
+                dbg_dump_budg(c, &c->bi);
                 dbg_dump_lprops(c);
                 cmt_retries = dbg_check_lprops(c);
                 up_write(&c->commit_sem);
@@ -665,6 +669,7 @@ out_free:
 
 out_release:
         release_head(c, BASEHD);
+        kfree(dent);
 out_ro:
         ubifs_ro_mode(c, err);
         if (last_reference)
@@ -689,7 +694,7 @@ int ubifs_jnl_write_data(struct ubifs_info *c, const struct inode *inode,
 {
         struct ubifs_data_node *data;
         int err, lnum, offs, compr_type, out_len;
-        int dlen = UBIFS_DATA_NODE_SZ + UBIFS_BLOCK_SIZE * WORST_COMPR_FACTOR;
+        int dlen = COMPRESSED_DATA_NODE_BUF_SZ, allocated = 1;
         struct ubifs_inode *ui = ubifs_inode(inode);
 
         dbg_jnl("ino %lu, blk %u, len %d, key %s",
@@ -697,9 +702,19 @@ int ubifs_jnl_write_data(struct ubifs_info *c, const struct inode *inode,
                 DBGKEY(key));
         ubifs_assert(len <= UBIFS_BLOCK_SIZE);
 
-        data = kmalloc(dlen, GFP_NOFS);
-        if (!data)
-                return -ENOMEM;
+        data = kmalloc(dlen, GFP_NOFS | __GFP_NOWARN);
+        if (!data) {
+                /*
+                 * Fall-back to the write reserve buffer. Note, we might be
+                 * currently on the memory reclaim path, when the kernel is
+                 * trying to free some memory by writing out dirty pages. The
+                 * write reserve buffer helps us to guarantee that we are
+                 * always able to write the data.
+                 */
+                allocated = 0;
+                mutex_lock(&c->write_reserve_mutex);
+                data = c->write_reserve_buf;
+        }
 
         data->ch.node_type = UBIFS_DATA_NODE;
         key_write(c, key, &data->key);
@@ -735,7 +750,10 @@ int ubifs_jnl_write_data(struct ubifs_info *c, const struct inode *inode,
                 goto out_ro;
 
         finish_reservation(c);
-        kfree(data);
+        if (!allocated)
+                mutex_unlock(&c->write_reserve_mutex);
+        else
+                kfree(data);
         return 0;
 
 out_release:
@@ -744,7 +762,10 @@ out_ro:
         ubifs_ro_mode(c, err);
         finish_reservation(c);
 out_free:
-        kfree(data);
+        if (!allocated)
+                mutex_unlock(&c->write_reserve_mutex);
+        else
+                kfree(data);
         return err;
 }
 
diff --git a/fs/ubifs/key.h b/fs/ubifs/key.h
index 0f530c684f0b..92a8491a8f8c 100644
--- a/fs/ubifs/key.h
+++ b/fs/ubifs/key.h
@@ -306,6 +306,20 @@ static inline void trun_key_init(const struct ubifs_info *c,
 }
 
 /**
+ * invalid_key_init - initialize invalid node key.
+ * @c: UBIFS file-system description object
+ * @key: key to initialize
+ *
+ * This is a helper function which marks a @key object as invalid.
+ */
+static inline void invalid_key_init(const struct ubifs_info *c,
+                                    union ubifs_key *key)
+{
+        key->u32[0] = 0xDEADBEAF;
+        key->u32[1] = UBIFS_INVALID_KEY;
+}
+
+/**
  * key_type - get key type.
  * @c: UBIFS file-system description object
  * @key: key to get type of
diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c
index c345e125f42c..affea9494ae2 100644
--- a/fs/ubifs/log.c
+++ b/fs/ubifs/log.c
@@ -100,20 +100,6 @@ struct ubifs_wbuf *ubifs_get_wbuf(struct ubifs_info *c, int lnum)
 }
 
 /**
- * next_log_lnum - switch to the next log LEB.
- * @c: UBIFS file-system description object
- * @lnum: current log LEB
- */
-static inline int next_log_lnum(const struct ubifs_info *c, int lnum)
-{
-        lnum += 1;
-        if (lnum > c->log_last)
-                lnum = UBIFS_LOG_LNUM;
-
-        return lnum;
-}
-
-/**
  * empty_log_bytes - calculate amount of empty space in the log.
  * @c: UBIFS file-system description object
  */
@@ -159,7 +145,7 @@ void ubifs_add_bud(struct ubifs_info *c, struct ubifs_bud *bud)
                 jhead = &c->jheads[bud->jhead];
                 list_add_tail(&bud->list, &jhead->buds_list);
         } else
-                ubifs_assert(c->replaying && (c->vfs_sb->s_flags & MS_RDONLY));
+                ubifs_assert(c->replaying && c->ro_mount);
 
         /*
          * Note, although this is a new bud, we anyway account this space now,
@@ -175,26 +161,6 @@ void ubifs_add_bud(struct ubifs_info *c, struct ubifs_bud *bud)
 }
 
 /**
- * ubifs_create_buds_lists - create journal head buds lists for remount rw.
- * @c: UBIFS file-system description object
- */
-void ubifs_create_buds_lists(struct ubifs_info *c)
-{
-        struct rb_node *p;
-
-        spin_lock(&c->buds_lock);
-        p = rb_first(&c->buds);
-        while (p) {
-                struct ubifs_bud *bud = rb_entry(p, struct ubifs_bud, rb);
-                struct ubifs_jhead *jhead = &c->jheads[bud->jhead];
-
-                list_add_tail(&bud->list, &jhead->buds_list);
-                p = rb_next(p);
-        }
-        spin_unlock(&c->buds_lock);
-}
-
-/**
  * ubifs_add_bud_to_log - add a new bud to the log.
  * @c: UBIFS file-system description object
  * @jhead: journal head the bud belongs to
@@ -223,8 +189,8 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs)
         }
 
         mutex_lock(&c->log_mutex);
-
-        if (c->ro_media) {
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        if (c->ro_error) {
                 err = -EROFS;
                 goto out_unlock;
         }
@@ -277,7 +243,7 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs)
         ref->jhead = cpu_to_le32(jhead);
 
         if (c->lhead_offs > c->leb_size - c->ref_node_alsz) {
-                c->lhead_lnum = next_log_lnum(c, c->lhead_lnum);
+                c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum);
                 c->lhead_offs = 0;
         }
 
@@ -445,7 +411,7 @@ int ubifs_log_start_commit(struct ubifs_info *c, int *ltail_lnum)
 
         /* Switch to the next log LEB */
         if (c->lhead_offs) {
-                c->lhead_lnum = next_log_lnum(c, c->lhead_lnum);
+                c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum);
                 c->lhead_offs = 0;
         }
 
@@ -466,7 +432,7 @@ int ubifs_log_start_commit(struct ubifs_info *c, int *ltail_lnum)
 
         c->lhead_offs += len;
         if (c->lhead_offs == c->leb_size) {
-                c->lhead_lnum = next_log_lnum(c, c->lhead_lnum);
+                c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum);
                 c->lhead_offs = 0;
         }
 
@@ -553,7 +519,7 @@ int ubifs_log_post_commit(struct ubifs_info *c, int old_ltail_lnum)
         }
         mutex_lock(&c->log_mutex);
         for (lnum = old_ltail_lnum; lnum != c->ltail_lnum;
-             lnum = next_log_lnum(c, lnum)) {
+             lnum = ubifs_next_log_lnum(c, lnum)) {
                 dbg_log("unmap log LEB %d", lnum);
                 err = ubifs_leb_unmap(c, lnum);
                 if (err)
@@ -662,7 +628,7 @@ static int add_node(struct ubifs_info *c, void *buf, int *lnum, int *offs,
                 err = ubifs_leb_change(c, *lnum, buf, sz, UBI_SHORTTERM);
                 if (err)
                         return err;
-                *lnum = next_log_lnum(c, *lnum);
+                *lnum = ubifs_next_log_lnum(c, *lnum);
                 *offs = 0;
         }
         memcpy(buf + *offs, node, len);
@@ -732,7 +698,7 @@ int ubifs_consolidate_log(struct ubifs_info *c)
                 ubifs_scan_destroy(sleb);
                 if (lnum == c->lhead_lnum)
                         break;
-                lnum = next_log_lnum(c, lnum);
+                lnum = ubifs_next_log_lnum(c, lnum);
         }
         if (offs) {
                 int sz = ALIGN(offs, c->min_io_size);
@@ -752,7 +718,7 @@ int ubifs_consolidate_log(struct ubifs_info *c)
         /* Unmap remaining LEBs */
         lnum = write_lnum;
         do {
-                lnum = next_log_lnum(c, lnum);
+                lnum = ubifs_next_log_lnum(c, lnum);
                 err = ubifs_leb_unmap(c, lnum);
                 if (err)
                         return err;
diff --git a/fs/ubifs/lprops.c b/fs/ubifs/lprops.c
index 4d4ca388889b..667884f4a615 100644
--- a/fs/ubifs/lprops.c
+++ b/fs/ubifs/lprops.c
@@ -1007,21 +1007,11 @@ out:
 }
 
 /**
- * struct scan_check_data - data provided to scan callback function.
- * @lst: LEB properties statistics
- * @err: error code
- */
-struct scan_check_data {
-        struct ubifs_lp_stats lst;
-        int err;
-};
-
-/**
  * scan_check_cb - scan callback.
  * @c: the UBIFS file-system description object
  * @lp: LEB properties to scan
  * @in_tree: whether the LEB properties are in main memory
- * @data: information passed to and from the caller of the scan
+ * @lst: lprops statistics to update
  *
  * This function returns a code that indicates whether the scan should continue
  * (%LPT_SCAN_CONTINUE), whether the LEB properties should be added to the tree
@@ -1030,12 +1020,12 @@ struct scan_check_data {
  */
 static int scan_check_cb(struct ubifs_info *c,
                          const struct ubifs_lprops *lp, int in_tree,
-                         struct scan_check_data *data)
+                         struct ubifs_lp_stats *lst)
 {
         struct ubifs_scan_leb *sleb;
         struct ubifs_scan_node *snod;
-        struct ubifs_lp_stats *lst = &data->lst;
-        int cat, lnum = lp->lnum, is_idx = 0, used = 0, free, dirty;
+        int cat, lnum = lp->lnum, is_idx = 0, used = 0, free, dirty, ret;
+        void *buf = NULL;
 
         cat = lp->flags & LPROPS_CAT_MASK;
         if (cat != LPROPS_UNCAT) {
@@ -1043,7 +1033,7 @@ static int scan_check_cb(struct ubifs_info *c,
                 if (cat != (lp->flags & LPROPS_CAT_MASK)) {
                         ubifs_err("bad LEB category %d expected %d",
                                   (lp->flags & LPROPS_CAT_MASK), cat);
-                        goto out;
+                        return -EINVAL;
                 }
         }
 
@@ -1077,7 +1067,7 @@ static int scan_check_cb(struct ubifs_info *c,
                         }
                         if (!found) {
                                 ubifs_err("bad LPT list (category %d)", cat);
-                                goto out;
+                                return -EINVAL;
                         }
                 }
         }
@@ -1089,36 +1079,40 @@ static int scan_check_cb(struct ubifs_info *c,
                 if ((lp->hpos != -1 && heap->arr[lp->hpos]->lnum != lnum) ||
                     lp != heap->arr[lp->hpos]) {
                         ubifs_err("bad LPT heap (category %d)", cat);
-                        goto out;
+                        return -EINVAL;
                 }
         }
 
-        sleb = ubifs_scan(c, lnum, 0, c->dbg->buf, 0);
-        if (IS_ERR(sleb)) {
-                /*
-                 * After an unclean unmount, empty and freeable LEBs
-                 * may contain garbage.
-                 */
-                if (lp->free == c->leb_size) {
-                        ubifs_err("scan errors were in empty LEB "
-                                  "- continuing checking");
-                        lst->empty_lebs += 1;
-                        lst->total_free += c->leb_size;
-                        lst->total_dark += ubifs_calc_dark(c, c->leb_size);
-                        return LPT_SCAN_CONTINUE;
-                }
+        buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+        if (!buf)
+                return -ENOMEM;
 
-                if (lp->free + lp->dirty == c->leb_size &&
-                    !(lp->flags & LPROPS_INDEX)) {
-                        ubifs_err("scan errors were in freeable LEB "
-                                  "- continuing checking");
-                        lst->total_free  += lp->free;
-                        lst->total_dirty += lp->dirty;
-                        lst->total_dark  +=  ubifs_calc_dark(c, c->leb_size);
-                        return LPT_SCAN_CONTINUE;
+        /*
+         * After an unclean unmount, empty and freeable LEBs
+         * may contain garbage - do not scan them.
+         */
+        if (lp->free == c->leb_size) {
+                lst->empty_lebs += 1;
+                lst->total_free += c->leb_size;
+                lst->total_dark += ubifs_calc_dark(c, c->leb_size);
+                return LPT_SCAN_CONTINUE;
+        }
+        if (lp->free + lp->dirty == c->leb_size &&
+            !(lp->flags & LPROPS_INDEX)) {
+                lst->total_free  += lp->free;
+                lst->total_dirty += lp->dirty;
+                lst->total_dark  +=  ubifs_calc_dark(c, c->leb_size);
+                return LPT_SCAN_CONTINUE;
+        }
+
+        sleb = ubifs_scan(c, lnum, 0, buf, 0);
+        if (IS_ERR(sleb)) {
+                ret = PTR_ERR(sleb);
+                if (ret == -EUCLEAN) {
+                        dbg_dump_lprops(c);
+                        dbg_dump_budg(c, &c->bi);
                 }
-                data->err = PTR_ERR(sleb);
-                return LPT_SCAN_STOP;
+                goto out;
         }
 
         is_idx = -1;
@@ -1236,6 +1230,7 @@ static int scan_check_cb(struct ubifs_info *c,
         }
 
         ubifs_scan_destroy(sleb);
+        vfree(buf);
         return LPT_SCAN_CONTINUE;
 
 out_print:
@@ -1245,9 +1240,10 @@ out_print:
         dbg_dump_leb(c, lnum);
 out_destroy:
         ubifs_scan_destroy(sleb);
+        ret = -EINVAL;
 out:
-        data->err = -EINVAL;
-        return LPT_SCAN_STOP;
+        vfree(buf);
+        return ret;
 }
 
 /**
@@ -1264,8 +1260,7 @@ out:
 int dbg_check_lprops(struct ubifs_info *c)
 {
         int i, err;
-        struct scan_check_data data;
-        struct ubifs_lp_stats *lst = &data.lst;
+        struct ubifs_lp_stats lst;
 
         if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
                 return 0;
@@ -1280,29 +1275,23 @@ int dbg_check_lprops(struct ubifs_info *c)
                         return err;
         }
 
-        memset(lst, 0, sizeof(struct ubifs_lp_stats));
-
-        data.err = 0;
+        memset(&lst, 0, sizeof(struct ubifs_lp_stats));
         err = ubifs_lpt_scan_nolock(c, c->main_first, c->leb_cnt - 1,
                                     (ubifs_lpt_scan_callback)scan_check_cb,
-                                    &data);
+                                    &lst);
         if (err && err != -ENOSPC)
                 goto out;
-        if (data.err) {
-                err = data.err;
-                goto out;
-        }
 
-        if (lst->empty_lebs != c->lst.empty_lebs ||
-            lst->idx_lebs != c->lst.idx_lebs ||
-            lst->total_free != c->lst.total_free ||
-            lst->total_dirty != c->lst.total_dirty ||
-            lst->total_used != c->lst.total_used) {
+        if (lst.empty_lebs != c->lst.empty_lebs ||
+            lst.idx_lebs != c->lst.idx_lebs ||
+            lst.total_free != c->lst.total_free ||
+            lst.total_dirty != c->lst.total_dirty ||
+            lst.total_used != c->lst.total_used) {
                 ubifs_err("bad overall accounting");
                 ubifs_err("calculated: empty_lebs %d, idx_lebs %d, "
                           "total_free %lld, total_dirty %lld, total_used %lld",
-                          lst->empty_lebs, lst->idx_lebs, lst->total_free,
-                          lst->total_dirty, lst->total_used);
+                          lst.empty_lebs, lst.idx_lebs, lst.total_free,
+                          lst.total_dirty, lst.total_used);
                 ubifs_err("read from lprops: empty_lebs %d, idx_lebs %d, "
                           "total_free %lld, total_dirty %lld, total_used %lld",
                           c->lst.empty_lebs, c->lst.idx_lebs, c->lst.total_free,
@@ -1311,11 +1300,11 @@ int dbg_check_lprops(struct ubifs_info *c)
                 goto out;
         }
 
-        if (lst->total_dead != c->lst.total_dead ||
-            lst->total_dark != c->lst.total_dark) {
+        if (lst.total_dead != c->lst.total_dead ||
+            lst.total_dark != c->lst.total_dark) {
                 ubifs_err("bad dead/dark space accounting");
                 ubifs_err("calculated: total_dead %lld, total_dark %lld",
-                          lst->total_dead, lst->total_dark);
+                          lst.total_dead, lst.total_dark);
                 ubifs_err("read from lprops: total_dead %lld, total_dark %lld",
                           c->lst.total_dead, c->lst.total_dark);
                 err = -EINVAL;
diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c
index 0084a33c4c69..ef5155e109a2 100644
--- a/fs/ubifs/lpt.c
+++ b/fs/ubifs/lpt.c
@@ -1270,10 +1270,9 @@ static int read_pnode(struct ubifs_info *c, struct ubifs_nnode *parent, int iip)
         lnum = branch->lnum;
         offs = branch->offs;
         pnode = kzalloc(sizeof(struct ubifs_pnode), GFP_NOFS);
-        if (!pnode) {
-                err = -ENOMEM;
-                goto out;
-        }
+        if (!pnode)
+                return -ENOMEM;
+
         if (lnum == 0) {
                 /*
                  * This pnode was not written which just means that the LEB
@@ -1363,6 +1362,7 @@ static int read_lsave(struct ubifs_info *c)
                 goto out;
         for (i = 0; i < c->lsave_cnt; i++) {
                 int lnum = c->lsave[i];
+                struct ubifs_lprops *lprops;
 
                 /*
                  * Due to automatic resizing, the values in the lsave table
@@ -1370,7 +1370,11 @@ static int read_lsave(struct ubifs_info *c)
                  */
                 if (lnum >= c->leb_cnt)
                         continue;
-                ubifs_lpt_lookup(c, lnum);
+                lprops = ubifs_lpt_lookup(c, lnum);
+                if (IS_ERR(lprops)) {
+                        err = PTR_ERR(lprops);
+                        goto out;
+                }
         }
 out:
         vfree(buf);
diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c
index d12535b7fc78..dfcb5748a7dc 100644
--- a/fs/ubifs/lpt_commit.c
+++ b/fs/ubifs/lpt_commit.c
@@ -29,6 +29,12 @@
 #include <linux/slab.h>
 #include "ubifs.h"
 
+#ifdef CONFIG_UBIFS_FS_DEBUG
+static int dbg_populate_lsave(struct ubifs_info *c);
+#else
+#define dbg_populate_lsave(c) 0
+#endif
+
 /**
  * first_dirty_cnode - find first dirty cnode.
  * @c: UBIFS file-system description object
@@ -586,7 +592,7 @@ static struct ubifs_pnode *next_pnode_to_dirty(struct ubifs_info *c,
                         if (nnode->nbranch[iip].lnum)
                                 break;
                 }
-       } while (iip >= UBIFS_LPT_FANOUT);
+        } while (iip >= UBIFS_LPT_FANOUT);
 
         /* Go right */
         nnode = ubifs_get_nnode(c, nnode, iip);
@@ -705,6 +711,9 @@ static int make_tree_dirty(struct ubifs_info *c)
         struct ubifs_pnode *pnode;
 
         pnode = pnode_lookup(c, 0);
+        if (IS_ERR(pnode))
+                return PTR_ERR(pnode);
+
         while (pnode) {
                 do_make_pnode_dirty(c, pnode);
                 pnode = next_pnode_to_dirty(c, pnode);
@@ -812,6 +821,10 @@ static void populate_lsave(struct ubifs_info *c)
                 c->lpt_drty_flgs |= LSAVE_DIRTY;
                 ubifs_add_lpt_dirt(c, c->lsave_lnum, c->lsave_sz);
         }
+
+        if (dbg_populate_lsave(c))
+                return;
+
         list_for_each_entry(lprops, &c->empty_list, list) {
                 c->lsave[cnt++] = lprops->lnum;
                 if (cnt >= c->lsave_cnt)
@@ -1625,29 +1638,35 @@ static int dbg_check_ltab_lnum(struct ubifs_info *c, int lnum)
 {
         int err, len = c->leb_size, dirty = 0, node_type, node_num, node_len;
         int ret;
-        void *buf = c->dbg->buf;
+        void *buf, *p;
 
         if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
                 return 0;
 
+        buf = p = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+        if (!buf) {
+                ubifs_err("cannot allocate memory for ltab checking");
+                return 0;
+        }
+
         dbg_lp("LEB %d", lnum);
         err = ubi_read(c->ubi, lnum, buf, 0, c->leb_size);
         if (err) {
                 dbg_msg("ubi_read failed, LEB %d, error %d", lnum, err);
-                return err;
+                goto out;
         }
         while (1) {
-                if (!is_a_node(c, buf, len)) {
+                if (!is_a_node(c, p, len)) {
                         int i, pad_len;
 
-                        pad_len = get_pad_len(c, buf, len);
+                        pad_len = get_pad_len(c, p, len);
                         if (pad_len) {
-                                buf += pad_len;
+                                p += pad_len;
                                 len -= pad_len;
                                 dirty += pad_len;
                                 continue;
                         }
-                        if (!dbg_is_all_ff(buf, len)) {
+                        if (!dbg_is_all_ff(p, len)) {
                                 dbg_msg("invalid empty space in LEB %d at %d",
                                         lnum, c->leb_size - len);
                                 err = -EINVAL;
@@ -1665,16 +1684,21 @@ static int dbg_check_ltab_lnum(struct ubifs_info *c, int lnum)
                                         lnum, dirty, c->ltab[i].dirty);
                                 err = -EINVAL;
                         }
-                        return err;
+                        goto out;
                 }
-                node_type = get_lpt_node_type(c, buf, &node_num);
+                node_type = get_lpt_node_type(c, p, &node_num);
                 node_len = get_lpt_node_len(c, node_type);
                 ret = dbg_is_node_dirty(c, node_type, lnum, c->leb_size - len);
                 if (ret == 1)
                         dirty += node_len;
-                buf += node_len;
+                p += node_len;
                 len -= node_len;
         }
+
+        err = 0;
+out:
+        vfree(buf);
+        return err;
 }
 
 /**
@@ -1867,25 +1891,31 @@ int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len)
 static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
 {
         int err, len = c->leb_size, node_type, node_num, node_len, offs;
-        void *buf = c->dbg->buf;
+        void *buf, *p;
 
         printk(KERN_DEBUG "(pid %d) start dumping LEB %d\n",
                current->pid, lnum);
+        buf = p = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+        if (!buf) {
+                ubifs_err("cannot allocate memory to dump LPT");
+                return;
+        }
+
         err = ubi_read(c->ubi, lnum, buf, 0, c->leb_size);
         if (err) {
                 ubifs_err("cannot read LEB %d, error %d", lnum, err);
-                return;
+                goto out;
         }
         while (1) {
                 offs = c->leb_size - len;
-                if (!is_a_node(c, buf, len)) {
+                if (!is_a_node(c, p, len)) {
                         int pad_len;
 
-                        pad_len = get_pad_len(c, buf, len);
+                        pad_len = get_pad_len(c, p, len);
                         if (pad_len) {
                                 printk(KERN_DEBUG "LEB %d:%d, pad %d bytes\n",
                                        lnum, offs, pad_len);
-                                buf += pad_len;
+                                p += pad_len;
                                 len -= pad_len;
                                 continue;
                         }
@@ -1895,7 +1925,7 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
                         break;
                 }
 
-                node_type = get_lpt_node_type(c, buf, &node_num);
+                node_type = get_lpt_node_type(c, p, &node_num);
                 switch (node_type) {
                 case UBIFS_LPT_PNODE:
                 {
@@ -1920,7 +1950,7 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
                         else
                                 printk(KERN_DEBUG "LEB %d:%d, nnode, ",
                                        lnum, offs);
-                        err = ubifs_unpack_nnode(c, buf, &nnode);
+                        err = ubifs_unpack_nnode(c, p, &nnode);
                         for (i = 0; i < UBIFS_LPT_FANOUT; i++) {
                                 printk(KERN_CONT "%d:%d", nnode.nbranch[i].lnum,
                                        nnode.nbranch[i].offs);
@@ -1941,15 +1971,18 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
                         break;
                 default:
                         ubifs_err("LPT node type %d not recognized", node_type);
-                        return;
+                        goto out;
                 }
 
-                buf += node_len;
+                p += node_len;
                 len -= node_len;
         }
 
         printk(KERN_DEBUG "(pid %d) finish dumping LEB %d\n",
                current->pid, lnum);
+out:
+        vfree(buf);
+        return;
 }
 
 /**
@@ -1971,4 +2004,47 @@ void dbg_dump_lpt_lebs(const struct ubifs_info *c)
                current->pid);
 }
 
+/**
+ * dbg_populate_lsave - debugging version of 'populate_lsave()'
+ * @c: UBIFS file-system description object
+ *
+ * This is a debugging version for 'populate_lsave()' which populates lsave
+ * with random LEBs instead of useful LEBs, which is good for test coverage.
+ * Returns zero if lsave has not been populated (this debugging feature is
+ * disabled) an non-zero if lsave has been populated.
+ */
+static int dbg_populate_lsave(struct ubifs_info *c)
+{
+        struct ubifs_lprops *lprops;
+        struct ubifs_lpt_heap *heap;
+        int i;
+
+        if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+                return 0;
+        if (random32() & 3)
+                return 0;
+
+        for (i = 0; i < c->lsave_cnt; i++)
+                c->lsave[i] = c->main_first;
+
+        list_for_each_entry(lprops, &c->empty_list, list)
+                c->lsave[random32() % c->lsave_cnt] = lprops->lnum;
+        list_for_each_entry(lprops, &c->freeable_list, list)
+                c->lsave[random32() % c->lsave_cnt] = lprops->lnum;
+        list_for_each_entry(lprops, &c->frdi_idx_list, list)
+                c->lsave[random32() % c->lsave_cnt] = lprops->lnum;
+
+        heap = &c->lpt_heap[LPROPS_DIRTY_IDX - 1];
+        for (i = 0; i < heap->cnt; i++)
+                c->lsave[random32() % c->lsave_cnt] = heap->arr[i]->lnum;
+        heap = &c->lpt_heap[LPROPS_DIRTY - 1];
+        for (i = 0; i < heap->cnt; i++)
+                c->lsave[random32() % c->lsave_cnt] = heap->arr[i]->lnum;
+        heap = &c->lpt_heap[LPROPS_FREE - 1];
+        for (i = 0; i < heap->cnt; i++)
+                c->lsave[random32() % c->lsave_cnt] = heap->arr[i]->lnum;
+
+        return 1;
+}
+
 #endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/master.c b/fs/ubifs/master.c
index 28beaeedadc0..278c2382e8c2 100644
--- a/fs/ubifs/master.c
+++ b/fs/ubifs/master.c
@@ -148,7 +148,7 @@ static int validate_master(const struct ubifs_info *c)
         }
 
         main_sz = (long long)c->main_lebs * c->leb_size;
-        if (c->old_idx_sz & 7 || c->old_idx_sz >= main_sz) {
+        if (c->bi.old_idx_sz & 7 || c->bi.old_idx_sz >= main_sz) {
                 err = 9;
                 goto out;
         }
@@ -218,7 +218,7 @@ static int validate_master(const struct ubifs_info *c)
         }
 
         if (c->lst.total_dead + c->lst.total_dark +
-            c->lst.total_used + c->old_idx_sz > main_sz) {
+            c->lst.total_used + c->bi.old_idx_sz > main_sz) {
                 err = 21;
                 goto out;
         }
@@ -286,7 +286,7 @@ int ubifs_read_master(struct ubifs_info *c)
         c->gc_lnum         = le32_to_cpu(c->mst_node->gc_lnum);
         c->ihead_lnum      = le32_to_cpu(c->mst_node->ihead_lnum);
         c->ihead_offs      = le32_to_cpu(c->mst_node->ihead_offs);
-        c->old_idx_sz      = le64_to_cpu(c->mst_node->index_size);
+        c->bi.old_idx_sz   = le64_to_cpu(c->mst_node->index_size);
         c->lpt_lnum        = le32_to_cpu(c->mst_node->lpt_lnum);
         c->lpt_offs        = le32_to_cpu(c->mst_node->lpt_offs);
         c->nhead_lnum      = le32_to_cpu(c->mst_node->nhead_lnum);
@@ -305,7 +305,7 @@ int ubifs_read_master(struct ubifs_info *c)
         c->lst.total_dead  = le64_to_cpu(c->mst_node->total_dead);
         c->lst.total_dark  = le64_to_cpu(c->mst_node->total_dark);
 
-        c->calc_idx_sz = c->old_idx_sz;
+        c->calc_idx_sz = c->bi.old_idx_sz;
 
         if (c->mst_node->flags & cpu_to_le32(UBIFS_MST_NO_ORPHS))
                 c->no_orphs = 1;
@@ -361,7 +361,8 @@ int ubifs_write_master(struct ubifs_info *c)
 {
         int err, lnum, offs, len;
 
-        if (c->ro_media)
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        if (c->ro_error)
                 return -EROFS;
 
         lnum = UBIFS_MST_LNUM;
diff --git a/fs/ubifs/misc.h b/fs/ubifs/misc.h
index 4fa81d867e41..0b5296a9a4c5 100644
--- a/fs/ubifs/misc.h
+++ b/fs/ubifs/misc.h
@@ -132,7 +132,8 @@ static inline int ubifs_leb_unmap(const struct ubifs_info *c, int lnum)
 {
         int err;
 
-        if (c->ro_media)
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        if (c->ro_error)
                 return -EROFS;
         err = ubi_leb_unmap(c->ubi, lnum);
         if (err) {
@@ -159,7 +160,8 @@ static inline int ubifs_leb_write(const struct ubifs_info *c, int lnum,
 {
         int err;
 
-        if (c->ro_media)
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        if (c->ro_error)
                 return -EROFS;
         err = ubi_leb_write(c->ubi, lnum, buf, offs, len, dtype);
         if (err) {
@@ -186,7 +188,8 @@ static inline int ubifs_leb_change(const struct ubifs_info *c, int lnum,
 {
         int err;
 
-        if (c->ro_media)
+        ubifs_assert(!c->ro_media && !c->ro_mount);
+        if (c->ro_error)
                 return -EROFS;
         err = ubi_leb_change(c->ubi, lnum, buf, len, dtype);
         if (err) {
@@ -337,4 +340,21 @@ static inline void ubifs_release_lprops(struct ubifs_info *c)
         mutex_unlock(&c->lp_mutex);
 }
 
+/**
+ * ubifs_next_log_lnum - switch to the next log LEB.
+ * @c: UBIFS file-system description object
+ * @lnum: current log LEB
+ *
+ * This helper function returns the log LEB number which goes next after LEB
+ * 'lnum'.
+ */
+static inline int ubifs_next_log_lnum(const struct ubifs_info *c, int lnum)
+{
+        lnum += 1;
+        if (lnum > c->log_last)
+                lnum = UBIFS_LOG_LNUM;
+
+        return lnum;
+}
+
 #endif /* __UBIFS_MISC_H__ */
diff --git a/fs/ubifs/orphan.c b/fs/ubifs/orphan.c
index 82009c74b6a3..a5422fffbd69 100644
--- a/fs/ubifs/orphan.c
+++ b/fs/ubifs/orphan.c
@@ -673,7 +673,8 @@ static int kill_orphans(struct ubifs_info *c)
                 sleb = ubifs_scan(c, lnum, 0, c->sbuf, 1);
                 if (IS_ERR(sleb)) {
                         if (PTR_ERR(sleb) == -EUCLEAN)
-                                sleb = ubifs_recover_leb(c, lnum, 0, c->sbuf, 0);
+                                sleb = ubifs_recover_leb(c, lnum, 0,
+                                                         c->sbuf, -1);
                         if (IS_ERR(sleb)) {
                                 err = PTR_ERR(sleb);
                                 break;
@@ -892,15 +893,22 @@ static int dbg_read_orphans(struct check_info *ci, struct ubifs_scan_leb *sleb)
 static int dbg_scan_orphans(struct ubifs_info *c, struct check_info *ci)
 {
         int lnum, err = 0;
+        void *buf;
 
         /* Check no-orphans flag and skip this if no orphans */
         if (c->no_orphs)
                 return 0;
 
+        buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+        if (!buf) {
+                ubifs_err("cannot allocate memory to check orphans");
+                return 0;
+        }
+
         for (lnum = c->orph_first; lnum <= c->orph_last; lnum++) {
                 struct ubifs_scan_leb *sleb;
 
-                sleb = ubifs_scan(c, lnum, 0, c->dbg->buf, 0);
+                sleb = ubifs_scan(c, lnum, 0, buf, 0);
                 if (IS_ERR(sleb)) {
                         err = PTR_ERR(sleb);
                         break;
@@ -912,6 +920,7 @@ static int dbg_scan_orphans(struct ubifs_info *c, struct check_info *ci)
                         break;
         }
 
+        vfree(buf);
         return err;
 }
 
diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c
index daae9e1f5382..783d8e0beb76 100644
--- a/fs/ubifs/recovery.c
+++ b/fs/ubifs/recovery.c
@@ -28,6 +28,23 @@
  * UBIFS always cleans away all remnants of an unclean un-mount, so that
  * errors do not accumulate. However UBIFS defers recovery if it is mounted
  * read-only, and the flash is not modified in that case.
+ *
+ * The general UBIFS approach to the recovery is that it recovers from
+ * corruptions which could be caused by power cuts, but it refuses to recover
+ * from corruption caused by other reasons. And UBIFS tries to distinguish
+ * between these 2 reasons of corruptions and silently recover in the former
+ * case and loudly complain in the latter case.
+ *
+ * UBIFS writes only to erased LEBs, so it writes only to the flash space
+ * containing only 0xFFs. UBIFS also always writes strictly from the beginning
+ * of the LEB to the end. And UBIFS assumes that the underlying flash media
+ * writes in @c->max_write_size bytes at a time.
+ *
+ * Hence, if UBIFS finds a corrupted node at offset X, it expects only the min.
+ * I/O unit corresponding to offset X to contain corrupted data, all the
+ * following min. I/O units have to contain empty space (all 0xFFs). If this is
+ * not true, the corruption cannot be the result of a power cut, and UBIFS
+ * refuses to mount.
  */
 
 #include <linux/crc32.h>
@@ -292,7 +309,7 @@ int ubifs_recover_master_node(struct ubifs_info *c)
 
         memcpy(c->mst_node, mst, UBIFS_MST_NODE_SZ);
 
-        if ((c->vfs_sb->s_flags & MS_RDONLY)) {
+        if (c->ro_mount) {
                 /* Read-only mode. Keep a copy for switching to rw mode */
                 c->rcvrd_mst_node = kmalloc(sz, GFP_KERNEL);
                 if (!c->rcvrd_mst_node) {
@@ -300,6 +317,32 @@ int ubifs_recover_master_node(struct ubifs_info *c)
                         goto out_free;
                 }
                 memcpy(c->rcvrd_mst_node, c->mst_node, UBIFS_MST_NODE_SZ);
+
+                /*
+                 * We had to recover the master node, which means there was an
+                 * unclean reboot. However, it is possible that the master node
+                 * is clean at this point, i.e., %UBIFS_MST_DIRTY is not set.
+                 * E.g., consider the following chain of events:
+                 *
+                 * 1. UBIFS was cleanly unmounted, so the master node is clean
+                 * 2. UBIFS is being mounted R/W and starts changing the master
+                 *    node in the first (%UBIFS_MST_LNUM). A power cut happens,
+                 *    so this LEB ends up with some amount of garbage at the
+                 *    end.
+                 * 3. UBIFS is being mounted R/O. We reach this place and
+                 *    recover the master node from the second LEB
+                 *    (%UBIFS_MST_LNUM + 1). But we cannot update the media
+                 *    because we are being mounted R/O. We have to defer the
+                 *    operation.
+                 * 4. However, this master node (@c->mst_node) is marked as
+                 *    clean (since the step 1). And if we just return, the
+                 *    mount code will be confused and won't recover the master
+                 *    node when it is re-mounter R/W later.
+                 *
+                 *    Thus, to force the recovery by marking the master node as
+                 *    dirty.
+                 */
+                c->mst_node->flags |= cpu_to_le32(UBIFS_MST_DIRTY);
         } else {
                 /* Write the recovered master node */
                 c->max_sqnum = le64_to_cpu(mst->ch.sqnum) - 1;
@@ -362,8 +405,9 @@ int ubifs_write_rcvrd_mst_node(struct ubifs_info *c)
  * @offs: offset to check
  *
  * This function returns %1 if @offs was in the last write to the LEB whose data
- * is in @buf, otherwise %0 is returned.  The determination is made by checking
- * for subsequent empty space starting from the next @c->min_io_size boundary.
+ * is in @buf, otherwise %0 is returned. The determination is made by checking
+ * for subsequent empty space starting from the next @c->max_write_size
+ * boundary.
  */
 static int is_last_write(const struct ubifs_info *c, void *buf, int offs)
 {
@@ -371,10 +415,10 @@ static int is_last_write(const struct ubifs_info *c, void *buf, int offs)
         uint8_t *p;
 
         /*
-         * Round up to the next @c->min_io_size boundary i.e. @offs is in the
-         * last wbuf written. After that should be empty space.
+         * Round up to the next @c->max_write_size boundary i.e. @offs is in
+         * the last wbuf written. After that should be empty space.
          */
-        empty_offs = ALIGN(offs + 1, c->min_io_size);
+        empty_offs = ALIGN(offs + 1, c->max_write_size);
         check_len = c->leb_size - empty_offs;
         p = buf + empty_offs - offs;
         return is_empty(p, check_len);
@@ -429,7 +473,7 @@ static int no_more_nodes(const struct ubifs_info *c, void *buf, int len,
         int skip, dlen = le32_to_cpu(ch->len);
 
         /* Check for empty space after the corrupt node's common header */
-        skip = ALIGN(offs + UBIFS_CH_SZ, c->min_io_size) - offs;
+        skip = ALIGN(offs + UBIFS_CH_SZ, c->max_write_size) - offs;
         if (is_empty(buf + skip, len - skip))
                 return 1;
         /*
@@ -441,7 +485,7 @@ static int no_more_nodes(const struct ubifs_info *c, void *buf, int len,
                 return 0;
         }
         /* Now we know the corrupt node's length we can skip over it */
-        skip = ALIGN(offs + dlen, c->min_io_size) - offs;
+        skip = ALIGN(offs + dlen, c->max_write_size) - offs;
         /* After which there should be empty space */
         if (is_empty(buf + skip, len - skip))
                 return 1;
@@ -469,7 +513,7 @@ static int fix_unclean_leb(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
                 endpt = snod->offs + snod->len;
         }
 
-        if ((c->vfs_sb->s_flags & MS_RDONLY) && !c->remounting_rw) {
+        if (c->ro_mount && !c->remounting_rw) {
                 /* Add to recovery list */
                 struct ubifs_unclean_leb *ucleb;
 
@@ -520,16 +564,15 @@ static int fix_unclean_leb(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 }
 
 /**
- * drop_incomplete_group - drop nodes from an incomplete group.
+ * drop_last_group - drop the last group of nodes.
  * @sleb: scanned LEB information
  * @offs: offset of dropped nodes is returned here
  *
- * This function returns %1 if nodes are dropped and %0 otherwise.
+ * This is a helper function for 'ubifs_recover_leb()' which drops the last
+ * group of nodes of the scanned LEB.
  */
-static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs)
+static void drop_last_group(struct ubifs_scan_leb *sleb, int *offs)
 {
-        int dropped = 0;
-
         while (!list_empty(&sleb->nodes)) {
                 struct ubifs_scan_node *snod;
                 struct ubifs_ch *ch;
@@ -538,15 +581,40 @@ static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs)
                                   list);
                 ch = snod->node;
                 if (ch->group_type != UBIFS_IN_NODE_GROUP)
-                        return dropped;
-                dbg_rcvry("dropping node at %d:%d", sleb->lnum, snod->offs);
+                        break;
+
+                dbg_rcvry("dropping grouped node at %d:%d",
+                          sleb->lnum, snod->offs);
+                *offs = snod->offs;
+                list_del(&snod->list);
+                kfree(snod);
+                sleb->nodes_cnt -= 1;
+        }
+}
+
+/**
+ * drop_last_node - drop the last node.
+ * @sleb: scanned LEB information
+ * @offs: offset of dropped nodes is returned here
+ * @grouped: non-zero if whole group of nodes have to be dropped
+ *
+ * This is a helper function for 'ubifs_recover_leb()' which drops the last
+ * node of the scanned LEB.
+ */
+static void drop_last_node(struct ubifs_scan_leb *sleb, int *offs)
+{
+        struct ubifs_scan_node *snod;
+
+        if (!list_empty(&sleb->nodes)) {
+                snod = list_entry(sleb->nodes.prev, struct ubifs_scan_node,
+                                  list);
+
+                dbg_rcvry("dropping last node at %d:%d", sleb->lnum, snod->offs);
                 *offs = snod->offs;
                 list_del(&snod->list);
                 kfree(snod);
                 sleb->nodes_cnt -= 1;
-                dropped = 1;
         }
-        return dropped;
 }
 
 /**
@@ -555,7 +623,8 @@ static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs)
  * @lnum: LEB number
  * @offs: offset
  * @sbuf: LEB-sized buffer to use
- * @grouped: nodes may be grouped for recovery
+ * @jhead: journal head number this LEB belongs to (%-1 if the LEB does not
+ *         belong to any journal head)
  *
  * This function does a scan of a LEB, but caters for errors that might have
  * been caused by the unclean unmount from which we are attempting to recover.
@@ -563,25 +632,21 @@ static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs)
  * found, and a negative error code in case of failure.
  */
 struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum,
-                                         int offs, void *sbuf, int grouped)
+                                         int offs, void *sbuf, int jhead)
 {
-        int err, len = c->leb_size - offs, need_clean = 0, quiet = 1;
-        int empty_chkd = 0, start = offs;
+        int ret = 0, err, len = c->leb_size - offs, start = offs, min_io_unit;
+        int grouped = jhead == -1 ? 0 : c->jheads[jhead].grouped;
         struct ubifs_scan_leb *sleb;
         void *buf = sbuf + offs;
 
-        dbg_rcvry("%d:%d", lnum, offs);
+        dbg_rcvry("%d:%d, jhead %d, grouped %d", lnum, offs, jhead, grouped);
 
         sleb = ubifs_start_scan(c, lnum, offs, sbuf);
         if (IS_ERR(sleb))
                 return sleb;
 
-        if (sleb->ecc)
-                need_clean = 1;
-
+        ubifs_assert(len >= 8);
         while (len >= 8) {
-                int ret;
-
                 dbg_scan("look at LEB %d:%d (%d bytes left)",
                          lnum, offs, len);
 
@@ -591,8 +656,7 @@ struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum,
                  * Scan quietly until there is an error from which we cannot
                  * recover
                  */
-                ret = ubifs_scan_a_node(c, buf, len, lnum, offs, quiet);
-
+                ret = ubifs_scan_a_node(c, buf, len, lnum, offs, 1);
                 if (ret == SCANNED_A_NODE) {
                         /* A valid node, and not a padding node */
                         struct ubifs_ch *ch = buf;
@@ -605,104 +669,126 @@ struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum,
                         offs += node_len;
                         buf += node_len;
                         len -= node_len;
-                        continue;
-                }
-
-                if (ret > 0) {
+                } else if (ret > 0) {
                         /* Padding bytes or a valid padding node */
                         offs += ret;
                         buf += ret;
                         len -= ret;
-                        continue;
-                }
-
-                if (ret == SCANNED_EMPTY_SPACE) {
-                        if (!is_empty(buf, len)) {
-                                if (!is_last_write(c, buf, offs))
-                                        break;
-                                clean_buf(c, &buf, lnum, &offs, &len);
-                                need_clean = 1;
-                        }
-                        empty_chkd = 1;
+                } else if (ret == SCANNED_EMPTY_SPACE ||
+                           ret == SCANNED_GARBAGE     ||
+                           ret == SCANNED_A_BAD_PAD_NODE ||
+                           ret == SCANNED_A_CORRUPT_NODE) {
+                        dbg_rcvry("found corruption - %d", ret);
                         break;
-                }
-
-                if (ret == SCANNED_GARBAGE || ret == SCANNED_A_BAD_PAD_NODE)
-                        if (is_last_write(c, buf, offs)) {
-                                clean_buf(c, &buf, lnum, &offs, &len);
-                                need_clean = 1;
-                                empty_chkd = 1;
-                                break;
-                        }
-
-                if (ret == SCANNED_A_CORRUPT_NODE)
-                        if (no_more_nodes(c, buf, len, lnum, offs)) {
-                                clean_buf(c, &buf, lnum, &offs, &len);
-                                need_clean = 1;
-                                empty_chkd = 1;
-                                break;
-                        }
-
-                if (quiet) {
-                        /* Redo the last scan but noisily */
-                        quiet = 0;
-                        continue;
-                }
-
-                switch (ret) {
-                case SCANNED_GARBAGE:
-                        dbg_err("garbage");
-                        goto corrupted;
-                case SCANNED_A_CORRUPT_NODE:
-                case SCANNED_A_BAD_PAD_NODE:
-                        dbg_err("bad node");
-                        goto corrupted;
-                default:
-                        dbg_err("unknown");
+                } else {
+                        dbg_err("unexpected return value %d", ret);
                         err = -EINVAL;
                         goto error;
                 }
         }
 
-        if (!empty_chkd && !is_empty(buf, len)) {
-                if (is_last_write(c, buf, offs)) {
-                        clean_buf(c, &buf, lnum, &offs, &len);
-                        need_clean = 1;
-                } else {
+        if (ret == SCANNED_GARBAGE || ret == SCANNED_A_BAD_PAD_NODE) {
+                if (!is_last_write(c, buf, offs))
+                        goto corrupted_rescan;
+        } else if (ret == SCANNED_A_CORRUPT_NODE) {
+                if (!no_more_nodes(c, buf, len, lnum, offs))
+                        goto corrupted_rescan;
+        } else if (!is_empty(buf, len)) {
+                if (!is_last_write(c, buf, offs)) {
                         int corruption = first_non_ff(buf, len);
 
+                        /*
+                         * See header comment for this file for more
+                         * explanations about the reasons we have this check.
+                         */
                         ubifs_err("corrupt empty space LEB %d:%d, corruption "
                                   "starts at %d", lnum, offs, corruption);
                         /* Make sure we dump interesting non-0xFF data */
-                        offs = corruption;
+                        offs += corruption;
                         buf += corruption;
                         goto corrupted;
                 }
         }
 
-        /* Drop nodes from incomplete group */
-        if (grouped && drop_incomplete_group(sleb, &offs)) {
-                buf = sbuf + offs;
-                len = c->leb_size - offs;
-                clean_buf(c, &buf, lnum, &offs, &len);
-                need_clean = 1;
-        }
+        min_io_unit = round_down(offs, c->min_io_size);
+        if (grouped)
+                /*
+                 * If nodes are grouped, always drop the incomplete group at
+                 * the end.
+                 */
+                drop_last_group(sleb, &offs);
 
-        if (offs % c->min_io_size) {
-                clean_buf(c, &buf, lnum, &offs, &len);
-                need_clean = 1;
+        if (jhead == GCHD) {
+                /*
+                 * If this LEB belongs to the GC head then while we are in the
+                 * middle of the same min. I/O unit keep dropping nodes. So
+                 * basically, what we want is to make sure that the last min.
+                 * I/O unit where we saw the corruption is dropped completely
+                 * with all the uncorrupted nodes which may possibly sit there.
+                 *
+                 * In other words, let's name the min. I/O unit where the
+                 * corruption starts B, and the previous min. I/O unit A. The
+                 * below code tries to deal with a situation when half of B
+                 * contains valid nodes or the end of a valid node, and the
+                 * second half of B contains corrupted data or garbage. This
+                 * means that UBIFS had been writing to B just before the power
+                 * cut happened. I do not know how realistic is this scenario
+                 * that half of the min. I/O unit had been written successfully
+                 * and the other half not, but this is possible in our 'failure
+                 * mode emulation' infrastructure at least.
+                 *
+                 * So what is the problem, why we need to drop those nodes? Why
+                 * can't we just clean-up the second half of B by putting a
+                 * padding node there? We can, and this works fine with one
+                 * exception which was reproduced with power cut emulation
+                 * testing and happens extremely rarely.
+                 *
+                 * Imagine the file-system is full, we run GC which starts
+                 * moving valid nodes from LEB X to LEB Y (obviously, LEB Y is
+                 * the current GC head LEB). The @c->gc_lnum is -1, which means
+                 * that GC will retain LEB X and will try to continue. Imagine
+                 * that LEB X is currently the dirtiest LEB, and the amount of
+                 * used space in LEB Y is exactly the same as amount of free
+                 * space in LEB X.
+                 *
+                 * And a power cut happens when nodes are moved from LEB X to
+                 * LEB Y. We are here trying to recover LEB Y which is the GC
+                 * head LEB. We find the min. I/O unit B as described above.
+                 * Then we clean-up LEB Y by padding min. I/O unit. And later
+                 * 'ubifs_rcvry_gc_commit()' function fails, because it cannot
+                 * find a dirty LEB which could be GC'd into LEB Y! Even LEB X
+                 * does not match because the amount of valid nodes there does
+                 * not fit the free space in LEB Y any more! And this is
+                 * because of the padding node which we added to LEB Y. The
+                 * user-visible effect of this which I once observed and
+                 * analysed is that we cannot mount the file-system with
+                 * -ENOSPC error.
+                 *
+                 * So obviously, to make sure that situation does not happen we
+                 * should free min. I/O unit B in LEB Y completely and the last
+                 * used min. I/O unit in LEB Y should be A. This is basically
+                 * what the below code tries to do.
+                 */
+                while (offs > min_io_unit)
+                        drop_last_node(sleb, &offs);
         }
 
+        buf = sbuf + offs;
+        len = c->leb_size - offs;
+
+        clean_buf(c, &buf, lnum, &offs, &len);
         ubifs_end_scan(c, sleb, lnum, offs);
 
-        if (need_clean) {
-                err = fix_unclean_leb(c, sleb, start);
-                if (err)
-                        goto error;
-        }
+        err = fix_unclean_leb(c, sleb, start);
+        if (err)
+                goto error;
 
         return sleb;
 
+corrupted_rescan:
+        /* Re-scan the corrupted data with verbose messages */
+        dbg_err("corruptio %d", ret);
+        ubifs_scan_a_node(c, buf, len, lnum, offs, 1);
 corrupted:
         ubifs_scanned_corruption(c, lnum, offs, buf);
         err = -EUCLEAN;
@@ -772,7 +858,8 @@ out_free:
  * @sbuf: LEB-sized buffer to use
  *
  * This function does a scan of a LEB, but caters for errors that might have
- * been caused by the unclean unmount from which we are attempting to recover.
+ * been caused by unclean reboots from which we are attempting to recover
+ * (assume that only the last log LEB can be corrupted by an unclean reboot).
  *
  * This function returns %0 on success and a negative error code on failure.
  */
@@ -818,7 +905,7 @@ struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum,
                 }
                 ubifs_scan_destroy(sleb);
         }
-        return ubifs_recover_leb(c, lnum, offs, sbuf, 0);
+        return ubifs_recover_leb(c, lnum, offs, sbuf, -1);
 }
 
 /**
@@ -835,12 +922,8 @@ struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum,
 static int recover_head(const struct ubifs_info *c, int lnum, int offs,
                         void *sbuf)
 {
-        int len, err;
+        int len = c->max_write_size, err;
 
-        if (c->min_io_size > 1)
-                len = c->min_io_size;
-        else
-                len = 512;
         if (offs + len > c->leb_size)
                 len = c->leb_size - offs;
 
@@ -883,7 +966,7 @@ int ubifs_recover_inl_heads(const struct ubifs_info *c, void *sbuf)
 {
         int err;
 
-        ubifs_assert(!(c->vfs_sb->s_flags & MS_RDONLY) || c->remounting_rw);
+        ubifs_assert(!c->ro_mount || c->remounting_rw);
 
         dbg_rcvry("checking index head at %d:%d", c->ihead_lnum, c->ihead_offs);
         err = recover_head(c, c->ihead_lnum, c->ihead_offs, sbuf);
@@ -1025,6 +1108,53 @@ int ubifs_clean_lebs(const struct ubifs_info *c, void *sbuf)
 }
 
 /**
+ * grab_empty_leb - grab an empty LEB to use as GC LEB and run commit.
+ * @c: UBIFS file-system description object
+ *
+ * This is a helper function for 'ubifs_rcvry_gc_commit()' which grabs an empty
+ * LEB to be used as GC LEB (@c->gc_lnum), and then runs the commit. Returns
+ * zero in case of success and a negative error code in case of failure.
+ */
+static int grab_empty_leb(struct ubifs_info *c)
+{
+        int lnum, err;
+
+        /*
+         * Note, it is very important to first search for an empty LEB and then
+         * run the commit, not vice-versa. The reason is that there might be
+         * only one empty LEB at the moment, the one which has been the
+         * @c->gc_lnum just before the power cut happened. During the regular
+         * UBIFS operation (not now) @c->gc_lnum is marked as "taken", so no
+         * one but GC can grab it. But at this moment this single empty LEB is
+         * not marked as taken, so if we run commit - what happens? Right, the
+         * commit will grab it and write the index there. Remember that the
+         * index always expands as long as there is free space, and it only
+         * starts consolidating when we run out of space.
+         *
+         * IOW, if we run commit now, we might not be able to find a free LEB
+         * after this.
+         */
+        lnum = ubifs_find_free_leb_for_idx(c);
+        if (lnum < 0) {
+                dbg_err("could not find an empty LEB");
+                dbg_dump_lprops(c);
+                dbg_dump_budg(c, &c->bi);
+                return lnum;
+        }
+
+        /* Reset the index flag */
+        err = ubifs_change_one_lp(c, lnum, LPROPS_NC, LPROPS_NC, 0,
+                                  LPROPS_INDEX, 0);
+        if (err)
+                return err;
+
+        c->gc_lnum = lnum;
+        dbg_rcvry("found empty LEB %d, run commit", lnum);
+
+        return ubifs_run_commit(c);
+}
+
+/**
  * ubifs_rcvry_gc_commit - recover the GC LEB number and run the commit.
  * @c: UBIFS file-system description object
  *
@@ -1046,71 +1176,26 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c)
 {
         struct ubifs_wbuf *wbuf = &c->jheads[GCHD].wbuf;
         struct ubifs_lprops lp;
-        int lnum, err;
+        int err;
+
+        dbg_rcvry("GC head LEB %d, offs %d", wbuf->lnum, wbuf->offs);
 
         c->gc_lnum = -1;
-        if (wbuf->lnum == -1) {
-                dbg_rcvry("no GC head LEB");
-                goto find_free;
-        }
-        /*
-         * See whether the used space in the dirtiest LEB fits in the GC head
-         * LEB.
-         */
-        if (wbuf->offs == c->leb_size) {
-                dbg_rcvry("no room in GC head LEB");
-                goto find_free;
-        }
+        if (wbuf->lnum == -1 || wbuf->offs == c->leb_size)
+                return grab_empty_leb(c);
+
         err = ubifs_find_dirty_leb(c, &lp, wbuf->offs, 2);
         if (err) {
-                /*
-                 * There are no dirty or empty LEBs subject to here being
-                 * enough for the index. Try to use
-                 * 'ubifs_find_free_leb_for_idx()', which will return any empty
-                 * LEBs (ignoring index requirements). If the index then
-                 * doesn't have enough LEBs the recovery commit will fail -
-                 * which is the  same result anyway i.e. recovery fails. So
-                 * there is no problem ignoring index  requirements and just
-                 * grabbing a free LEB since we have already established there
-                 * is not a dirty LEB we could have used instead.
-                 */
-                if (err == -ENOSPC) {
-                        dbg_rcvry("could not find a dirty LEB");
-                        goto find_free;
-                }
-                return err;
-        }
-        ubifs_assert(!(lp.flags & LPROPS_INDEX));
-        lnum = lp.lnum;
-        if (lp.free + lp.dirty == c->leb_size) {
-                /* An empty LEB was returned */
-                if (lp.free != c->leb_size) {
-                        err = ubifs_change_one_lp(c, lnum, c->leb_size,
-                                                  0, 0, 0, 0);
-                        if (err)
-                                return err;
-                }
-                err = ubifs_leb_unmap(c, lnum);
-                if (err)
-                        return err;
-                c->gc_lnum = lnum;
-                dbg_rcvry("allocated LEB %d for GC", lnum);
-                /* Run the commit */
-                dbg_rcvry("committing");
-                return ubifs_run_commit(c);
-        }
-        /*
-         * There was no empty LEB so the used space in the dirtiest LEB must fit
-         * in the GC head LEB.
-         */
-        if (lp.free + lp.dirty < wbuf->offs) {
-                dbg_rcvry("LEB %d doesn't fit in GC head LEB %d:%d",
-                          lnum, wbuf->lnum, wbuf->offs);
-                err = ubifs_return_leb(c, lnum);
-                if (err)
+                if (err != -ENOSPC)
                         return err;
-                goto find_free;
+
+                dbg_rcvry("could not find a dirty LEB");
+                return grab_empty_leb(c);
         }
+
+        ubifs_assert(!(lp.flags & LPROPS_INDEX));
+        ubifs_assert(lp.free + lp.dirty >= wbuf->offs);
+
         /*
          * We run the commit before garbage collection otherwise subsequent
          * mounts will see the GC and orphan deletion in a different order.
@@ -1119,11 +1204,8 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c)
         err = ubifs_run_commit(c);
         if (err)
                 return err;
-        /*
-         * The data in the dirtiest LEB fits in the GC head LEB, so do the GC
-         * - use locking to keep 'ubifs_assert()' happy.
-         */
-        dbg_rcvry("GC'ing LEB %d", lnum);
+
+        dbg_rcvry("GC'ing LEB %d", lp.lnum);
         mutex_lock_nested(&wbuf->io_mutex, wbuf->jhead);
         err = ubifs_garbage_collect_leb(c, &lp);
         if (err >= 0) {
@@ -1139,37 +1221,17 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c)
                         err = -EINVAL;
                 return err;
         }
-        if (err != LEB_RETAINED) {
-                dbg_err("GC returned %d", err);
+
+        ubifs_assert(err == LEB_RETAINED);
+        if (err != LEB_RETAINED)
                 return -EINVAL;
-        }
+
         err = ubifs_leb_unmap(c, c->gc_lnum);
         if (err)
                 return err;
-        dbg_rcvry("allocated LEB %d for GC", lnum);
-        return 0;
 
-find_free:
-        /*
-         * There is no GC head LEB or the free space in the GC head LEB is too
-         * small, or there are not dirty LEBs. Allocate gc_lnum by calling
-         * 'ubifs_find_free_leb_for_idx()' so GC is not run.
-         */
-        lnum = ubifs_find_free_leb_for_idx(c);
-        if (lnum < 0) {
-                dbg_err("could not find an empty LEB");
-                return lnum;
-        }
-        /* And reset the index flag */
-        err = ubifs_change_one_lp(c, lnum, LPROPS_NC, LPROPS_NC, 0,
-                                  LPROPS_INDEX, 0);
-        if (err)
-                return err;
-        c->gc_lnum = lnum;
-        dbg_rcvry("allocated LEB %d for GC", lnum);
-        /* Run the commit */
-        dbg_rcvry("committing");
-        return ubifs_run_commit(c);
+        dbg_rcvry("allocated LEB %d for GC", lp.lnum);
+        return 0;
 }
 
 /**
@@ -1411,7 +1473,7 @@ static int fix_size_in_place(struct ubifs_info *c, struct size_entry *e)
         err = ubi_leb_change(c->ubi, lnum, c->sbuf, len, UBI_UNKNOWN);
         if (err)
                 goto out;
-        dbg_rcvry("inode %lu at %d:%d size %lld -> %lld ",
+        dbg_rcvry("inode %lu at %d:%d size %lld -> %lld",
                   (unsigned long)e->inum, lnum, offs, i_size, e->d_size);
         return 0;
 
@@ -1460,20 +1522,27 @@ int ubifs_recover_size(struct ubifs_info *c)
                                 e->i_size = le64_to_cpu(ino->size);
                         }
                 }
+
                 if (e->exists && e->i_size < e->d_size) {
-                        if (!e->inode && (c->vfs_sb->s_flags & MS_RDONLY)) {
+                        if (c->ro_mount) {
                                 /* Fix the inode size and pin it in memory */
                                 struct inode *inode;
+                                struct ubifs_inode *ui;
+
+                                ubifs_assert(!e->inode);
 
                                 inode = ubifs_iget(c->vfs_sb, e->inum);
                                 if (IS_ERR(inode))
                                         return PTR_ERR(inode);
+
+                                ui = ubifs_inode(inode);
                                 if (inode->i_size < e->d_size) {
                                         dbg_rcvry("ino %lu size %lld -> %lld",
                                                   (unsigned long)e->inum,
-                                                  e->d_size, inode->i_size);
+                                                  inode->i_size, e->d_size);
                                         inode->i_size = e->d_size;
-                                        ubifs_inode(inode)->ui_size = e->d_size;
+                                        ui->ui_size = e->d_size;
+                                        ui->synced_i_size = e->d_size;
                                         e->inode = inode;
                                         this = rb_next(this);
                                         continue;
@@ -1488,9 +1557,11 @@ int ubifs_recover_size(struct ubifs_info *c)
                                         iput(e->inode);
                         }
                 }
+
                 this = rb_next(this);
                 rb_erase(&e->rb, &c->size_tree);
                 kfree(e);
         }
+
         return 0;
 }
diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c
index 5c2d6d759a3e..5e97161ce4d3 100644
--- a/fs/ubifs/replay.c
+++ b/fs/ubifs/replay.c
@@ -33,43 +33,32 @@
  */
 
 #include "ubifs.h"
-
-/*
- * Replay flags.
- *
- * REPLAY_DELETION: node was deleted
- * REPLAY_REF: node is a reference node
- */
-enum {
-        REPLAY_DELETION = 1,
-        REPLAY_REF = 2,
-};
+#include <linux/list_sort.h>
 
 /**
- * struct replay_entry - replay tree entry.
+ * struct replay_entry - replay list entry.
  * @lnum: logical eraseblock number of the node
  * @offs: node offset
  * @len: node length
+ * @deletion: non-zero if this entry corresponds to a node deletion
  * @sqnum: node sequence number
- * @flags: replay flags
- * @rb: links the replay tree
+ * @list: links the replay list
  * @key: node key
  * @nm: directory entry name
  * @old_size: truncation old size
  * @new_size: truncation new size
- * @free: amount of free space in a bud
- * @dirty: amount of dirty space in a bud from padding and deletion nodes
  *
- * UBIFS journal replay must compare node sequence numbers, which means it must
- * build a tree of node information to insert into the TNC.
+ * The replay process first scans all buds and builds the replay list, then
+ * sorts the replay list in nodes sequence number order, and then inserts all
+ * the replay entries to the TNC.
  */
 struct replay_entry {
         int lnum;
         int offs;
         int len;
+        unsigned int deletion:1;
         unsigned long long sqnum;
-        int flags;
-        struct rb_node rb;
+        struct list_head list;
         union ubifs_key key;
         union {
                 struct qstr nm;
@@ -77,10 +66,6 @@ struct replay_entry {
                         loff_t old_size;
                         loff_t new_size;
                 };
-                struct {
-                        int free;
-                        int dirty;
-                };
         };
 };
 
@@ -88,57 +73,64 @@ struct replay_entry {
  * struct bud_entry - entry in the list of buds to replay.
  * @list: next bud in the list
  * @bud: bud description object
- * @free: free bytes in the bud
  * @sqnum: reference node sequence number
+ * @free: free bytes in the bud
+ * @dirty: dirty bytes in the bud
  */
 struct bud_entry {
         struct list_head list;
         struct ubifs_bud *bud;
-        int free;
         unsigned long long sqnum;
+        int free;
+        int dirty;
 };
 
 /**
  * set_bud_lprops - set free and dirty space used by a bud.
  * @c: UBIFS file-system description object
- * @r: replay entry of bud
+ * @b: bud entry which describes the bud
+ *
+ * This function makes sure the LEB properties of bud @b are set correctly
+ * after the replay. Returns zero in case of success and a negative error code
+ * in case of failure.
  */
-static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r)
+static int set_bud_lprops(struct ubifs_info *c, struct bud_entry *b)
 {
         const struct ubifs_lprops *lp;
         int err = 0, dirty;
 
         ubifs_get_lprops(c);
 
-        lp = ubifs_lpt_lookup_dirty(c, r->lnum);
+        lp = ubifs_lpt_lookup_dirty(c, b->bud->lnum);
         if (IS_ERR(lp)) {
                 err = PTR_ERR(lp);
                 goto out;
         }
 
         dirty = lp->dirty;
-        if (r->offs == 0 && (lp->free != c->leb_size || lp->dirty != 0)) {
+        if (b->bud->start == 0 && (lp->free != c->leb_size || lp->dirty != 0)) {
                 /*
                  * The LEB was added to the journal with a starting offset of
                  * zero which means the LEB must have been empty. The LEB
-                 * property values should be lp->free == c->leb_size and
-                 * lp->dirty == 0, but that is not the case. The reason is that
-                 * the LEB was garbage collected. The garbage collector resets
-                 * the free and dirty space without recording it anywhere except
-                 * lprops, so if there is not a commit then lprops does not have
-                 * that information next time the file system is mounted.
+                 * property values should be @lp->free == @c->leb_size and
+                 * @lp->dirty == 0, but that is not the case. The reason is that
+                 * the LEB had been garbage collected before it became the bud,
+                 * and there was not commit inbetween. The garbage collector
+                 * resets the free and dirty space without recording it
+                 * anywhere except lprops, so if there was no commit then
+                 * lprops does not have that information.
                  *
                  * We do not need to adjust free space because the scan has told
                  * us the exact value which is recorded in the replay entry as
-                 * r->free.
+                 * @b->free.
                  *
                  * However we do need to subtract from the dirty space the
                  * amount of space that the garbage collector reclaimed, which
                  * is the whole LEB minus the amount of space that was free.
                  */
-                dbg_mnt("bud LEB %d was GC'd (%d free, %d dirty)", r->lnum,
+                dbg_mnt("bud LEB %d was GC'd (%d free, %d dirty)", b->bud->lnum,
                         lp->free, lp->dirty);
-                dbg_gc("bud LEB %d was GC'd (%d free, %d dirty)", r->lnum,
+                dbg_gc("bud LEB %d was GC'd (%d free, %d dirty)", b->bud->lnum,
                         lp->free, lp->dirty);
                 dirty -= c->leb_size - lp->free;
                 /*
@@ -150,21 +142,48 @@ static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r)
                  */
                 if (dirty != 0)
                         dbg_msg("LEB %d lp: %d free %d dirty "
-                                "replay: %d free %d dirty", r->lnum, lp->free,
-                                lp->dirty, r->free, r->dirty);
+                                "replay: %d free %d dirty", b->bud->lnum,
+                                lp->free, lp->dirty, b->free, b->dirty);
         }
-        lp = ubifs_change_lp(c, lp, r->free, dirty + r->dirty,
+        lp = ubifs_change_lp(c, lp, b->free, dirty + b->dirty,
                              lp->flags | LPROPS_TAKEN, 0);
         if (IS_ERR(lp)) {
                 err = PTR_ERR(lp);
                 goto out;
         }
+
+        /* Make sure the journal head points to the latest bud */
+        err = ubifs_wbuf_seek_nolock(&c->jheads[b->bud->jhead].wbuf,
+                                     b->bud->lnum, c->leb_size - b->free,
+                                     UBI_SHORTTERM);
+
 out:
         ubifs_release_lprops(c);
         return err;
 }
 
 /**
+ * set_buds_lprops - set free and dirty space for all replayed buds.
+ * @c: UBIFS file-system description object
+ *
+ * This function sets LEB properties for all replayed buds. Returns zero in
+ * case of success and a negative error code in case of failure.
+ */
+static int set_buds_lprops(struct ubifs_info *c)
+{
+        struct bud_entry *b;
+        int err;
+
+        list_for_each_entry(b, &c->replay_buds, list) {
+                err = set_bud_lprops(c, b);
+                if (err)
+                        return err;
+        }
+
+        return 0;
+}
+
+/**
  * trun_remove_range - apply a replay entry for a truncation to the TNC.
  * @c: UBIFS file-system description object
  * @r: replay entry of truncation
@@ -200,24 +219,22 @@ static int trun_remove_range(struct ubifs_info *c, struct replay_entry *r)
  */
 static int apply_replay_entry(struct ubifs_info *c, struct replay_entry *r)
 {
-        int err, deletion = ((r->flags & REPLAY_DELETION) != 0);
+        int err;
 
-        dbg_mnt("LEB %d:%d len %d flgs %d sqnum %llu %s", r->lnum,
-                r->offs, r->len, r->flags, r->sqnum, DBGKEY(&r->key));
+        dbg_mnt("LEB %d:%d len %d deletion %d sqnum %llu %s", r->lnum,
+                r->offs, r->len, r->deletion, r->sqnum, DBGKEY(&r->key));
 
         /* Set c->replay_sqnum to help deal with dangling branches. */
         c->replay_sqnum = r->sqnum;
 
-        if (r->flags & REPLAY_REF)
-                err = set_bud_lprops(c, r);
-        else if (is_hash_key(c, &r->key)) {
-                if (deletion)
+        if (is_hash_key(c, &r->key)) {
+                if (r->deletion)
                         err = ubifs_tnc_remove_nm(c, &r->key, &r->nm);
                 else
                         err = ubifs_tnc_add_nm(c, &r->key, r->lnum, r->offs,
                                                r->len, &r->nm);
         } else {
-                if (deletion)
+                if (r->deletion)
                         switch (key_type(c, &r->key)) {
                         case UBIFS_INO_KEY:
                         {
@@ -240,7 +257,7 @@ static int apply_replay_entry(struct ubifs_info *c, struct replay_entry *r)
                         return err;
 
                 if (c->need_recovery)
-                        err = ubifs_recover_size_accum(c, &r->key, deletion,
+                        err = ubifs_recover_size_accum(c, &r->key, r->deletion,
                                                        r->new_size);
         }
 
@@ -248,68 +265,77 @@ static int apply_replay_entry(struct ubifs_info *c, struct replay_entry *r)
 }
 
 /**
- * destroy_replay_tree - destroy the replay.
- * @c: UBIFS file-system description object
+ * replay_entries_cmp - compare 2 replay entries.
+ * @priv: UBIFS file-system description object
+ * @a: first replay entry
+ * @a: second replay entry
  *
- * Destroy the replay tree.
+ * This is a comparios function for 'list_sort()' which compares 2 replay
+ * entries @a and @b by comparing their sequence numer.  Returns %1 if @a has
+ * greater sequence number and %-1 otherwise.
  */
-static void destroy_replay_tree(struct ubifs_info *c)
+static int replay_entries_cmp(void *priv, struct list_head *a,
+                              struct list_head *b)
 {
-        struct rb_node *this = c->replay_tree.rb_node;
-        struct replay_entry *r;
-
-        while (this) {
-                if (this->rb_left) {
-                        this = this->rb_left;
-                        continue;
-                } else if (this->rb_right) {
-                        this = this->rb_right;
-                        continue;
-                }
-                r = rb_entry(this, struct replay_entry, rb);
-                this = rb_parent(this);
-                if (this) {
-                        if (this->rb_left == &r->rb)
-                                this->rb_left = NULL;
-                        else
-                                this->rb_right = NULL;
-                }
-                if (is_hash_key(c, &r->key))
-                        kfree(r->nm.name);
-                kfree(r);
-        }
-        c->replay_tree = RB_ROOT;
+        struct replay_entry *ra, *rb;
+
+        cond_resched();
+        if (a == b)
+                return 0;
+
+        ra = list_entry(a, struct replay_entry, list);
+        rb = list_entry(b, struct replay_entry, list);
+        ubifs_assert(ra->sqnum != rb->sqnum);
+        if (ra->sqnum > rb->sqnum)
+                return 1;
+        return -1;
 }
 
 /**
- * apply_replay_tree - apply the replay tree to the TNC.
+ * apply_replay_list - apply the replay list to the TNC.
  * @c: UBIFS file-system description object
  *
- * Apply the replay tree.
- * Returns zero in case of success and a negative error code in case of
- * failure.
+ * Apply all entries in the replay list to the TNC. Returns zero in case of
+ * success and a negative error code in case of failure.
  */
-static int apply_replay_tree(struct ubifs_info *c)
+static int apply_replay_list(struct ubifs_info *c)
 {
-        struct rb_node *this = rb_first(&c->replay_tree);
+        struct replay_entry *r;
+        int err;
 
-        while (this) {
-                struct replay_entry *r;
-                int err;
+        list_sort(c, &c->replay_list, &replay_entries_cmp);
 
+        list_for_each_entry(r, &c->replay_list, list) {
                 cond_resched();
 
-                r = rb_entry(this, struct replay_entry, rb);
                 err = apply_replay_entry(c, r);
                 if (err)
                         return err;
-                this = rb_next(this);
         }
+
         return 0;
 }
 
 /**
- * insert_node - insert a node to the replay tree.
+ * destroy_replay_list - destroy the replay.
+ * @c: UBIFS file-system description object
+ *
+ * Destroy the replay list.
+ */
+static void destroy_replay_list(struct ubifs_info *c)
+{
+        struct replay_entry *r, *tmp;
+
+        list_for_each_entry_safe(r, tmp, &c->replay_list, list) {
+                if (is_hash_key(c, &r->key))
+                        kfree(r->nm.name);
+                list_del(&r->list);
+                kfree(r);
+        }
+}
+
+/**
+ * insert_node - insert a node to the replay list
  * @c: UBIFS file-system description object
  * @lnum: node logical eraseblock number
  * @offs: node offset
@@ -321,39 +347,25 @@ static int apply_replay_tree(struct ubifs_info *c)
  * @old_size: truncation old size
  * @new_size: truncation new size
  *
- * This function inserts a scanned non-direntry node to the replay tree. The
- * replay tree is an RB-tree containing @struct replay_entry elements which are
- * indexed by the sequence number. The replay tree is applied at the very end
- * of the replay process. Since the tree is sorted in sequence number order,
- * the older modifications are applied first. This function returns zero in
- * case of success and a negative error code in case of failure.
+ * This function inserts a scanned non-direntry node to the replay list. The
+ * replay list contains @struct replay_entry elements, and we sort this list in
+ * sequence number order before applying it. The replay list is applied at the
+ * very end of the replay process. Since the list is sorted in sequence number
+ * order, the older modifications are applied first. This function returns zero
+ * in case of success and a negative error code in case of failure.
  */
 static int insert_node(struct ubifs_info *c, int lnum, int offs, int len,
                        union ubifs_key *key, unsigned long long sqnum,
                        int deletion, int *used, loff_t old_size,
                        loff_t new_size)
 {
-        struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL;
         struct replay_entry *r;
 
+        dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key));
+
         if (key_inum(c, key) >= c->highest_inum)
                 c->highest_inum = key_inum(c, key);
 
-        dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key));
-        while (*p) {
-                parent = *p;
-                r = rb_entry(parent, struct replay_entry, rb);
-                if (sqnum < r->sqnum) {
-                        p = &(*p)->rb_left;
-                        continue;
-                } else if (sqnum > r->sqnum) {
-                        p = &(*p)->rb_right;
-                        continue;
-                }
-                ubifs_err("duplicate sqnum in replay");
-                return -EINVAL;
-        }
-
         r = kzalloc(sizeof(struct replay_entry), GFP_KERNEL);
         if (!r)
                 return -ENOMEM;
@@ -363,19 +375,18 @@ static int insert_node(struct ubifs_info *c, int lnum, int offs, int len,
         r->lnum = lnum;
         r->offs = offs;
         r->len = len;
+        r->deletion = !!deletion;
         r->sqnum = sqnum;
-        r->flags = (deletion ? REPLAY_DELETION : 0);
+        key_copy(c, key, &r->key);
         r->old_size = old_size;
         r->new_size = new_size;
-        key_copy(c, key, &r->key);
 
-        rb_link_node(&r->rb, parent, p);
-        rb_insert_color(&r->rb, &c->replay_tree);
+        list_add_tail(&r->list, &c->replay_list);
         return 0;
 }
 
 /**
- * insert_dent - insert a directory entry node into the replay tree.
+ * insert_dent - insert a directory entry node into the replay list.
  * @c: UBIFS file-system description object
  * @lnum: node logical eraseblock number
  * @offs: node offset
@@ -387,43 +398,25 @@ static int insert_node(struct ubifs_info *c, int lnum, int offs, int len,
  * @deletion: non-zero if this is a deletion
  * @used: number of bytes in use in a LEB
  *
- * This function inserts a scanned directory entry node to the replay tree.
- * Returns zero in case of success and a negative error code in case of
- * failure.
- *
- * This function is also used for extended attribute entries because they are
- * implemented as directory entry nodes.
+ * This function inserts a scanned directory entry node or an extended
+ * attribute entry to the replay list. Returns zero in case of success and a
+ * negative error code in case of failure.
  */
 static int insert_dent(struct ubifs_info *c, int lnum, int offs, int len,
                        union ubifs_key *key, const char *name, int nlen,
                        unsigned long long sqnum, int deletion, int *used)
 {
-        struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL;
         struct replay_entry *r;
         char *nbuf;
 
+        dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key));
         if (key_inum(c, key) >= c->highest_inum)
                 c->highest_inum = key_inum(c, key);
 
-        dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key));
-        while (*p) {
-                parent = *p;
-                r = rb_entry(parent, struct replay_entry, rb);
-                if (sqnum < r->sqnum) {
-                        p = &(*p)->rb_left;
-                        continue;
-                }
-                if (sqnum > r->sqnum) {
-                        p = &(*p)->rb_right;
-                        continue;
-                }
-                ubifs_err("duplicate sqnum in replay");
-                return -EINVAL;
-        }
-
         r = kzalloc(sizeof(struct replay_entry), GFP_KERNEL);
         if (!r)
                 return -ENOMEM;
+
         nbuf = kmalloc(nlen + 1, GFP_KERNEL);
         if (!nbuf) {
                 kfree(r);
@@ -435,17 +428,15 @@ static int insert_dent(struct ubifs_info *c, int lnum, int offs, int len,
         r->lnum = lnum;
         r->offs = offs;
         r->len = len;
+        r->deletion = !!deletion;
         r->sqnum = sqnum;
+        key_copy(c, key, &r->key);
         r->nm.len = nlen;
         memcpy(nbuf, name, nlen);
         nbuf[nlen] = '\0';
         r->nm.name = nbuf;
-        r->flags = (deletion ? REPLAY_DELETION : 0);
-        key_copy(c, key, &r->key);
 
-        ubifs_assert(!*p);
-        rb_link_node(&r->rb, parent, p);
-        rb_insert_color(&r->rb, &c->replay_tree);
+        list_add_tail(&r->list, &c->replay_list);
         return 0;
 }
 
@@ -482,29 +473,91 @@ int ubifs_validate_entry(struct ubifs_info *c,
 }
 
 /**
+ * is_last_bud - check if the bud is the last in the journal head.
+ * @c: UBIFS file-system description object
+ * @bud: bud description object
+ *
+ * This function checks if bud @bud is the last bud in its journal head. This
+ * information is then used by 'replay_bud()' to decide whether the bud can
+ * have corruptions or not. Indeed, only last buds can be corrupted by power
+ * cuts. Returns %1 if this is the last bud, and %0 if not.
+ */
+static int is_last_bud(struct ubifs_info *c, struct ubifs_bud *bud)
+{
+        struct ubifs_jhead *jh = &c->jheads[bud->jhead];
+        struct ubifs_bud *next;
+        uint32_t data;
+        int err;
+
+        if (list_is_last(&bud->list, &jh->buds_list))
+                return 1;
+
+        /*
+         * The following is a quirk to make sure we work correctly with UBIFS
+         * images used with older UBIFS.
+         *
+         * Normally, the last bud will be the last in the journal head's list
+         * of bud. However, there is one exception if the UBIFS image belongs
+         * to older UBIFS. This is fairly unlikely: one would need to use old
+         * UBIFS, then have a power cut exactly at the right point, and then
+         * try to mount this image with new UBIFS.
+         *
+         * The exception is: it is possible to have 2 buds A and B, A goes
+         * before B, and B is the last, bud B is contains no data, and bud A is
+         * corrupted at the end. The reason is that in older versions when the
+         * journal code switched the next bud (from A to B), it first added a
+         * log reference node for the new bud (B), and only after this it
+         * synchronized the write-buffer of current bud (A). But later this was
+         * changed and UBIFS started to always synchronize the write-buffer of
+         * the bud (A) before writing the log reference for the new bud (B).
+         *
+         * But because older UBIFS always synchronized A's write-buffer before
+         * writing to B, we can recognize this exceptional situation but
+         * checking the contents of bud B - if it is empty, then A can be
+         * treated as the last and we can recover it.
+         *
+         * TODO: remove this piece of code in a couple of years (today it is
+         * 16.05.2011).
+         */
+        next = list_entry(bud->list.next, struct ubifs_bud, list);
+        if (!list_is_last(&next->list, &jh->buds_list))
+                return 0;
+
+        err = ubi_read(c->ubi, next->lnum, (char *)&data,
+                       next->start, 4);
+        if (err)
+                return 0;
+
+        return data == 0xFFFFFFFF;
+}
+
+/**
  * replay_bud - replay a bud logical eraseblock.
  * @c: UBIFS file-system description object
- * @lnum: bud logical eraseblock number to replay
- * @offs: bud start offset
- * @jhead: journal head to which this bud belongs
- * @free: amount of free space in the bud is returned here
- * @dirty: amount of dirty space from padding and deletion nodes is returned
- * here
+ * @b: bud entry which describes the bud
  *
- * This function returns zero in case of success and a negative error code in
- * case of failure.
+ * This function replays bud @bud, recovers it if needed, and adds all nodes
+ * from this bud to the replay list. Returns zero in case of success and a
+ * negative error code in case of failure.
  */
-static int replay_bud(struct ubifs_info *c, int lnum, int offs, int jhead,
-                      int *free, int *dirty)
+static int replay_bud(struct ubifs_info *c, struct bud_entry *b)
 {
-        int err = 0, used = 0;
+        int is_last = is_last_bud(c, b->bud);
+        int err = 0, used = 0, lnum = b->bud->lnum, offs = b->bud->start;
         struct ubifs_scan_leb *sleb;
         struct ubifs_scan_node *snod;
-        struct ubifs_bud *bud;
 
-        dbg_mnt("replay bud LEB %d, head %d", lnum, jhead);
-        if (c->need_recovery)
-                sleb = ubifs_recover_leb(c, lnum, offs, c->sbuf, jhead != GCHD);
+        dbg_mnt("replay bud LEB %d, head %d, offs %d, is_last %d",
+                lnum, b->bud->jhead, offs, is_last);
+
+        if (c->need_recovery && is_last)
+                /*
+                 * Recover only last LEBs in the journal heads, because power
+                 * cuts may cause corruptions only in these LEBs, because only
+                 * these LEBs could possibly be written to at the power cut
+                 * time.
+                 */
+                sleb = ubifs_recover_leb(c, lnum, offs, c->sbuf, b->bud->jhead);
         else
                 sleb = ubifs_scan(c, lnum, offs, c->sbuf, 0);
         if (IS_ERR(sleb))
@@ -620,20 +673,13 @@ static int replay_bud(struct ubifs_info *c, int lnum, int offs, int jhead,
                         goto out;
         }
 
-        bud = ubifs_search_bud(c, lnum);
-        if (!bud)
-                BUG();
-
+        ubifs_assert(ubifs_search_bud(c, lnum));
         ubifs_assert(sleb->endpt - offs >= used);
         ubifs_assert(sleb->endpt % c->min_io_size == 0);
 
-        if (sleb->endpt + c->min_io_size <= c->leb_size &&
-            !(c->vfs_sb->s_flags & MS_RDONLY))
-                err = ubifs_wbuf_seek_nolock(&c->jheads[jhead].wbuf, lnum,
-                                             sleb->endpt, UBI_SHORTTERM);
-
-        *dirty = sleb->endpt - offs - used;
-        *free = c->leb_size - sleb->endpt;
+        b->dirty = sleb->endpt - offs - used;
+        b->free = c->leb_size - sleb->endpt;
+        dbg_mnt("bud LEB %d replied: dirty %d, free %d", lnum, b->dirty, b->free);
 
 out:
         ubifs_scan_destroy(sleb);
@@ -647,55 +693,6 @@ out_dump:
 }
 
 /**
- * insert_ref_node - insert a reference node to the replay tree.
- * @c: UBIFS file-system description object
- * @lnum: node logical eraseblock number
- * @offs: node offset
- * @sqnum: sequence number
- * @free: amount of free space in bud
- * @dirty: amount of dirty space from padding and deletion nodes
- *
- * This function inserts a reference node to the replay tree and returns zero
- * in case of success or a negative error code in case of failure.
- */
-static int insert_ref_node(struct ubifs_info *c, int lnum, int offs,
-                           unsigned long long sqnum, int free, int dirty)
-{
-        struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL;
-        struct replay_entry *r;
-
-        dbg_mnt("add ref LEB %d:%d", lnum, offs);
-        while (*p) {
-                parent = *p;
-                r = rb_entry(parent, struct replay_entry, rb);
-                if (sqnum < r->sqnum) {
-                        p = &(*p)->rb_left;
-                        continue;
-                } else if (sqnum > r->sqnum) {
-                        p = &(*p)->rb_right;
-                        continue;
-                }
-                ubifs_err("duplicate sqnum in replay tree");
-                return -EINVAL;
-        }
-
-        r = kzalloc(sizeof(struct replay_entry), GFP_KERNEL);
-        if (!r)
-                return -ENOMEM;
-
-        r->lnum = lnum;
-        r->offs = offs;
-        r->sqnum = sqnum;
-        r->flags = REPLAY_REF;
-        r->free = free;
-        r->dirty = dirty;
-
-        rb_link_node(&r->rb, parent, p);
-        rb_insert_color(&r->rb, &c->replay_tree);
-        return 0;
-}
-
-/**
  * replay_buds - replay all buds.
  * @c: UBIFS file-system description object
  *
@@ -705,17 +702,16 @@ static int insert_ref_node(struct ubifs_info *c, int lnum, int offs,
 static int replay_buds(struct ubifs_info *c)
 {
         struct bud_entry *b;
-        int err, uninitialized_var(free), uninitialized_var(dirty);
+        int err;
+        unsigned long long prev_sqnum = 0;
 
         list_for_each_entry(b, &c->replay_buds, list) {
-                err = replay_bud(c, b->bud->lnum, b->bud->start, b->bud->jhead,
-                                 &free, &dirty);
-                if (err)
-                        return err;
-                err = insert_ref_node(c, b->bud->lnum, b->bud->start, b->sqnum,
-                                      free, dirty);
+                err = replay_bud(c, b);
                 if (err)
                         return err;
+
+                ubifs_assert(b->sqnum > prev_sqnum);
+                prev_sqnum = b->sqnum;
         }
 
         return 0;
@@ -840,6 +836,11 @@ static int replay_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf)
         if (IS_ERR(sleb)) {
                 if (PTR_ERR(sleb) != -EUCLEAN || !c->need_recovery)
                         return PTR_ERR(sleb);
+                /*
+                 * Note, the below function will recover this log LEB only if
+                 * it is the last, because unclean reboots can possibly corrupt
+                 * only the tail of the log.
+                 */
                 sleb = ubifs_recover_log_leb(c, lnum, offs, sbuf);
                 if (IS_ERR(sleb))
                         return PTR_ERR(sleb);
@@ -851,7 +852,6 @@ static int replay_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf)
         }
 
         node = sleb->buf;
-
         snod = list_entry(sleb->nodes.next, struct ubifs_scan_node, list);
         if (c->cs_sqnum == 0) {
                 /*
@@ -898,7 +898,6 @@ static int replay_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf)
         }
 
         list_for_each_entry(snod, &sleb->nodes, list) {
-
                 cond_resched();
 
                 if (snod->sqnum >= SQNUM_WATERMARK) {
@@ -1011,7 +1010,6 @@ out:
 int ubifs_replay_journal(struct ubifs_info *c)
 {
         int err, i, lnum, offs, free;
-        void *sbuf = NULL;
 
         BUILD_BUG_ON(UBIFS_TRUN_KEY > 5);
 
@@ -1026,14 +1024,8 @@ int ubifs_replay_journal(struct ubifs_info *c)
                 return -EINVAL;
         }
 
-        sbuf = vmalloc(c->leb_size);
-        if (!sbuf)
-                return -ENOMEM;
-
         dbg_mnt("start replaying the journal");
-
         c->replaying = 1;
-
         lnum = c->ltail_lnum = c->lhead_lnum;
         offs = c->lhead_offs;
 
@@ -1046,7 +1038,7 @@ int ubifs_replay_journal(struct ubifs_info *c)
                         lnum = UBIFS_LOG_LNUM;
                         offs = 0;
                 }
-                err = replay_log_leb(c, lnum, offs, sbuf);
+                err = replay_log_leb(c, lnum, offs, c->sbuf);
                 if (err == 1)
                         /* We hit the end of the log */
                         break;
@@ -1059,27 +1051,30 @@ int ubifs_replay_journal(struct ubifs_info *c)
         if (err)
                 goto out;
 
-        err = apply_replay_tree(c);
+        err = apply_replay_list(c);
+        if (err)
+                goto out;
+
+        err = set_buds_lprops(c);
         if (err)
                 goto out;
 
         /*
-         * UBIFS budgeting calculations use @c->budg_uncommitted_idx variable
-         * to roughly estimate index growth. Things like @c->min_idx_lebs
+         * UBIFS budgeting calculations use @c->bi.uncommitted_idx variable
+         * to roughly estimate index growth. Things like @c->bi.min_idx_lebs
          * depend on it. This means we have to initialize it to make sure
          * budgeting works properly.
          */
-        c->budg_uncommitted_idx = atomic_long_read(&c->dirty_zn_cnt);
-        c->budg_uncommitted_idx *= c->max_idx_node_sz;
+        c->bi.uncommitted_idx = atomic_long_read(&c->dirty_zn_cnt);
+        c->bi.uncommitted_idx *= c->max_idx_node_sz;
 
         ubifs_assert(c->bud_bytes <= c->max_bud_bytes || c->need_recovery);
         dbg_mnt("finished, log head LEB %d:%d, max_sqnum %llu, "
                 "highest_inum %lu", c->lhead_lnum, c->lhead_offs, c->max_sqnum,
                 (unsigned long)c->highest_inum);
 out:
-        destroy_replay_tree(c);
+        destroy_replay_list(c);
         destroy_bud_list(c);
-        vfree(sbuf);
         c->replaying = 0;
         return err;
 }
diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c
index 96cb62c8a9dd..c606f010e8df 100644
--- a/fs/ubifs/sb.c
+++ b/fs/ubifs/sb.c
@@ -475,7 +475,8 @@ failed:
  * @c: UBIFS file-system description object
  *
  * This function returns a pointer to the superblock node or a negative error
- * code.
+ * code. Note, the user of this function is responsible of kfree()'ing the
+ * returned superblock buffer.
  */
 struct ubifs_sb_node *ubifs_read_sb_node(struct ubifs_info *c)
 {
@@ -542,11 +543,8 @@ int ubifs_read_superblock(struct ubifs_info *c)
          * due to the unavailability of time-travelling equipment.
          */
         if (c->fmt_version > UBIFS_FORMAT_VERSION) {
-                struct super_block *sb = c->vfs_sb;
-                int mounting_ro = sb->s_flags & MS_RDONLY;
-
-                ubifs_assert(!c->ro_media || mounting_ro);
-                if (!mounting_ro ||
+                ubifs_assert(!c->ro_media || c->ro_mount);
+                if (!c->ro_mount ||
                     c->ro_compat_version > UBIFS_RO_COMPAT_VERSION) {
                         ubifs_err("on-flash format version is w%d/r%d, but "
                                   "software only supports up to version "
@@ -619,12 +617,13 @@ int ubifs_read_superblock(struct ubifs_info *c)
         c->vfs_sb->s_time_gran = le32_to_cpu(sup->time_gran);
         memcpy(&c->uuid, &sup->uuid, 16);
         c->big_lpt = !!(sup_flags & UBIFS_FLG_BIGLPT);
+        c->space_fixup = !!(sup_flags & UBIFS_FLG_SPACE_FIXUP);
 
         /* Automatically increase file system size to the maximum size */
         c->old_leb_cnt = c->leb_cnt;
         if (c->leb_cnt < c->vi.size && c->leb_cnt < c->max_leb_cnt) {
                 c->leb_cnt = min_t(int, c->max_leb_cnt, c->vi.size);
-                if (c->vfs_sb->s_flags & MS_RDONLY)
+                if (c->ro_mount)
                         dbg_mnt("Auto resizing (ro) from %d LEBs to %d LEBs",
                                 c->old_leb_cnt, c->leb_cnt);
                 else {
@@ -653,3 +652,152 @@ out:
         kfree(sup);
         return err;
 }
+
+/**
+ * fixup_leb - fixup/unmap an LEB containing free space.
+ * @c: UBIFS file-system description object
+ * @lnum: the LEB number to fix up
+ * @len: number of used bytes in LEB (starting at offset 0)
+ *
+ * This function reads the contents of the given LEB number @lnum, then fixes
+ * it up, so that empty min. I/O units in the end of LEB are actually erased on
+ * flash (rather than being just all-0xff real data). If the LEB is completely
+ * empty, it is simply unmapped.
+ */
+static int fixup_leb(struct ubifs_info *c, int lnum, int len)
+{
+        int err;
+
+        ubifs_assert(len >= 0);
+        ubifs_assert(len % c->min_io_size == 0);
+        ubifs_assert(len < c->leb_size);
+
+        if (len == 0) {
+                dbg_mnt("unmap empty LEB %d", lnum);
+                return ubi_leb_unmap(c->ubi, lnum);
+        }
+
+        dbg_mnt("fixup LEB %d, data len %d", lnum, len);
+        err = ubi_read(c->ubi, lnum, c->sbuf, 0, len);
+        if (err)
+                return err;
+
+        return ubi_leb_change(c->ubi, lnum, c->sbuf, len, UBI_UNKNOWN);
+}
+
+/**
+ * fixup_free_space - find & remap all LEBs containing free space.
+ * @c: UBIFS file-system description object
+ *
+ * This function walks through all LEBs in the filesystem and fiexes up those
+ * containing free/empty space.
+ */
+static int fixup_free_space(struct ubifs_info *c)
+{
+        int lnum, err = 0;
+        struct ubifs_lprops *lprops;
+
+        ubifs_get_lprops(c);
+
+        /* Fixup LEBs in the master area */
+        for (lnum = UBIFS_MST_LNUM; lnum < UBIFS_LOG_LNUM; lnum++) {
+                err = fixup_leb(c, lnum, c->mst_offs + c->mst_node_alsz);
+                if (err)
+                        goto out;
+        }
+
+        /* Unmap unused log LEBs */
+        lnum = ubifs_next_log_lnum(c, c->lhead_lnum);
+        while (lnum != c->ltail_lnum) {
+                err = fixup_leb(c, lnum, 0);
+                if (err)
+                        goto out;
+                lnum = ubifs_next_log_lnum(c, lnum);
+        }
+
+        /* Fixup the current log head */
+        err = fixup_leb(c, c->lhead_lnum, c->lhead_offs);
+        if (err)
+                goto out;
+
+        /* Fixup LEBs in the LPT area */
+        for (lnum = c->lpt_first; lnum <= c->lpt_last; lnum++) {
+                int free = c->ltab[lnum - c->lpt_first].free;
+
+                if (free > 0) {
+                        err = fixup_leb(c, lnum, c->leb_size - free);
+                        if (err)
+                                goto out;
+                }
+        }
+
+        /* Unmap LEBs in the orphans area */
+        for (lnum = c->orph_first; lnum <= c->orph_last; lnum++) {
+                err = fixup_leb(c, lnum, 0);
+                if (err)
+                        goto out;
+        }
+
+        /* Fixup LEBs in the main area */
+        for (lnum = c->main_first; lnum < c->leb_cnt; lnum++) {
+                lprops = ubifs_lpt_lookup(c, lnum);
+                if (IS_ERR(lprops)) {
+                        err = PTR_ERR(lprops);
+                        goto out;
+                }
+
+                if (lprops->free > 0) {
+                        err = fixup_leb(c, lnum, c->leb_size - lprops->free);
+                        if (err)
+                                goto out;
+                }
+        }
+
+out:
+        ubifs_release_lprops(c);
+        return err;
+}
+
+/**
+ * ubifs_fixup_free_space - find & fix all LEBs with free space.
+ * @c: UBIFS file-system description object
+ *
+ * This function fixes up LEBs containing free space on first mount, if the
+ * appropriate flag was set when the FS was created. Each LEB with one or more
+ * empty min. I/O unit (i.e. free-space-count > 0) is re-written, to make sure
+ * the free space is actually erased. E.g., this is necessary for some NAND
+ * chips, since the free space may have been programmed like real "0xff" data
+ * (generating a non-0xff ECC), causing future writes to the not-really-erased
+ * NAND pages to behave badly. After the space is fixed up, the superblock flag
+ * is cleared, so that this is skipped for all future mounts.
+ */
+int ubifs_fixup_free_space(struct ubifs_info *c)
+{
+        int err;
+        struct ubifs_sb_node *sup;
+
+        ubifs_assert(c->space_fixup);
+        ubifs_assert(!c->ro_mount);
+
+        ubifs_msg("start fixing up free space");
+
+        err = fixup_free_space(c);
+        if (err)
+                return err;
+
+        sup = ubifs_read_sb_node(c);
+        if (IS_ERR(sup))
+                return PTR_ERR(sup);
+
+        /* Free-space fixup is no longer required */
+        c->space_fixup = 0;
+        sup->flags &= cpu_to_le32(~UBIFS_FLG_SPACE_FIXUP);
+
+        err = ubifs_write_sb_node(c, sup);
+        kfree(sup);
+        if (err)
+                return err;
+
+        ubifs_msg("free space fixup complete");
+        return err;
+}
diff --git a/fs/ubifs/scan.c b/fs/ubifs/scan.c
index 96c525384191..36216b46f772 100644
--- a/fs/ubifs/scan.c
+++ b/fs/ubifs/scan.c
@@ -197,7 +197,7 @@ int ubifs_add_snod(const struct ubifs_info *c, struct ubifs_scan_leb *sleb,
         struct ubifs_ino_node *ino = buf;
         struct ubifs_scan_node *snod;
 
-        snod = kzalloc(sizeof(struct ubifs_scan_node), GFP_NOFS);
+        snod = kmalloc(sizeof(struct ubifs_scan_node), GFP_NOFS);
         if (!snod)
                 return -ENOMEM;
 
@@ -212,13 +212,15 @@ int ubifs_add_snod(const struct ubifs_info *c, struct ubifs_scan_leb *sleb,
         case UBIFS_DENT_NODE:
         case UBIFS_XENT_NODE:
         case UBIFS_DATA_NODE:
-        case UBIFS_TRUN_NODE:
                 /*
                  * The key is in the same place in all keyed
                  * nodes.
                  */
                 key_read(c, &ino->key, &snod->key);
                 break;
+        default:
+                invalid_key_init(c, &snod->key);
+                break;
         }
         list_add_tail(&snod->list, &sleb->nodes);
         sleb->nodes_cnt += 1;
@@ -326,7 +328,7 @@ struct ubifs_scan_leb *ubifs_scan(const struct ubifs_info *c, int lnum,
                 if (!quiet)
                         ubifs_err("empty space starts at non-aligned offset %d",
                                   offs);
-                goto corrupted;;
+                goto corrupted;
         }
 
         ubifs_end_scan(c, sleb, lnum, offs);
diff --git a/fs/ubifs/shrinker.c b/fs/ubifs/shrinker.c
index 0b201114a5ad..9e1d05666fed 100644
--- a/fs/ubifs/shrinker.c
+++ b/fs/ubifs/shrinker.c
@@ -250,7 +250,7 @@ static int kick_a_thread(void)
                         dirty_zn_cnt = atomic_long_read(&c->dirty_zn_cnt);
 
                         if (!dirty_zn_cnt || c->cmt_state == COMMIT_BROKEN ||
-                            c->ro_media) {
+                            c->ro_mount || c->ro_error) {
                                 mutex_unlock(&c->umount_mutex);
                                 continue;
                         }
@@ -277,13 +277,18 @@ static int kick_a_thread(void)
         return 0;
 }
 
-int ubifs_shrinker(struct shrinker *shrink, int nr, gfp_t gfp_mask)
+int ubifs_shrinker(struct shrinker *shrink, struct shrink_control *sc)
 {
+        int nr = sc->nr_to_scan;
         int freed, contention = 0;
         long clean_zn_cnt = atomic_long_read(&ubifs_clean_zn_cnt);
 
         if (nr == 0)
-                return clean_zn_cnt;
+                /*
+                 * Due to the way UBIFS updates the clean znode counter it may
+                 * temporarily be negative.
+                 */
+                return clean_zn_cnt >= 0 ? clean_zn_cnt : 1;
 
         if (!clean_zn_cnt) {
                 /*
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index cd5900b85d38..529be0582029 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -272,12 +272,20 @@ static struct inode *ubifs_alloc_inode(struct super_block *sb)
         return &ui->vfs_inode;
 };
 
+static void ubifs_i_callback(struct rcu_head *head)
+{
+        struct inode *inode = container_of(head, struct inode, i_rcu);
+        struct ubifs_inode *ui = ubifs_inode(inode);
+        INIT_LIST_HEAD(&inode->i_dentry);
+        kmem_cache_free(ubifs_inode_slab, ui);
+}
+
 static void ubifs_destroy_inode(struct inode *inode)
 {
         struct ubifs_inode *ui = ubifs_inode(inode);
 
         kfree(ui->data);
-        kmem_cache_free(ubifs_inode_slab, inode);
+        call_rcu(&inode->i_rcu, ubifs_i_callback);
 }
 
 /*
@@ -367,14 +375,14 @@ out:
                 ubifs_release_dirty_inode_budget(c, ui);
         else {
                 /* We've deleted something - clean the "no space" flags */
-                c->nospace = c->nospace_rp = 0;
+                c->bi.nospace = c->bi.nospace_rp = 0;
                 smp_wmb();
         }
 done:
         end_writeback(inode);
 }
 
-static void ubifs_dirty_inode(struct inode *inode)
+static void ubifs_dirty_inode(struct inode *inode, int flags)
 {
         struct ubifs_inode *ui = ubifs_inode(inode);
 
@@ -504,9 +512,12 @@ static int init_constants_early(struct ubifs_info *c)
 
         c->leb_cnt = c->vi.size;
         c->leb_size = c->vi.usable_leb_size;
+        c->leb_start = c->di.leb_start;
         c->half_leb_size = c->leb_size / 2;
         c->min_io_size = c->di.min_io_size;
         c->min_io_shift = fls(c->min_io_size) - 1;
+        c->max_write_size = c->di.max_write_size;
+        c->max_write_shift = fls(c->max_write_size) - 1;
 
         if (c->leb_size < UBIFS_MIN_LEB_SZ) {
                 ubifs_err("too small LEBs (%d bytes), min. is %d bytes",
@@ -526,6 +537,18 @@ static int init_constants_early(struct ubifs_info *c)
         }
 
         /*
+         * Maximum write size has to be greater or equivalent to min. I/O
+         * size, and be multiple of min. I/O size.
+         */
+        if (c->max_write_size < c->min_io_size ||
+            c->max_write_size % c->min_io_size ||
+            !is_power_of_2(c->max_write_size)) {
+                ubifs_err("bad write buffer size %d for %d min. I/O unit",
+                          c->max_write_size, c->min_io_size);
+                return -EINVAL;
+        }
+
+        /*
          * UBIFS aligns all node to 8-byte boundary, so to make function in
          * io.c simpler, assume minimum I/O unit size to be 8 bytes if it is
          * less than 8.
@@ -533,6 +556,10 @@ static int init_constants_early(struct ubifs_info *c)
         if (c->min_io_size < 8) {
                 c->min_io_size = 8;
                 c->min_io_shift = 3;
+                if (c->max_write_size < c->min_io_size) {
+                        c->max_write_size = c->min_io_size;
+                        c->max_write_shift = c->min_io_shift;
+                }
         }
 
         c->ref_node_alsz = ALIGN(UBIFS_REF_NODE_SZ, c->min_io_size);
@@ -667,11 +694,11 @@ static int init_constants_sb(struct ubifs_info *c)
          * be compressed and direntries are of the maximum size.
          *
          * Note, data, which may be stored in inodes is budgeted separately, so
-         * it is not included into 'c->inode_budget'.
+         * it is not included into 'c->bi.inode_budget'.
          */
-        c->page_budget = UBIFS_MAX_DATA_NODE_SZ * UBIFS_BLOCKS_PER_PAGE;
-        c->inode_budget = UBIFS_INO_NODE_SZ;
-        c->dent_budget = UBIFS_MAX_DENT_NODE_SZ;
+        c->bi.page_budget = UBIFS_MAX_DATA_NODE_SZ * UBIFS_BLOCKS_PER_PAGE;
+        c->bi.inode_budget = UBIFS_INO_NODE_SZ;
+        c->bi.dent_budget = UBIFS_MAX_DENT_NODE_SZ;
 
         /*
          * When the amount of flash space used by buds becomes
@@ -715,7 +742,7 @@ static void init_constants_master(struct ubifs_info *c)
 {
         long long tmp64;
 
-        c->min_idx_lebs = ubifs_calc_min_idx_lebs(c);
+        c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c);
         c->report_rp_size = ubifs_reported_space(c, c->rp_size);
 
         /*
@@ -784,15 +811,18 @@ static int alloc_wbufs(struct ubifs_info *c)
 
                 c->jheads[i].wbuf.sync_callback = &bud_wbuf_callback;
                 c->jheads[i].wbuf.jhead = i;
+                c->jheads[i].grouped = 1;
         }
 
         c->jheads[BASEHD].wbuf.dtype = UBI_SHORTTERM;
         /*
          * Garbage Collector head likely contains long-term data and
-         * does not need to be synchronized by timer.
+         * does not need to be synchronized by timer. Also GC head nodes are
+         * not grouped.
          */
         c->jheads[GCHD].wbuf.dtype = UBI_LONGTERM;
         c->jheads[GCHD].wbuf.no_timer = 1;
+        c->jheads[GCHD].grouped = 0;
 
         return 0;
 }
@@ -1117,8 +1147,8 @@ static int check_free_space(struct ubifs_info *c)
 {
         ubifs_assert(c->dark_wm > 0);
         if (c->lst.total_free + c->lst.total_dirty < c->dark_wm) {
-                ubifs_err("insufficient free space to mount in read/write mode");
-                dbg_dump_budg(c);
+                ubifs_err("insufficient free space to mount in R/W mode");
+                dbg_dump_budg(c, &c->bi);
                 dbg_dump_lprops(c);
                 return -ENOSPC;
         }
@@ -1137,11 +1167,11 @@ static int check_free_space(struct ubifs_info *c)
  */
 static int mount_ubifs(struct ubifs_info *c)
 {
-        struct super_block *sb = c->vfs_sb;
-        int err, mounted_read_only = (sb->s_flags & MS_RDONLY);
+        int err;
         long long x;
         size_t sz;
 
+        c->ro_mount = !!(c->vfs_sb->s_flags & MS_RDONLY);
         err = init_constants_early(c);
         if (err)
                 return err;
@@ -1154,7 +1184,7 @@ static int mount_ubifs(struct ubifs_info *c)
         if (err)
                 goto out_free;
 
-        if (c->empty && (mounted_read_only || c->ro_media)) {
+        if (c->empty && (c->ro_mount || c->ro_media)) {
                 /*
                  * This UBI volume is empty, and read-only, or the file system
                  * is mounted read-only - we cannot format it.
@@ -1165,7 +1195,7 @@ static int mount_ubifs(struct ubifs_info *c)
                 goto out_free;
         }
 
-        if (c->ro_media && !mounted_read_only) {
+        if (c->ro_media && !c->ro_mount) {
                 ubifs_err("cannot mount read-write - read-only media");
                 err = -EROFS;
                 goto out_free;
@@ -1185,7 +1215,7 @@ static int mount_ubifs(struct ubifs_info *c)
         if (!c->sbuf)
                 goto out_free;
 
-        if (!mounted_read_only) {
+        if (!c->ro_mount) {
                 c->ileb_buf = vmalloc(c->leb_size);
                 if (!c->ileb_buf)
                         goto out_free;
@@ -1194,11 +1224,14 @@ static int mount_ubifs(struct ubifs_info *c)
         if (c->bulk_read == 1)
                 bu_init(c);
 
-        /*
-         * We have to check all CRCs, even for data nodes, when we mount the FS
-         * (specifically, when we are replaying).
-         */
-        c->always_chk_crc = 1;
+        if (!c->ro_mount) {
+                c->write_reserve_buf = kmalloc(COMPRESSED_DATA_NODE_BUF_SZ,
+                                               GFP_KERNEL);
+                if (!c->write_reserve_buf)
+                        goto out_free;
+        }
+
+        c->mounting = 1;
 
         err = ubifs_read_superblock(c);
         if (err)
@@ -1227,12 +1260,12 @@ static int mount_ubifs(struct ubifs_info *c)
                 goto out_free;
         }
 
-        sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, c->vi.vol_id);
-        if (!mounted_read_only) {
-                err = alloc_wbufs(c);
-                if (err)
-                        goto out_cbuf;
+        err = alloc_wbufs(c);
+        if (err)
+                goto out_cbuf;
 
+        sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, c->vi.vol_id);
+        if (!c->ro_mount) {
                 /* Create background thread */
                 c->bgt = kthread_create(ubifs_bg_thread, c, "%s", c->bgt_name);
                 if (IS_ERR(c->bgt)) {
@@ -1254,12 +1287,25 @@ static int mount_ubifs(struct ubifs_info *c)
         if ((c->mst_node->flags & cpu_to_le32(UBIFS_MST_DIRTY)) != 0) {
                 ubifs_msg("recovery needed");
                 c->need_recovery = 1;
-                if (!mounted_read_only) {
-                        err = ubifs_recover_inl_heads(c, c->sbuf);
-                        if (err)
-                                goto out_master;
-                }
-        } else if (!mounted_read_only) {
+        }
+
+        if (c->need_recovery && !c->ro_mount) {
+                err = ubifs_recover_inl_heads(c, c->sbuf);
+                if (err)
+                        goto out_master;
+        }
+
+        err = ubifs_lpt_init(c, 1, !c->ro_mount);
+        if (err)
+                goto out_master;
+
+        if (!c->ro_mount && c->space_fixup) {
+                err = ubifs_fixup_free_space(c);
+                if (err)
+                        goto out_master;
+        }
+
+        if (!c->ro_mount) {
                 /*
                  * Set the "dirty" flag so that if we reboot uncleanly we
                  * will notice this immediately on the next mount.
@@ -1267,14 +1313,10 @@ static int mount_ubifs(struct ubifs_info *c)
                 c->mst_node->flags |= cpu_to_le32(UBIFS_MST_DIRTY);
                 err = ubifs_write_master(c);
                 if (err)
-                        goto out_master;
+                        goto out_lpt;
         }
 
-        err = ubifs_lpt_init(c, 1, !mounted_read_only);
-        if (err)
-                goto out_lpt;
-
-        err = dbg_check_idx_size(c, c->old_idx_sz);
+        err = dbg_check_idx_size(c, c->bi.old_idx_sz);
         if (err)
                 goto out_lpt;
 
@@ -1283,13 +1325,13 @@ static int mount_ubifs(struct ubifs_info *c)
                 goto out_journal;
 
         /* Calculate 'min_idx_lebs' after journal replay */
-        c->min_idx_lebs = ubifs_calc_min_idx_lebs(c);
+        c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c);
 
-        err = ubifs_mount_orphans(c, c->need_recovery, mounted_read_only);
+        err = ubifs_mount_orphans(c, c->need_recovery, c->ro_mount);
         if (err)
                 goto out_orphans;
 
-        if (!mounted_read_only) {
+        if (!c->ro_mount) {
                 int lnum;
 
                 err = check_free_space(c);
@@ -1351,7 +1393,7 @@ static int mount_ubifs(struct ubifs_info *c)
         spin_unlock(&ubifs_infos_lock);
 
         if (c->need_recovery) {
-                if (mounted_read_only)
+                if (c->ro_mount)
                         ubifs_msg("recovery deferred");
                 else {
                         c->need_recovery = 0;
@@ -1374,11 +1416,11 @@ static int mount_ubifs(struct ubifs_info *c)
         if (err)
                 goto out_infos;
 
-        c->always_chk_crc = 0;
+        c->mounting = 0;
 
         ubifs_msg("mounted UBI device %d, volume %d, name \"%s\"",
                   c->vi.ubi_num, c->vi.vol_id, c->vi.name);
-        if (mounted_read_only)
+        if (c->ro_mount)
                 ubifs_msg("mounted read-only");
         x = (long long)c->main_lebs * c->leb_size;
         ubifs_msg("file system size:   %lld bytes (%lld KiB, %lld MiB, %d "
@@ -1395,6 +1437,7 @@ static int mount_ubifs(struct ubifs_info *c)
 
         dbg_msg("compiled on:         " __DATE__ " at " __TIME__);
         dbg_msg("min. I/O unit size:  %d bytes", c->min_io_size);
+        dbg_msg("max. write size:     %d bytes", c->max_write_size);
         dbg_msg("LEB size:            %d bytes (%d KiB)",
                 c->leb_size, c->leb_size >> 10);
         dbg_msg("data journal heads:  %d",
@@ -1411,7 +1454,8 @@ static int mount_ubifs(struct ubifs_info *c)
                 c->main_lebs, c->main_first, c->leb_cnt - 1);
         dbg_msg("index LEBs:          %d", c->lst.idx_lebs);
         dbg_msg("total index bytes:   %lld (%lld KiB, %lld MiB)",
-                c->old_idx_sz, c->old_idx_sz >> 10, c->old_idx_sz >> 20);
+                c->bi.old_idx_sz, c->bi.old_idx_sz >> 10,
+                c->bi.old_idx_sz >> 20);
         dbg_msg("key hash type:       %d", c->key_hash_type);
         dbg_msg("tree fanout:         %d", c->fanout);
         dbg_msg("reserved GC LEB:     %d", c->gc_lnum);
@@ -1424,9 +1468,9 @@ static int mount_ubifs(struct ubifs_info *c)
                 UBIFS_TRUN_NODE_SZ, UBIFS_SB_NODE_SZ, UBIFS_MST_NODE_SZ);
         dbg_msg("node sizes:          ref %zu, cmt. start %zu, orph %zu",
                 UBIFS_REF_NODE_SZ, UBIFS_CS_NODE_SZ, UBIFS_ORPH_NODE_SZ);
-        dbg_msg("max. node sizes:     data %zu, inode %zu dentry %zu",
-                UBIFS_MAX_DATA_NODE_SZ, UBIFS_MAX_INO_NODE_SZ,
-                UBIFS_MAX_DENT_NODE_SZ);
+        dbg_msg("max. node sizes:     data %zu, inode %zu dentry %zu, idx %d",
+                UBIFS_MAX_DATA_NODE_SZ, UBIFS_MAX_INO_NODE_SZ,
+                UBIFS_MAX_DENT_NODE_SZ, ubifs_idx_node_sz(c, c->fanout));
         dbg_msg("dead watermark:      %d", c->dead_wm);
         dbg_msg("dark watermark:      %d", c->dark_wm);
         dbg_msg("LEB overhead:        %d", c->leb_overhead);
@@ -1466,6 +1510,7 @@ out_wbufs:
 out_cbuf:
         kfree(c->cbuf);
 out_free:
+        kfree(c->write_reserve_buf);
         kfree(c->bu.buf);
         vfree(c->ileb_buf);
         vfree(c->sbuf);
@@ -1504,6 +1549,7 @@ static void ubifs_umount(struct ubifs_info *c)
         kfree(c->cbuf);
         kfree(c->rcvrd_mst_node);
         kfree(c->mst_node);
+        kfree(c->write_reserve_buf);
         kfree(c->bu.buf);
         vfree(c->ileb_buf);
         vfree(c->sbuf);
@@ -1535,7 +1581,7 @@ static int ubifs_remount_rw(struct ubifs_info *c)
         mutex_lock(&c->umount_mutex);
         dbg_save_space_info(c);
         c->remounting_rw = 1;
-        c->always_chk_crc = 1;
+        c->ro_mount = 0;
 
         err = check_free_space(c);
         if (err)
@@ -1551,6 +1597,7 @@ static int ubifs_remount_rw(struct ubifs_info *c)
                 }
                 sup->leb_cnt = cpu_to_le32(c->leb_cnt);
                 err = ubifs_write_sb_node(c, sup);
+                kfree(sup);
                 if (err)
                         goto out;
         }
@@ -1590,16 +1637,14 @@ static int ubifs_remount_rw(struct ubifs_info *c)
                 goto out;
         }
 
-        err = ubifs_lpt_init(c, 0, 1);
-        if (err)
+        c->write_reserve_buf = kmalloc(COMPRESSED_DATA_NODE_BUF_SZ, GFP_KERNEL);
+        if (!c->write_reserve_buf)
                 goto out;
 
-        err = alloc_wbufs(c);
+        err = ubifs_lpt_init(c, 0, 1);
         if (err)
                 goto out;
 
-        ubifs_create_buds_lists(c);
-
         /* Create background thread */
         c->bgt = kthread_create(ubifs_bg_thread, c, "%s", c->bgt_name);
         if (IS_ERR(c->bgt)) {
@@ -1634,20 +1679,37 @@ static int ubifs_remount_rw(struct ubifs_info *c)
         if (err)
                 goto out;
 
+        dbg_gen("re-mounted read-write");
+        c->remounting_rw = 0;
+
         if (c->need_recovery) {
                 c->need_recovery = 0;
                 ubifs_msg("deferred recovery completed");
+        } else {
+                /*
+                 * Do not run the debugging space check if the were doing
+                 * recovery, because when we saved the information we had the
+                 * file-system in a state where the TNC and lprops has been
+                 * modified in memory, but all the I/O operations (including a
+                 * commit) were deferred. So the file-system was in
+                 * "non-committed" state. Now the file-system is in committed
+                 * state, and of course the amount of free space will change
+                 * because, for example, the old index size was imprecise.
+                 */
+                err = dbg_check_space_info(c);
+        }
+
+        if (c->space_fixup) {
+                err = ubifs_fixup_free_space(c);
+                if (err)
+                        goto out;
         }
 
-        dbg_gen("re-mounted read-write");
-        c->vfs_sb->s_flags &= ~MS_RDONLY;
-        c->remounting_rw = 0;
-        c->always_chk_crc = 0;
-        err = dbg_check_space_info(c);
         mutex_unlock(&c->umount_mutex);
         return err;
 
 out:
+        c->ro_mount = 1;
         vfree(c->orph_buf);
         c->orph_buf = NULL;
         if (c->bgt) {
@@ -1655,11 +1717,12 @@ out:
                 c->bgt = NULL;
         }
         free_wbufs(c);
+        kfree(c->write_reserve_buf);
+        c->write_reserve_buf = NULL;
         vfree(c->ileb_buf);
         c->ileb_buf = NULL;
         ubifs_lpt_free(c, 1);
         c->remounting_rw = 0;
-        c->always_chk_crc = 0;
         mutex_unlock(&c->umount_mutex);
         return err;
 }
@@ -1676,7 +1739,7 @@ static void ubifs_remount_ro(struct ubifs_info *c)
         int i, err;
 
         ubifs_assert(!c->need_recovery);
-        ubifs_assert(!(c->vfs_sb->s_flags & MS_RDONLY));
+        ubifs_assert(!c->ro_mount);
 
         mutex_lock(&c->umount_mutex);
         if (c->bgt) {
@@ -1686,10 +1749,8 @@ static void ubifs_remount_ro(struct ubifs_info *c)
 
         dbg_save_space_info(c);
 
-        for (i = 0; i < c->jhead_cnt; i++) {
+        for (i = 0; i < c->jhead_cnt; i++)
                 ubifs_wbuf_sync(&c->jheads[i].wbuf);
-                hrtimer_cancel(&c->jheads[i].wbuf.timer);
-        }
 
         c->mst_node->flags &= ~cpu_to_le32(UBIFS_MST_DIRTY);
         c->mst_node->flags |= cpu_to_le32(UBIFS_MST_NO_ORPHS);
@@ -1698,12 +1759,14 @@ static void ubifs_remount_ro(struct ubifs_info *c)
         if (err)
                 ubifs_ro_mode(c, err);
 
-        free_wbufs(c);
         vfree(c->orph_buf);
         c->orph_buf = NULL;
+        kfree(c->write_reserve_buf);
+        c->write_reserve_buf = NULL;
         vfree(c->ileb_buf);
         c->ileb_buf = NULL;
         ubifs_lpt_free(c, 1);
+        c->ro_mount = 1;
         err = dbg_check_space_info(c);
         if (err)
                 ubifs_ro_mode(c, err);
@@ -1723,10 +1786,11 @@ static void ubifs_put_super(struct super_block *sb)
          * of the media. For example, there will be dirty inodes if we failed
          * to write them back because of I/O errors.
          */
-        ubifs_assert(atomic_long_read(&c->dirty_pg_cnt) == 0);
-        ubifs_assert(c->budg_idx_growth == 0);
-        ubifs_assert(c->budg_dd_growth == 0);
-        ubifs_assert(c->budg_data_growth == 0);
+        if (!c->ro_error) {
+                ubifs_assert(c->bi.idx_growth == 0);
+                ubifs_assert(c->bi.dd_growth == 0);
+                ubifs_assert(c->bi.data_growth == 0);
+        }
 
         /*
          * The 'c->umount_lock' prevents races between UBIFS memory shrinker
@@ -1735,7 +1799,7 @@ static void ubifs_put_super(struct super_block *sb)
          * the mutex is locked.
          */
         mutex_lock(&c->umount_mutex);
-        if (!(c->vfs_sb->s_flags & MS_RDONLY)) {
+        if (!c->ro_mount) {
                 /*
                  * First of all kill the background thread to make sure it does
                  * not interfere with un-mounting and freeing resources.
@@ -1745,23 +1809,22 @@ static void ubifs_put_super(struct super_block *sb)
                         c->bgt = NULL;
                 }
 
-                /* Synchronize write-buffers */
-                if (c->jheads)
-                        for (i = 0; i < c->jhead_cnt; i++)
-                                ubifs_wbuf_sync(&c->jheads[i].wbuf);
-
                 /*
-                 * On fatal errors c->ro_media is set to 1, in which case we do
+                 * On fatal errors c->ro_error is set to 1, in which case we do
                  * not write the master node.
                  */
-                if (!c->ro_media) {
+                if (!c->ro_error) {
+                        int err;
+
+                        /* Synchronize write-buffers */
+                        for (i = 0; i < c->jhead_cnt; i++)
+                                ubifs_wbuf_sync(&c->jheads[i].wbuf);
+
                         /*
                          * We are being cleanly unmounted which means the
                          * orphans were killed - indicate this in the master
                          * node. Also save the reserved GC LEB number.
                          */
-                        int err;
-
                         c->mst_node->flags &= ~cpu_to_le32(UBIFS_MST_DIRTY);
                         c->mst_node->flags |= cpu_to_le32(UBIFS_MST_NO_ORPHS);
                         c->mst_node->gc_lnum = cpu_to_le32(c->gc_lnum);
@@ -1774,6 +1837,10 @@ static void ubifs_put_super(struct super_block *sb)
                                  */
                                 ubifs_err("failed to write master node, "
                                           "error %d", err);
+                } else {
+                        for (i = 0; i < c->jhead_cnt; i++)
+                                /* Make sure write-buffer timers are canceled */
+                                hrtimer_cancel(&c->jheads[i].wbuf.timer);
                 }
         }
 
@@ -1781,7 +1848,6 @@ static void ubifs_put_super(struct super_block *sb)
         bdi_destroy(&c->bdi);
         ubi_close_volume(c->ubi);
         mutex_unlock(&c->umount_mutex);
-        kfree(c);
 }
 
 static int ubifs_remount_fs(struct super_block *sb, int *flags, char *data)
@@ -1797,17 +1863,21 @@ static int ubifs_remount_fs(struct super_block *sb, int *flags, char *data)
                 return err;
         }
 
-        if ((sb->s_flags & MS_RDONLY) && !(*flags & MS_RDONLY)) {
+        if (c->ro_mount && !(*flags & MS_RDONLY)) {
+                if (c->ro_error) {
+                        ubifs_msg("cannot re-mount R/W due to prior errors");
+                        return -EROFS;
+                }
                 if (c->ro_media) {
-                        ubifs_msg("cannot re-mount due to prior errors");
+                        ubifs_msg("cannot re-mount R/W - UBI volume is R/O");
                         return -EROFS;
                 }
                 err = ubifs_remount_rw(c);
                 if (err)
                         return err;
-        } else if (!(sb->s_flags & MS_RDONLY) && (*flags & MS_RDONLY)) {
-                if (c->ro_media) {
-                        ubifs_msg("cannot re-mount due to prior errors");
+        } else if (!c->ro_mount && (*flags & MS_RDONLY)) {
+                if (c->ro_error) {
+                        ubifs_msg("cannot re-mount R/O due to prior errors");
                         return -EROFS;
                 }
                 ubifs_remount_ro(c);
@@ -1900,59 +1970,65 @@ static struct ubi_volume_desc *open_ubi(const char *name, int mode)
         return ERR_PTR(-EINVAL);
 }
 
-static int ubifs_fill_super(struct super_block *sb, void *data, int silent)
+static struct ubifs_info *alloc_ubifs_info(struct ubi_volume_desc *ubi)
 {
-        struct ubi_volume_desc *ubi = sb->s_fs_info;
         struct ubifs_info *c;
-        struct inode *root;
-        int err;
 
         c = kzalloc(sizeof(struct ubifs_info), GFP_KERNEL);
-        if (!c)
-                return -ENOMEM;
+        if (c) {
+                spin_lock_init(&c->cnt_lock);
+                spin_lock_init(&c->cs_lock);
+                spin_lock_init(&c->buds_lock);
+                spin_lock_init(&c->space_lock);
+                spin_lock_init(&c->orphan_lock);
+                init_rwsem(&c->commit_sem);
+                mutex_init(&c->lp_mutex);
+                mutex_init(&c->tnc_mutex);
+                mutex_init(&c->log_mutex);
+                mutex_init(&c->mst_mutex);
+                mutex_init(&c->umount_mutex);
+                mutex_init(&c->bu_mutex);
+                mutex_init(&c->write_reserve_mutex);
+                init_waitqueue_head(&c->cmt_wq);
+                c->buds = RB_ROOT;
+                c->old_idx = RB_ROOT;
+                c->size_tree = RB_ROOT;
+                c->orph_tree = RB_ROOT;
+                INIT_LIST_HEAD(&c->infos_list);
+                INIT_LIST_HEAD(&c->idx_gc);
+                INIT_LIST_HEAD(&c->replay_list);
+                INIT_LIST_HEAD(&c->replay_buds);
+                INIT_LIST_HEAD(&c->uncat_list);
+                INIT_LIST_HEAD(&c->empty_list);
+                INIT_LIST_HEAD(&c->freeable_list);
+                INIT_LIST_HEAD(&c->frdi_idx_list);
+                INIT_LIST_HEAD(&c->unclean_leb_list);
+                INIT_LIST_HEAD(&c->old_buds);
+                INIT_LIST_HEAD(&c->orph_list);
+                INIT_LIST_HEAD(&c->orph_new);
+                c->no_chk_data_crc = 1;
+
+                c->highest_inum = UBIFS_FIRST_INO;
+                c->lhead_lnum = c->ltail_lnum = UBIFS_LOG_LNUM;
+
+                ubi_get_volume_info(ubi, &c->vi);
+                ubi_get_device_info(c->vi.ubi_num, &c->di);
+        }
+        return c;
+}
 
-        spin_lock_init(&c->cnt_lock);
-        spin_lock_init(&c->cs_lock);
-        spin_lock_init(&c->buds_lock);
-        spin_lock_init(&c->space_lock);
-        spin_lock_init(&c->orphan_lock);
-        init_rwsem(&c->commit_sem);
-        mutex_init(&c->lp_mutex);
-        mutex_init(&c->tnc_mutex);
-        mutex_init(&c->log_mutex);
-        mutex_init(&c->mst_mutex);
-        mutex_init(&c->umount_mutex);
-        mutex_init(&c->bu_mutex);
-        init_waitqueue_head(&c->cmt_wq);
-        c->buds = RB_ROOT;
-        c->old_idx = RB_ROOT;
-        c->size_tree = RB_ROOT;
-        c->orph_tree = RB_ROOT;
-        INIT_LIST_HEAD(&c->infos_list);
-        INIT_LIST_HEAD(&c->idx_gc);
-        INIT_LIST_HEAD(&c->replay_list);
-        INIT_LIST_HEAD(&c->replay_buds);
-        INIT_LIST_HEAD(&c->uncat_list);
-        INIT_LIST_HEAD(&c->empty_list);
-        INIT_LIST_HEAD(&c->freeable_list);
-        INIT_LIST_HEAD(&c->frdi_idx_list);
-        INIT_LIST_HEAD(&c->unclean_leb_list);
-        INIT_LIST_HEAD(&c->old_buds);
-        INIT_LIST_HEAD(&c->orph_list);
-        INIT_LIST_HEAD(&c->orph_new);
+static int ubifs_fill_super(struct super_block *sb, void *data, int silent)
+{
+        struct ubifs_info *c = sb->s_fs_info;
+        struct inode *root;
+        int err;
 
         c->vfs_sb = sb;
-        c->highest_inum = UBIFS_FIRST_INO;
-        c->lhead_lnum = c->ltail_lnum = UBIFS_LOG_LNUM;
-
-        ubi_get_volume_info(ubi, &c->vi);
-        ubi_get_device_info(c->vi.ubi_num, &c->di);
-
         /* Re-open the UBI device in read-write mode */
         c->ubi = ubi_open_volume(c->vi.ubi_num, c->vi.vol_id, UBI_READWRITE);
         if (IS_ERR(c->ubi)) {
                 err = PTR_ERR(c->ubi);
-                goto out_free;
+                goto out;
         }
 
         /*
@@ -1965,7 +2041,6 @@ static int ubifs_fill_super(struct super_block *sb, void *data, int silent)
          */
         c->bdi.name = "ubifs",
         c->bdi.capabilities = BDI_CAP_MAP_COPY;
-        c->bdi.unplug_io_fn = default_unplug_io_fn;
         err  = bdi_init(&c->bdi);
         if (err)
                 goto out_close;
@@ -2019,24 +2094,29 @@ out_bdi:
         bdi_destroy(&c->bdi);
 out_close:
         ubi_close_volume(c->ubi);
-out_free:
-        kfree(c);
+out:
         return err;
 }
 
 static int sb_test(struct super_block *sb, void *data)
 {
-        dev_t *dev = data;
+        struct ubifs_info *c1 = data;
         struct ubifs_info *c = sb->s_fs_info;
 
-        return c->vi.cdev == *dev;
+        return c->vi.cdev == c1->vi.cdev;
+}
+
+static int sb_set(struct super_block *sb, void *data)
+{
+        sb->s_fs_info = data;
+        return set_anon_super(sb, NULL);
 }
 
-static int ubifs_get_sb(struct file_system_type *fs_type, int flags,
-                        const char *name, void *data, struct vfsmount *mnt)
+static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags,
+                        const char *name, void *data)
 {
         struct ubi_volume_desc *ubi;
-        struct ubi_volume_info vi;
+        struct ubifs_info *c;
         struct super_block *sb;
         int err;
 
@@ -2049,34 +2129,37 @@ static int ubifs_get_sb(struct file_system_type *fs_type, int flags,
          */
         ubi = open_ubi(name, UBI_READONLY);
         if (IS_ERR(ubi)) {
-                ubifs_err("cannot open \"%s\", error %d",
-                          name, (int)PTR_ERR(ubi));
-                return PTR_ERR(ubi);
+                dbg_err("cannot open \"%s\", error %d",
+                        name, (int)PTR_ERR(ubi));
+                return ERR_CAST(ubi);
         }
-        ubi_get_volume_info(ubi, &vi);
 
-        dbg_gen("opened ubi%d_%d", vi.ubi_num, vi.vol_id);
+        c = alloc_ubifs_info(ubi);
+        if (!c) {
+                err = -ENOMEM;
+                goto out_close;
+        }
 
-        sb = sget(fs_type, &sb_test, &set_anon_super, &vi.cdev);
+        dbg_gen("opened ubi%d_%d", c->vi.ubi_num, c->vi.vol_id);
+
+        sb = sget(fs_type, sb_test, sb_set, c);
         if (IS_ERR(sb)) {
                 err = PTR_ERR(sb);
+                kfree(c);
                 goto out_close;
         }
 
         if (sb->s_root) {
+                struct ubifs_info *c1 = sb->s_fs_info;
+                kfree(c);
                 /* A new mount point for already mounted UBIFS */
                 dbg_gen("this ubi volume is already mounted");
-                if ((flags ^ sb->s_flags) & MS_RDONLY) {
+                if (!!(flags & MS_RDONLY) != c1->ro_mount) {
                         err = -EBUSY;
                         goto out_deact;
                 }
         } else {
                 sb->s_flags = flags;
-                /*
-                 * Pass 'ubi' to 'fill_super()' in sb->s_fs_info where it is
-                 * replaced by 'c'.
-                 */
-                sb->s_fs_info = ubi;
                 err = ubifs_fill_super(sb, data, flags & MS_SILENT ? 1 : 0);
                 if (err)
                         goto out_deact;
@@ -2087,21 +2170,27 @@ static int ubifs_get_sb(struct file_system_type *fs_type, int flags,
         /* 'fill_super()' opens ubi again so we must close it here */
         ubi_close_volume(ubi);
 
-        simple_set_mnt(mnt, sb);
-        return 0;
+        return dget(sb->s_root);
 
 out_deact:
         deactivate_locked_super(sb);
 out_close:
         ubi_close_volume(ubi);
-        return err;
+        return ERR_PTR(err);
+}
+
+static void kill_ubifs_super(struct super_block *s)
+{
+        struct ubifs_info *c = s->s_fs_info;
+        kill_anon_super(s);
+        kfree(c);
 }
 
 static struct file_system_type ubifs_fs_type = {
         .name    = "ubifs",
         .owner   = THIS_MODULE,
-        .get_sb  = ubifs_get_sb,
-        .kill_sb = kill_anon_super,
+        .mount   = ubifs_mount,
+        .kill_sb = kill_ubifs_super,
 };
 
 /*
diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c
index 2194915220e5..91b4213dde84 100644
--- a/fs/ubifs/tnc.c
+++ b/fs/ubifs/tnc.c
@@ -447,8 +447,11 @@ static int tnc_read_node_nm(struct ubifs_info *c, struct ubifs_zbranch *zbr,
  *
  * Note, this function does not check CRC of data nodes if @c->no_chk_data_crc
  * is true (it is controlled by corresponding mount option). However, if
- * @c->always_chk_crc is true, @c->no_chk_data_crc is ignored and CRC is always
- * checked.
+ * @c->mounting or @c->remounting_rw is true (we are mounting or re-mounting to
+ * R/W mode), @c->no_chk_data_crc is ignored and CRC is checked. This is
+ * because during mounting or re-mounting from R/O mode to R/W mode we may read
+ * journal nodes (when replying the journal or doing the recovery) and the
+ * journal nodes may potentially be corrupted, so checking is required.
  */
 static int try_read_node(const struct ubifs_info *c, void *buf, int type,
                          int len, int lnum, int offs)
@@ -476,7 +479,8 @@ static int try_read_node(const struct ubifs_info *c, void *buf, int type,
         if (node_len != len)
                 return 0;
 
-        if (type == UBIFS_DATA_NODE && !c->always_chk_crc && c->no_chk_data_crc)
+        if (type == UBIFS_DATA_NODE && c->no_chk_data_crc && !c->mounting &&
+            !c->remounting_rw)
                 return 1;
 
         crc = crc32(UBIFS_CRC32_INIT, buf + 8, node_len - 8);
@@ -1177,6 +1181,7 @@ int ubifs_lookup_level0(struct ubifs_info *c, const union ubifs_key *key,
         unsigned long time = get_seconds();
 
         dbg_tnc("search key %s", DBGKEY(key));
+        ubifs_assert(key_type(c, key) < UBIFS_INVALID_KEY);
 
         znode = c->zroot.znode;
         if (unlikely(!znode)) {
@@ -2552,11 +2557,11 @@ int ubifs_tnc_remove_nm(struct ubifs_info *c, const union ubifs_key *key,
                 if (err) {
                         /* Ensure the znode is dirtied */
                         if (znode->cnext || !ubifs_zn_dirty(znode)) {
-                                    znode = dirty_cow_bottom_up(c, znode);
-                                    if (IS_ERR(znode)) {
-                                            err = PTR_ERR(znode);
-                                            goto out_unlock;
-                                    }
+                                znode = dirty_cow_bottom_up(c, znode);
+                                if (IS_ERR(znode)) {
+                                        err = PTR_ERR(znode);
+                                        goto out_unlock;
+                                }
                         }
                         err = tnc_delete(c, znode, n);
                 }
@@ -2871,12 +2876,13 @@ static void tnc_destroy_cnext(struct ubifs_info *c)
  */
 void ubifs_tnc_close(struct ubifs_info *c)
 {
-        long clean_freed;
-
         tnc_destroy_cnext(c);
         if (c->zroot.znode) {
-                clean_freed = ubifs_destroy_tnc_subtree(c->zroot.znode);
-                atomic_long_sub(clean_freed, &ubifs_clean_zn_cnt);
+                long n;
+
+                ubifs_destroy_tnc_subtree(c->zroot.znode);
+                n = atomic_long_read(&c->clean_zn_cnt);
+                atomic_long_sub(n, &ubifs_clean_zn_cnt);
         }
         kfree(c->gap_lebs);
         kfree(c->ilebs);
@@ -2966,7 +2972,7 @@ static struct ubifs_znode *right_znode(struct ubifs_info *c,
  *
  * This function searches an indexing node by its first key @key and its
  * address @lnum:@offs. It looks up the indexing tree by pulling all indexing
- * nodes it traverses to TNC. This function is called fro indexing nodes which
+ * nodes it traverses to TNC. This function is called for indexing nodes which
  * were found on the media by scanning, for example when garbage-collecting or
  * when doing in-the-gaps commit. This means that the indexing node which is
  * looked for does not have to have exactly the same leftmost key @key, because
@@ -2988,6 +2994,8 @@ static struct ubifs_znode *lookup_znode(struct ubifs_info *c,
         struct ubifs_znode *znode, *zn;
         int n, nn;
 
+        ubifs_assert(key_type(c, key) < UBIFS_INVALID_KEY);
+
         /*
          * The arguments have probably been read off flash, so don't assume
          * they are valid.
diff --git a/fs/ubifs/tnc_commit.c b/fs/ubifs/tnc_commit.c
index 53288e5d604e..41920f357bbf 100644
--- a/fs/ubifs/tnc_commit.c
+++ b/fs/ubifs/tnc_commit.c
@@ -377,15 +377,13 @@ static int layout_in_gaps(struct ubifs_info *c, int cnt)
                                 c->gap_lebs = NULL;
                                 return err;
                         }
-                        if (!dbg_force_in_the_gaps_enabled) {
+                        if (dbg_force_in_the_gaps_enabled()) {
                                 /*
                                  * Do not print scary warnings if the debugging
                                  * option which forces in-the-gaps is enabled.
                                  */
-                                ubifs_err("out of space");
-                                spin_lock(&c->space_lock);
-                                dbg_dump_budg(c);
-                                spin_unlock(&c->space_lock);
+                                ubifs_warn("out of space");
+                                dbg_dump_budg(c, &c->bi);
                                 dbg_dump_lprops(c);
                         }
                         /* Try to commit anyway */
@@ -796,16 +794,16 @@ int ubifs_tnc_start_commit(struct ubifs_info *c, struct ubifs_zbranch *zroot)
         spin_lock(&c->space_lock);
         /*
          * Although we have not finished committing yet, update size of the
-         * committed index ('c->old_idx_sz') and zero out the index growth
+         * committed index ('c->bi.old_idx_sz') and zero out the index growth
          * budget. It is OK to do this now, because we've reserved all the
          * space which is needed to commit the index, and it is save for the
          * budgeting subsystem to assume the index is already committed,
          * even though it is not.
          */
-        ubifs_assert(c->min_idx_lebs == ubifs_calc_min_idx_lebs(c));
-        c->old_idx_sz = c->calc_idx_sz;
-        c->budg_uncommitted_idx = 0;
-        c->min_idx_lebs = ubifs_calc_min_idx_lebs(c);
+        ubifs_assert(c->bi.min_idx_lebs == ubifs_calc_min_idx_lebs(c));
+        c->bi.old_idx_sz = c->calc_idx_sz;
+        c->bi.uncommitted_idx = 0;
+        c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c);
         spin_unlock(&c->space_lock);
         mutex_unlock(&c->tnc_mutex);
 
diff --git a/fs/ubifs/ubifs-media.h b/fs/ubifs/ubifs-media.h
index 191ca7863fe7..e24380cf46ed 100644
--- a/fs/ubifs/ubifs-media.h
+++ b/fs/ubifs/ubifs-media.h
@@ -408,9 +408,11 @@ enum {
  * Superblock flags.
  *
  * UBIFS_FLG_BIGLPT: if "big" LPT model is used if set
+ * UBIFS_FLG_SPACE_FIXUP: first-mount "fixup" of free space within LEBs needed
  */
 enum {
         UBIFS_FLG_BIGLPT = 0x02,
+        UBIFS_FLG_SPACE_FIXUP = 0x04,
 };
 
 /**
@@ -434,7 +436,7 @@ struct ubifs_ch {
         __u8 node_type;
         __u8 group_type;
         __u8 padding[2];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * union ubifs_dev_desc - device node descriptor.
@@ -448,7 +450,7 @@ struct ubifs_ch {
 union ubifs_dev_desc {
         __le32 new;
         __le64 huge;
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_ino_node - inode node.
@@ -509,7 +511,7 @@ struct ubifs_ino_node {
         __le16 compr_type;
         __u8 padding2[26]; /* Watch 'zero_ino_node_unused()' if changing! */
         __u8 data[];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_dent_node - directory entry node.
@@ -534,7 +536,7 @@ struct ubifs_dent_node {
         __le16 nlen;
         __u8 padding2[4]; /* Watch 'zero_dent_node_unused()' if changing! */
         __u8 name[];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_data_node - data node.
@@ -555,7 +557,7 @@ struct ubifs_data_node {
         __le16 compr_type;
         __u8 padding[2]; /* Watch 'zero_data_node_unused()' if changing! */
         __u8 data[];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_trun_node - truncation node.
@@ -575,7 +577,7 @@ struct ubifs_trun_node {
         __u8 padding[12]; /* Watch 'zero_trun_node_unused()' if changing! */
         __le64 old_size;
         __le64 new_size;
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_pad_node - padding node.
@@ -586,7 +588,7 @@ struct ubifs_trun_node {
 struct ubifs_pad_node {
         struct ubifs_ch ch;
         __le32 pad_len;
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_sb_node - superblock node.
@@ -644,7 +646,7 @@ struct ubifs_sb_node {
         __u8 uuid[16];
         __le32 ro_compat_version;
         __u8 padding2[3968];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_mst_node - master node.
@@ -711,7 +713,7 @@ struct ubifs_mst_node {
         __le32 idx_lebs;
         __le32 leb_cnt;
         __u8 padding[344];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_ref_node - logical eraseblock reference node.
@@ -727,7 +729,7 @@ struct ubifs_ref_node {
         __le32 offs;
         __le32 jhead;
         __u8 padding[28];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_branch - key/reference/length branch
@@ -741,7 +743,7 @@ struct ubifs_branch {
         __le32 offs;
         __le32 len;
         __u8 key[];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_idx_node - indexing node.
@@ -755,7 +757,7 @@ struct ubifs_idx_node {
         __le16 child_cnt;
         __le16 level;
         __u8 branches[];
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_cs_node - commit start node.
@@ -765,7 +767,7 @@ struct ubifs_idx_node {
 struct ubifs_cs_node {
         struct ubifs_ch ch;
         __le64 cmt_no;
-} __attribute__ ((packed));
+} __packed;
 
 /**
  * struct ubifs_orph_node - orphan node.
@@ -777,6 +779,6 @@ struct ubifs_orph_node {
         struct ubifs_ch ch;
         __le64 cmt_no;
         __le64 inos[];
-} __attribute__ ((packed));
+} __packed;
 
 #endif /* __UBIFS_MEDIA_H__ */
diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h
index 0c9876b396dd..f79983d6f860 100644
--- a/fs/ubifs/ubifs.h
+++ b/fs/ubifs/ubifs.h
@@ -119,8 +119,12 @@
  * in TNC. However, when replaying, it is handy to introduce fake "truncation"
  * keys for truncation nodes because the code becomes simpler. So we define
  * %UBIFS_TRUN_KEY type.
+ *
+ * But otherwise, out of the journal reply scope, the truncation keys are
+ * invalid.
  */
-#define UBIFS_TRUN_KEY UBIFS_KEY_TYPES_CNT
+#define UBIFS_TRUN_KEY    UBIFS_KEY_TYPES_CNT
+#define UBIFS_INVALID_KEY UBIFS_KEY_TYPES_CNT
 
 /*
  * How much a directory entry/extended attribute entry adds to the parent/host
@@ -147,6 +151,12 @@
  */
 #define WORST_COMPR_FACTOR 2
 
+/*
+ * How much memory is needed for a buffer where we comress a data node.
+ */
+#define COMPRESSED_DATA_NODE_BUF_SZ \
+        (UBIFS_DATA_NODE_SZ + UBIFS_BLOCK_SIZE * WORST_COMPR_FACTOR)
+
 /* Maximum expected tree height for use by bottom_up_buf */
 #define BOTTOM_UP_HEIGHT 64
 
@@ -379,9 +389,9 @@ struct ubifs_gced_idx_leb {
  * The @ui_size is a "shadow" variable for @inode->i_size and UBIFS uses
  * @ui_size instead of @inode->i_size. The reason for this is that UBIFS cannot
  * make sure @inode->i_size is always changed under @ui_mutex, because it
- * cannot call 'truncate_setsize()' with @ui_mutex locked, because it would deadlock
- * with 'ubifs_writepage()' (see file.c). All the other inode fields are
- * changed under @ui_mutex, so they do not need "shadow" fields. Note, one
+ * cannot call 'truncate_setsize()' with @ui_mutex locked, because it would
+ * deadlock with 'ubifs_writepage()' (see file.c). All the other inode fields
+ * are changed under @ui_mutex, so they do not need "shadow" fields. Note, one
  * could consider to rework locking and base it on "shadow" fields.
  */
 struct ubifs_inode {
@@ -642,6 +652,7 @@ typedef int (*ubifs_lpt_scan_callback)(struct ubifs_info *c,
  * @offs: write-buffer offset in this logical eraseblock
  * @avail: number of bytes available in the write-buffer
  * @used:  number of used bytes in the write-buffer
+ * @size: write-buffer size (in [@c->min_io_size, @c->max_write_size] range)
  * @dtype: type of data stored in this LEB (%UBI_LONGTERM, %UBI_SHORTTERM,
  * %UBI_UNKNOWN)
  * @jhead: journal head the mutex belongs to (note, needed only to shut lockdep
@@ -676,6 +687,7 @@ struct ubifs_wbuf {
         int offs;
         int avail;
         int used;
+        int size;
         int dtype;
         int jhead;
         int (*sync_callback)(struct ubifs_info *c, int lnum, int free, int pad);
@@ -710,12 +722,14 @@ struct ubifs_bud {
  * struct ubifs_jhead - journal head.
  * @wbuf: head's write-buffer
  * @buds_list: list of bud LEBs belonging to this journal head
+ * @grouped: non-zero if UBIFS groups nodes when writing to this journal head
  *
  * Note, the @buds list is protected by the @c->buds_lock.
  */
 struct ubifs_jhead {
         struct ubifs_wbuf wbuf;
         struct list_head buds_list;
+        unsigned int grouped:1;
 };
 
 /**
@@ -925,6 +939,40 @@ struct ubifs_mount_opts {
         unsigned int compr_type:2;
 };
 
+/**
+ * struct ubifs_budg_info - UBIFS budgeting information.
+ * @idx_growth: amount of bytes budgeted for index growth
+ * @data_growth: amount of bytes budgeted for cached data
+ * @dd_growth: amount of bytes budgeted for cached data that will make
+ *             other data dirty
+ * @uncommitted_idx: amount of bytes were budgeted for growth of the index, but
+ *                   which still have to be taken into account because the index
+ *                   has not been committed so far
+ * @old_idx_sz: size of index on flash
+ * @min_idx_lebs: minimum number of LEBs required for the index
+ * @nospace: non-zero if the file-system does not have flash space (used as
+ *           optimization)
+ * @nospace_rp: the same as @nospace, but additionally means that even reserved
+ *              pool is full
+ * @page_budget: budget for a page (constant, nenver changed after mount)
+ * @inode_budget: budget for an inode (constant, nenver changed after mount)
+ * @dent_budget: budget for a directory entry (constant, nenver changed after
+ *               mount)
+ */
+struct ubifs_budg_info {
+        long long idx_growth;
+        long long data_growth;
+        long long dd_growth;
+        long long uncommitted_idx;
+        unsigned long long old_idx_sz;
+        int min_idx_lebs;
+        unsigned int nospace:1;
+        unsigned int nospace_rp:1;
+        int page_budget;
+        int inode_budget;
+        int dent_budget;
+};
+
 struct ubifs_debug_info;
 
 /**
@@ -968,6 +1016,7 @@ struct ubifs_debug_info;
  * @cmt_wq: wait queue to sleep on if the log is full and a commit is running
  *
  * @big_lpt: flag that LPT is too big to write whole during commit
+ * @space_fixup: flag indicating that free space in LEBs needs to be cleaned up
  * @no_chk_data_crc: do not check CRCs when reading data nodes (except during
  *                   recovery)
  * @bulk_read: enable bulk-reads
@@ -999,6 +1048,11 @@ struct ubifs_debug_info;
  * @bu_mutex: protects the pre-allocated bulk-read buffer and @c->bu
  * @bu: pre-allocated bulk-read information
  *
+ * @write_reserve_mutex: protects @write_reserve_buf
+ * @write_reserve_buf: on the write path we allocate memory, which might
+ *                     sometimes be unavailable, in which case we use this
+ *                     write reserve buffer
+ *
  * @log_lebs: number of logical eraseblocks in the log
  * @log_bytes: log size in bytes
  * @log_last: last LEB of the log
@@ -1020,7 +1074,12 @@ struct ubifs_debug_info;
  *
  * @min_io_size: minimal input/output unit size
  * @min_io_shift: number of bits in @min_io_size minus one
+ * @max_write_size: maximum amount of bytes the underlying flash can write at a
+ *                  time (MTD write buffer size)
+ * @max_write_shift: number of bits in @max_write_size minus one
  * @leb_size: logical eraseblock size in bytes
+ * @leb_start: starting offset of logical eraseblocks within physical
+ *             eraseblocks
  * @half_leb_size: half LEB size
  * @idx_leb_size: how many bytes of an LEB are effectively available when it is
  *                used to store indexing nodes (@leb_size - @max_idx_node_sz)
@@ -1028,37 +1087,21 @@ struct ubifs_debug_info;
  * @max_leb_cnt: maximum count of logical eraseblocks
  * @old_leb_cnt: count of logical eraseblocks before re-size
  * @ro_media: the underlying UBI volume is read-only
+ * @ro_mount: the file-system was mounted as read-only
+ * @ro_error: UBIFS switched to R/O mode because an error happened
  *
  * @dirty_pg_cnt: number of dirty pages (not used)
  * @dirty_zn_cnt: number of dirty znodes
  * @clean_zn_cnt: number of clean znodes
  *
- * @budg_idx_growth: amount of bytes budgeted for index growth
- * @budg_data_growth: amount of bytes budgeted for cached data
- * @budg_dd_growth: amount of bytes budgeted for cached data that will make
- *                  other data dirty
- * @budg_uncommitted_idx: amount of bytes were budgeted for growth of the index,
- *                        but which still have to be taken into account because
- *                        the index has not been committed so far
- * @space_lock: protects @budg_idx_growth, @budg_data_growth, @budg_dd_growth,
- *              @budg_uncommited_idx, @min_idx_lebs, @old_idx_sz, @lst,
- *              @nospace, and @nospace_rp;
- * @min_idx_lebs: minimum number of LEBs required for the index
- * @old_idx_sz: size of index on flash
+ * @space_lock: protects @bi and @lst
+ * @lst: lprops statistics
+ * @bi: budgeting information
  * @calc_idx_sz: temporary variable which is used to calculate new index size
  *               (contains accurate new index size at end of TNC commit start)
- * @lst: lprops statistics
- * @nospace: non-zero if the file-system does not have flash space (used as
- *           optimization)
- * @nospace_rp: the same as @nospace, but additionally means that even reserved
- *              pool is full
- *
- * @page_budget: budget for a page
- * @inode_budget: budget for an inode
- * @dent_budget: budget for a directory entry
  *
  * @ref_node_alsz: size of the LEB reference node aligned to the min. flash
- * I/O unit
+ *                 I/O unit
  * @mst_node_alsz: master node aligned size
  * @min_idx_node_sz: minimum indexing node aligned on 8-bytes boundary
  * @max_idx_node_sz: maximum indexing node aligned on 8-bytes boundary
@@ -1160,19 +1203,20 @@ struct ubifs_debug_info;
  * @rp_uid: reserved pool user ID
  * @rp_gid: reserved pool group ID
  *
- * @empty: if the UBI device is empty
- * @replay_tree: temporary tree used during journal replay
+ * @empty: %1 if the UBI device is empty
+ * @need_recovery: %1 if the file-system needs recovery
+ * @replaying: %1 during journal replay
+ * @mounting: %1 while mounting
+ * @remounting_rw: %1 while re-mounting from R/O mode to R/W mode
  * @replay_list: temporary list used during journal replay
  * @replay_buds: list of buds to replay
  * @cs_sqnum: sequence number of first node in the log (commit start node)
  * @replay_sqnum: sequence number of node currently being replayed
- * @need_recovery: file-system needs recovery
- * @replaying: set to %1 during journal replay
- * @unclean_leb_list: LEBs to recover when mounting ro to rw
- * @rcvrd_mst_node: recovered master node to write when mounting ro to rw
+ * @unclean_leb_list: LEBs to recover when re-mounting R/O mounted FS to R/W
+ *                    mode
+ * @rcvrd_mst_node: recovered master node to write when re-mounting R/O mounted
+ *                  FS to R/W mode
  * @size_tree: inode size information for recovery
- * @remounting_rw: set while remounting from ro to rw (sb flags have MS_RDONLY)
- * @always_chk_crc: always check CRCs (while mounting and remounting rw)
  * @mount_opts: UBIFS-specific mount options
  *
  * @dbg: debugging-related information
@@ -1212,6 +1256,7 @@ struct ubifs_info {
         wait_queue_head_t cmt_wq;
 
         unsigned int big_lpt:1;
+        unsigned int space_fixup:1;
         unsigned int no_chk_data_crc:1;
         unsigned int bulk_read:1;
         unsigned int default_compr:2;
@@ -1241,6 +1286,9 @@ struct ubifs_info {
         struct mutex bu_mutex;
         struct bu_info bu;
 
+        struct mutex write_reserve_mutex;
+        void *write_reserve_buf;
+
         int log_lebs;
         long long log_bytes;
         int log_last;
@@ -1262,33 +1310,27 @@ struct ubifs_info {
 
         int min_io_size;
         int min_io_shift;
+        int max_write_size;
+        int max_write_shift;
         int leb_size;
+        int leb_start;
         int half_leb_size;
         int idx_leb_size;
         int leb_cnt;
         int max_leb_cnt;
         int old_leb_cnt;
-        int ro_media;
+        unsigned int ro_media:1;
+        unsigned int ro_mount:1;
+        unsigned int ro_error:1;
 
         atomic_long_t dirty_pg_cnt;
         atomic_long_t dirty_zn_cnt;
         atomic_long_t clean_zn_cnt;
 
-        long long budg_idx_growth;
-        long long budg_data_growth;
-        long long budg_dd_growth;
-        long long budg_uncommitted_idx;
         spinlock_t space_lock;
-        int min_idx_lebs;
-        unsigned long long old_idx_sz;
-        unsigned long long calc_idx_sz;
         struct ubifs_lp_stats lst;
-        unsigned int nospace:1;
-        unsigned int nospace_rp:1;
-
-        int page_budget;
-        int inode_budget;
-        int dent_budget;
+        struct ubifs_budg_info bi;
+        unsigned long long calc_idx_sz;
 
         int ref_node_alsz;
         int mst_node_alsz;
@@ -1391,19 +1433,18 @@ struct ubifs_info {
         gid_t rp_gid;
 
         /* The below fields are used only during mounting and re-mounting */
-        int empty;
-        struct rb_root replay_tree;
+        unsigned int empty:1;
+        unsigned int need_recovery:1;
+        unsigned int replaying:1;
+        unsigned int mounting:1;
+        unsigned int remounting_rw:1;
         struct list_head replay_list;
         struct list_head replay_buds;
         unsigned long long cs_sqnum;
         unsigned long long replay_sqnum;
-        int need_recovery;
-        int replaying;
         struct list_head unclean_leb_list;
         struct ubifs_mst_node *rcvrd_mst_node;
         struct rb_root size_tree;
-        int remounting_rw;
-        int always_chk_crc;
         struct ubifs_mount_opts mount_opts;
 
 #ifdef CONFIG_UBIFS_FS_DEBUG
@@ -1575,7 +1616,7 @@ int ubifs_tnc_start_commit(struct ubifs_info *c, struct ubifs_zbranch *zroot);
 int ubifs_tnc_end_commit(struct ubifs_info *c);
 
 /* shrinker.c */
-int ubifs_shrinker(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask);
+int ubifs_shrinker(struct shrinker *shrink, struct shrink_control *sc);
 
 /* commit.c */
 int ubifs_bg_thread(void *info);
@@ -1594,6 +1635,7 @@ int ubifs_write_master(struct ubifs_info *c);
 int ubifs_read_superblock(struct ubifs_info *c);
 struct ubifs_sb_node *ubifs_read_sb_node(struct ubifs_info *c);
 int ubifs_write_sb_node(struct ubifs_info *c, struct ubifs_sb_node *sup);
+int ubifs_fixup_free_space(struct ubifs_info *c);
 
 /* replay.c */
 int ubifs_validate_entry(struct ubifs_info *c,
@@ -1702,7 +1744,7 @@ struct inode *ubifs_iget(struct super_block *sb, unsigned long inum);
 int ubifs_recover_master_node(struct ubifs_info *c);
 int ubifs_write_rcvrd_mst_node(struct ubifs_info *c);
 struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum,
-                                         int offs, void *sbuf, int grouped);
+                                         int offs, void *sbuf, int jhead);
 struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum,
                                              int offs, void *sbuf);
 int ubifs_recover_inl_heads(const struct ubifs_info *c, void *sbuf);
diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c
index c74400f88fe0..16f19f55e63f 100644
--- a/fs/ubifs/xattr.c
+++ b/fs/ubifs/xattr.c
@@ -56,6 +56,7 @@
  */
 
 #include "ubifs.h"
+#include <linux/fs.h>
 #include <linux/slab.h>
 #include <linux/xattr.h>
 #include <linux/posix_acl_xattr.h>
@@ -79,9 +80,8 @@ enum {
         SECURITY_XATTR,
 };
 
-static const struct inode_operations none_inode_operations;
-static const struct address_space_operations none_address_operations;
-static const struct file_operations none_file_operations;
+static const struct inode_operations empty_iops;
+static const struct file_operations empty_fops;
 
 /**
  * create_xattr - create an extended attribute.
@@ -130,9 +130,9 @@ static int create_xattr(struct ubifs_info *c, struct inode *host,
         }
 
         /* Re-define all operations to be "nothing" */
-        inode->i_mapping->a_ops = &none_address_operations;
-        inode->i_op = &none_inode_operations;
-        inode->i_fop = &none_file_operations;
+        inode->i_mapping->a_ops = &empty_aops;
+        inode->i_op = &empty_iops;
+        inode->i_fop = &empty_fops;
 
         inode->i_flags |= S_SYNC | S_NOATIME | S_NOCMTIME | S_NOQUOTA;
         ui = ubifs_inode(inode);