aboutsummaryrefslogtreecommitdiff
path: root/fs/ubifs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ubifs')
-rw-r--r--fs/ubifs/Kconfig35
-rw-r--r--fs/ubifs/Makefile5
-rw-r--r--fs/ubifs/budget.c116
-rw-r--r--fs/ubifs/commit.c90
-rw-r--r--fs/ubifs/compress.c7
-rw-r--r--fs/ubifs/debug.c1813
-rw-r--r--fs/ubifs/debug.h521
-rw-r--r--fs/ubifs/dir.c201
-rw-r--r--fs/ubifs/file.c127
-rw-r--r--fs/ubifs/find.c26
-rw-r--r--fs/ubifs/gc.c82
-rw-r--r--fs/ubifs/io.c418
-rw-r--r--fs/ubifs/ioctl.c8
-rw-r--r--fs/ubifs/journal.c83
-rw-r--r--fs/ubifs/log.c105
-rw-r--r--fs/ubifs/lprops.c209
-rw-r--r--fs/ubifs/lpt.c112
-rw-r--r--fs/ubifs/lpt_commit.c275
-rw-r--r--fs/ubifs/master.c16
-rw-r--r--fs/ubifs/misc.h120
-rw-r--r--fs/ubifs/orphan.c84
-rw-r--r--fs/ubifs/recovery.c561
-rw-r--r--fs/ubifs/replay.c514
-rw-r--r--fs/ubifs/sb.c217
-rw-r--r--fs/ubifs/scan.c33
-rw-r--r--fs/ubifs/shrinker.c25
-rw-r--r--fs/ubifs/super.c535
-rw-r--r--fs/ubifs/tnc.c164
-rw-r--r--fs/ubifs/tnc_commit.c187
-rw-r--r--fs/ubifs/tnc_misc.c46
-rw-r--r--fs/ubifs/ubifs-media.h30
-rw-r--r--fs/ubifs/ubifs.h231
-rw-r--r--fs/ubifs/xattr.c48
33 files changed, 3772 insertions, 3272 deletions
diff --git a/fs/ubifs/Kconfig b/fs/ubifs/Kconfig
index 830e3f76f44..ba66d508006 100644
--- a/fs/ubifs/Kconfig
+++ b/fs/ubifs/Kconfig
@@ -11,12 +11,6 @@ config UBIFS_FS
help
UBIFS is a file system for flash devices which works on top of UBI.
-config UBIFS_FS_XATTR
- bool "Extended attributes support"
- depends on UBIFS_FS
- help
- This option enables support of extended attributes.
-
config UBIFS_FS_ADVANCED_COMPR
bool "Advanced compression options"
depends on UBIFS_FS
@@ -41,32 +35,3 @@ config UBIFS_FS_ZLIB
default y
help
Zlib compresses better than LZO but it is slower. Say 'Y' if unsure.
-
-# Debugging-related stuff
-config UBIFS_FS_DEBUG
- bool "Enable debugging"
- 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.
diff --git a/fs/ubifs/Makefile b/fs/ubifs/Makefile
index 80e93c35e49..2c6f0cb816b 100644
--- a/fs/ubifs/Makefile
+++ b/fs/ubifs/Makefile
@@ -3,7 +3,4 @@ obj-$(CONFIG_UBIFS_FS) += ubifs.o
ubifs-y += shrinker.o journal.o file.o dir.o super.o sb.o io.o
ubifs-y += tnc.o master.o scan.o replay.o log.o commit.o gc.o orphan.o
ubifs-y += budget.o find.o tnc_commit.o compress.o lpt.o lprops.o
-ubifs-y += recovery.o ioctl.o lpt_commit.o tnc_misc.o
-
-ubifs-$(CONFIG_UBIFS_FS_DEBUG) += debug.o
-ubifs-$(CONFIG_UBIFS_FS_XATTR) += xattr.o
+ubifs-y += recovery.o ioctl.o lpt_commit.o tnc_misc.o xattr.o debug.o
diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c
index c8ff0d1ae5d..eb997e9c4ab 100644
--- a/fs/ubifs/budget.c
+++ b/fs/ubifs/budget.c
@@ -63,7 +63,7 @@
static void shrink_liability(struct ubifs_info *c, int nr_to_write)
{
down_read(&c->vfs_sb->s_umount);
- writeback_inodes_sb(c->vfs_sb);
+ writeback_inodes_sb(c->vfs_sb, WB_REASON_FS_FREE_SPACE);
up_read(&c->vfs_sb->s_umount);
}
@@ -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;
/*
@@ -272,8 +272,8 @@ long long ubifs_calc_available(const struct ubifs_info *c, int min_idx_lebs)
*/
static int can_use_rp(struct ubifs_info *c)
{
- if (current_fsuid() == c->rp_uid || capable(CAP_SYS_RESOURCE) ||
- (c->rp_gid != 0 && in_group_p(c->rp_gid)))
+ if (uid_eq(current_fsuid(), c->rp_uid) || capable(CAP_SYS_RESOURCE) ||
+ (!gid_eq(c->rp_gid, GLOBAL_ROOT_GID) && in_group_p(c->rp_gid)))
return 1;
return 0;
}
@@ -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.
@@ -342,14 +342,13 @@ static int do_budget_space(struct ubifs_info *c)
lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt -
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);
+ dbg_budg("out of indexing space: min_idx_lebs %d (old %d), 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 +359,7 @@ 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 +392,11 @@ 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 +412,12 @@ 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;
}
@@ -438,7 +437,6 @@ static int calc_dd_growth(const struct ubifs_info *c,
*/
int ubifs_budget_space(struct ubifs_info *c, struct ubifs_budget_req *req)
{
- int uninitialized_var(cmt_retries), uninitialized_var(wb_retries);
int err, idx_growth, data_growth, dd_growth, retried = 0;
ubifs_assert(req->new_page <= 1);
@@ -460,19 +458,19 @@ 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 +482,9 @@ 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 +504,9 @@ 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 +521,8 @@ 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 +551,23 @@ 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 +584,13 @@ 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 +610,7 @@ 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 +680,9 @@ 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 +695,8 @@ 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 02429d81ca3..ff8229340cd 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 && ubifs_zn_dirty(c->zroot.znode))
+ 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
@@ -70,6 +120,12 @@ static int do_commit(struct ubifs_info *c)
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);
@@ -126,7 +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);
@@ -162,12 +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:
@@ -237,8 +293,8 @@ int ubifs_bg_thread(void *info)
int err;
struct ubifs_info *c = info;
- dbg_msg("background thread \"%s\" started, PID %d",
- c->bgt_name, current->pid);
+ ubifs_msg("background thread \"%s\" started, PID %d",
+ c->bgt_name, current->pid);
set_freezable();
while (1) {
@@ -272,7 +328,7 @@ int ubifs_bg_thread(void *info)
cond_resched();
}
- dbg_msg("background thread \"%s\" stops", c->bgt_name);
+ ubifs_msg("background thread \"%s\" stops", c->bgt_name);
return 0;
}
@@ -362,7 +418,7 @@ int ubifs_run_commit(struct ubifs_info *c)
spin_lock(&c->cs_lock);
if (c->cmt_state == COMMIT_BROKEN) {
- err = -EINVAL;
+ err = -EROFS;
goto out;
}
@@ -388,7 +444,7 @@ int ubifs_run_commit(struct ubifs_info *c)
* re-check it.
*/
if (c->cmt_state == COMMIT_BROKEN) {
- err = -EINVAL;
+ err = -EROFS;
goto out_cmt_unlock;
}
@@ -440,7 +496,9 @@ int ubifs_gc_should_commit(struct ubifs_info *c)
return ret;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
+/*
+ * Everything below is related to debugging.
+ */
/**
* struct idx_node - hold index nodes during index tree traversal.
@@ -456,7 +514,7 @@ struct idx_node {
struct list_head list;
int iip;
union ubifs_key upper_key;
- struct ubifs_idx_node idx __attribute__((aligned(8)));
+ struct ubifs_idx_node idx __aligned(8);
};
/**
@@ -520,8 +578,8 @@ int dbg_check_old_index(struct ubifs_info *c, struct ubifs_zbranch *zroot)
struct idx_node *i;
size_t sz;
- if (!(ubifs_chk_flags & UBIFS_CHK_OLD_IDX))
- goto out;
+ if (!dbg_is_chk_index(c))
+ return 0;
INIT_LIST_HEAD(&list);
@@ -658,14 +716,14 @@ out:
return 0;
out_dump:
- dbg_err("dumping index node (iip=%d)", i->iip);
- dbg_dump_node(c, idx);
+ ubifs_err("dumping index node (iip=%d)", i->iip);
+ ubifs_dump_node(c, idx);
list_del(&i->list);
kfree(i);
if (!list_empty(&list)) {
i = list_entry(list.prev, struct idx_node, list);
- dbg_err("dumping parent index node");
- dbg_dump_node(c, &i->idx);
+ ubifs_err("dumping parent index node");
+ ubifs_dump_node(c, &i->idx);
}
out_free:
while (!list_empty(&list)) {
@@ -678,5 +736,3 @@ out_free:
err = -EINVAL;
return err;
}
-
-#endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/compress.c b/fs/ubifs/compress.c
index 11e4132f314..2bfa0953335 100644
--- a/fs/ubifs/compress.c
+++ b/fs/ubifs/compress.c
@@ -112,8 +112,7 @@ void ubifs_compress(const void *in_buf, int in_len, void *out_buf, int *out_len,
if (compr->comp_mutex)
mutex_unlock(compr->comp_mutex);
if (unlikely(err)) {
- ubifs_warn("cannot compress %d bytes, compressor %s, "
- "error %d, leave data uncompressed",
+ ubifs_warn("cannot compress %d bytes, compressor %s, error %d, leave data uncompressed",
in_len, compr->name, err);
goto no_compr;
}
@@ -176,8 +175,8 @@ int ubifs_decompress(const void *in_buf, int in_len, void *out_buf,
if (compr->decomp_mutex)
mutex_unlock(compr->decomp_mutex);
if (err)
- ubifs_err("cannot decompress %d bytes, compressor %s, "
- "error %d", in_len, compr->name, err);
+ ubifs_err("cannot decompress %d bytes, compressor %s, error %d",
+ in_len, compr->name, err);
return err;
}
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index 0bee4dbffc3..177b0152fef 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -27,33 +27,14 @@
* various local functions of those subsystems.
*/
-#define UBIFS_DBG_PRESERVE_UBI
-
-#include "ubifs.h"
#include <linux/module.h>
-#include <linux/moduleparam.h>
#include <linux/debugfs.h>
#include <linux/math64.h>
-#include <linux/slab.h>
-
-#ifdef CONFIG_UBIFS_FS_DEBUG
-
-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_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);
+#include <linux/uaccess.h>
+#include <linux/random.h>
+#include "ubifs.h"
-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");
+static DEFINE_SPINLOCK(dbg_lock);
static const char *get_key_fmt(int fmt)
{
@@ -95,8 +76,30 @@ static const char *get_key_type(int type)
}
}
-static void sprintf_key(const struct ubifs_info *c, const union ubifs_key *key,
- char *buffer)
+static const char *get_dent_type(int type)
+{
+ switch (type) {
+ case UBIFS_ITYPE_REG:
+ return "file";
+ case UBIFS_ITYPE_DIR:
+ return "dir";
+ case UBIFS_ITYPE_LNK:
+ return "symlink";
+ case UBIFS_ITYPE_BLK:
+ return "blkdev";
+ case UBIFS_ITYPE_CHR:
+ return "char dev";
+ case UBIFS_ITYPE_FIFO:
+ return "fifo";
+ case UBIFS_ITYPE_SOCK:
+ return "socket";
+ default:
+ return "unknown/invalid type";
+ }
+}
+
+const char *dbg_snprintf_key(const struct ubifs_info *c,
+ const union ubifs_key *key, char *buffer, int len)
{
char *p = buffer;
int type = key_type(c, key);
@@ -104,45 +107,34 @@ static void sprintf_key(const struct ubifs_info *c, const union ubifs_key *key,
if (c->key_fmt == UBIFS_SIMPLE_KEY_FMT) {
switch (type) {
case UBIFS_INO_KEY:
- sprintf(p, "(%lu, %s)", (unsigned long)key_inum(c, key),
- get_key_type(type));
+ len -= snprintf(p, len, "(%lu, %s)",
+ (unsigned long)key_inum(c, key),
+ get_key_type(type));
break;
case UBIFS_DENT_KEY:
case UBIFS_XENT_KEY:
- sprintf(p, "(%lu, %s, %#08x)",
- (unsigned long)key_inum(c, key),
- get_key_type(type), key_hash(c, key));
+ len -= snprintf(p, len, "(%lu, %s, %#08x)",
+ (unsigned long)key_inum(c, key),
+ get_key_type(type), key_hash(c, key));
break;
case UBIFS_DATA_KEY:
- sprintf(p, "(%lu, %s, %u)",
- (unsigned long)key_inum(c, key),
- get_key_type(type), key_block(c, key));
+ len -= snprintf(p, len, "(%lu, %s, %u)",
+ (unsigned long)key_inum(c, key),
+ get_key_type(type), key_block(c, key));
break;
case UBIFS_TRUN_KEY:
- sprintf(p, "(%lu, %s)",
- (unsigned long)key_inum(c, key),
- get_key_type(type));
+ len -= snprintf(p, len, "(%lu, %s)",
+ (unsigned long)key_inum(c, key),
+ get_key_type(type));
break;
default:
- sprintf(p, "(bad key type: %#08x, %#08x)",
- key->u32[0], key->u32[1]);
+ len -= snprintf(p, len, "(bad key type: %#08x, %#08x)",
+ key->u32[0], key->u32[1]);
}
} else
- sprintf(p, "bad key format %d", c->key_fmt);
-}
-
-const char *dbg_key_str0(const struct ubifs_info *c, const union ubifs_key *key)
-{
- /* dbg_lock must be held */
- sprintf_key(c, key, dbg_key_buf0);
- return dbg_key_buf0;
-}
-
-const char *dbg_key_str1(const struct ubifs_info *c, const union ubifs_key *key)
-{
- /* dbg_lock must be held */
- sprintf_key(c, key, dbg_key_buf1);
- return dbg_key_buf1;
+ len -= snprintf(p, len, "bad key format %d", c->key_fmt);
+ ubifs_assert(len > 0);
+ return p;
}
const char *dbg_ntype(int type)
@@ -227,68 +219,96 @@ const char *dbg_jhead(int jhead)
static void dump_ch(const struct ubifs_ch *ch)
{
- printk(KERN_DEBUG "\tmagic %#x\n", le32_to_cpu(ch->magic));
- printk(KERN_DEBUG "\tcrc %#x\n", le32_to_cpu(ch->crc));
- printk(KERN_DEBUG "\tnode_type %d (%s)\n", ch->node_type,
+ pr_err("\tmagic %#x\n", le32_to_cpu(ch->magic));
+ pr_err("\tcrc %#x\n", le32_to_cpu(ch->crc));
+ pr_err("\tnode_type %d (%s)\n", ch->node_type,
dbg_ntype(ch->node_type));
- printk(KERN_DEBUG "\tgroup_type %d (%s)\n", ch->group_type,
+ pr_err("\tgroup_type %d (%s)\n", ch->group_type,
dbg_gtype(ch->group_type));
- printk(KERN_DEBUG "\tsqnum %llu\n",
+ pr_err("\tsqnum %llu\n",
(unsigned long long)le64_to_cpu(ch->sqnum));
- printk(KERN_DEBUG "\tlen %u\n", le32_to_cpu(ch->len));
+ pr_err("\tlen %u\n", le32_to_cpu(ch->len));
}
-void dbg_dump_inode(const struct ubifs_info *c, const struct inode *inode)
+void ubifs_dump_inode(struct ubifs_info *c, const struct inode *inode)
{
const struct ubifs_inode *ui = ubifs_inode(inode);
+ struct qstr nm = { .name = NULL };
+ union ubifs_key key;
+ struct ubifs_dent_node *dent, *pdent = NULL;
+ int count = 2;
- printk(KERN_DEBUG "Dump in-memory inode:");
- printk(KERN_DEBUG "\tinode %lu\n", inode->i_ino);
- printk(KERN_DEBUG "\tsize %llu\n",
+ pr_err("Dump in-memory inode:");
+ pr_err("\tinode %lu\n", inode->i_ino);
+ pr_err("\tsize %llu\n",
(unsigned long long)i_size_read(inode));
- printk(KERN_DEBUG "\tnlink %u\n", inode->i_nlink);
- printk(KERN_DEBUG "\tuid %u\n", (unsigned int)inode->i_uid);
- printk(KERN_DEBUG "\tgid %u\n", (unsigned int)inode->i_gid);
- printk(KERN_DEBUG "\tatime %u.%u\n",
+ pr_err("\tnlink %u\n", inode->i_nlink);
+ pr_err("\tuid %u\n", (unsigned int)i_uid_read(inode));
+ pr_err("\tgid %u\n", (unsigned int)i_gid_read(inode));
+ pr_err("\tatime %u.%u\n",
(unsigned int)inode->i_atime.tv_sec,
(unsigned int)inode->i_atime.tv_nsec);
- printk(KERN_DEBUG "\tmtime %u.%u\n",
+ pr_err("\tmtime %u.%u\n",
(unsigned int)inode->i_mtime.tv_sec,
(unsigned int)inode->i_mtime.tv_nsec);
- printk(KERN_DEBUG "\tctime %u.%u\n",
+ pr_err("\tctime %u.%u\n",
(unsigned int)inode->i_ctime.tv_sec,
(unsigned int)inode->i_ctime.tv_nsec);
- printk(KERN_DEBUG "\tcreat_sqnum %llu\n", ui->creat_sqnum);
- printk(KERN_DEBUG "\txattr_size %u\n", ui->xattr_size);
- printk(KERN_DEBUG "\txattr_cnt %u\n", ui->xattr_cnt);
- printk(KERN_DEBUG "\txattr_names %u\n", ui->xattr_names);
- printk(KERN_DEBUG "\tdirty %u\n", ui->dirty);
- printk(KERN_DEBUG "\txattr %u\n", ui->xattr);
- printk(KERN_DEBUG "\tbulk_read %u\n", ui->xattr);
- printk(KERN_DEBUG "\tsynced_i_size %llu\n",
+ pr_err("\tcreat_sqnum %llu\n", ui->creat_sqnum);
+ pr_err("\txattr_size %u\n", ui->xattr_size);
+ pr_err("\txattr_cnt %u\n", ui->xattr_cnt);
+ pr_err("\txattr_names %u\n", ui->xattr_names);
+ pr_err("\tdirty %u\n", ui->dirty);
+ pr_err("\txattr %u\n", ui->xattr);
+ pr_err("\tbulk_read %u\n", ui->xattr);
+ pr_err("\tsynced_i_size %llu\n",
(unsigned long long)ui->synced_i_size);
- printk(KERN_DEBUG "\tui_size %llu\n",
+ pr_err("\tui_size %llu\n",
(unsigned long long)ui->ui_size);
- printk(KERN_DEBUG "\tflags %d\n", ui->flags);
- printk(KERN_DEBUG "\tcompr_type %d\n", ui->compr_type);
- printk(KERN_DEBUG "\tlast_page_read %lu\n", ui->last_page_read);
- printk(KERN_DEBUG "\tread_in_a_row %lu\n", ui->read_in_a_row);
- printk(KERN_DEBUG "\tdata_len %d\n", ui->data_len);
+ pr_err("\tflags %d\n", ui->flags);
+ pr_err("\tcompr_type %d\n", ui->compr_type);
+ pr_err("\tlast_page_read %lu\n", ui->last_page_read);
+ pr_err("\tread_in_a_row %lu\n", ui->read_in_a_row);
+ pr_err("\tdata_len %d\n", ui->data_len);
+
+ if (!S_ISDIR(inode->i_mode))
+ return;
+
+ pr_err("List of directory entries:\n");
+ ubifs_assert(!mutex_is_locked(&c->tnc_mutex));
+
+ lowest_dent_key(c, &key, inode->i_ino);
+ while (1) {
+ dent = ubifs_tnc_next_ent(c, &key, &nm);
+ if (IS_ERR(dent)) {
+ if (PTR_ERR(dent) != -ENOENT)
+ pr_err("error %ld\n", PTR_ERR(dent));
+ break;
+ }
+
+ pr_err("\t%d: %s (%s)\n",
+ count++, dent->name, get_dent_type(dent->type));
+
+ nm.name = dent->name;
+ nm.len = le16_to_cpu(dent->nlen);
+ kfree(pdent);
+ pdent = dent;
+ key_read(c, &dent->key, &key);
+ }
+ kfree(pdent);
}
-void dbg_dump_node(const struct ubifs_info *c, const void *node)
+void ubifs_dump_node(const struct ubifs_info *c, const void *node)
{
int i, n;
union ubifs_key key;
const struct ubifs_ch *ch = node;
-
- if (dbg_failure_mode)
- return;
+ char key_buf[DBG_KEY_BUF_LEN];
/* If the magic is incorrect, just hexdump the first bytes */
if (le32_to_cpu(ch->magic) != UBIFS_NODE_MAGIC) {
- printk(KERN_DEBUG "Not a node, first %zu bytes:", UBIFS_CH_SZ);
- print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1,
+ pr_err("Not a node, first %zu bytes:", UBIFS_CH_SZ);
+ print_hex_dump(KERN_ERR, "", DUMP_PREFIX_OFFSET, 32, 1,
(void *)node, UBIFS_CH_SZ, 1);
return;
}
@@ -301,8 +321,7 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
{
const struct ubifs_pad_node *pad = node;
- printk(KERN_DEBUG "\tpad_len %u\n",
- le32_to_cpu(pad->pad_len));
+ pr_err("\tpad_len %u\n", le32_to_cpu(pad->pad_len));
break;
}
case UBIFS_SB_NODE:
@@ -310,110 +329,77 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
const struct ubifs_sb_node *sup = node;
unsigned int sup_flags = le32_to_cpu(sup->flags);
- printk(KERN_DEBUG "\tkey_hash %d (%s)\n",
+ pr_err("\tkey_hash %d (%s)\n",
(int)sup->key_hash, get_key_hash(sup->key_hash));
- printk(KERN_DEBUG "\tkey_fmt %d (%s)\n",
+ pr_err("\tkey_fmt %d (%s)\n",
(int)sup->key_fmt, get_key_fmt(sup->key_fmt));
- printk(KERN_DEBUG "\tflags %#x\n", sup_flags);
- printk(KERN_DEBUG "\t big_lpt %u\n",
+ pr_err("\tflags %#x\n", sup_flags);
+ pr_err("\t big_lpt %u\n",
!!(sup_flags & UBIFS_FLG_BIGLPT));
- printk(KERN_DEBUG "\tmin_io_size %u\n",
- le32_to_cpu(sup->min_io_size));
- printk(KERN_DEBUG "\tleb_size %u\n",
- le32_to_cpu(sup->leb_size));
- printk(KERN_DEBUG "\tleb_cnt %u\n",
- le32_to_cpu(sup->leb_cnt));
- printk(KERN_DEBUG "\tmax_leb_cnt %u\n",
- le32_to_cpu(sup->max_leb_cnt));
- printk(KERN_DEBUG "\tmax_bud_bytes %llu\n",
+ pr_err("\t space_fixup %u\n",
+ !!(sup_flags & UBIFS_FLG_SPACE_FIXUP));
+ pr_err("\tmin_io_size %u\n", le32_to_cpu(sup->min_io_size));
+ pr_err("\tleb_size %u\n", le32_to_cpu(sup->leb_size));
+ pr_err("\tleb_cnt %u\n", le32_to_cpu(sup->leb_cnt));
+ pr_err("\tmax_leb_cnt %u\n", le32_to_cpu(sup->max_leb_cnt));
+ pr_err("\tmax_bud_bytes %llu\n",
(unsigned long long)le64_to_cpu(sup->max_bud_bytes));
- printk(KERN_DEBUG "\tlog_lebs %u\n",
- le32_to_cpu(sup->log_lebs));
- printk(KERN_DEBUG "\tlpt_lebs %u\n",
- le32_to_cpu(sup->lpt_lebs));
- printk(KERN_DEBUG "\torph_lebs %u\n",
- le32_to_cpu(sup->orph_lebs));
- printk(KERN_DEBUG "\tjhead_cnt %u\n",
- le32_to_cpu(sup->jhead_cnt));
- printk(KERN_DEBUG "\tfanout %u\n",
- le32_to_cpu(sup->fanout));
- printk(KERN_DEBUG "\tlsave_cnt %u\n",
- le32_to_cpu(sup->lsave_cnt));
- printk(KERN_DEBUG "\tdefault_compr %u\n",
+ pr_err("\tlog_lebs %u\n", le32_to_cpu(sup->log_lebs));
+ pr_err("\tlpt_lebs %u\n", le32_to_cpu(sup->lpt_lebs));
+ pr_err("\torph_lebs %u\n", le32_to_cpu(sup->orph_lebs));
+ pr_err("\tjhead_cnt %u\n", le32_to_cpu(sup->jhead_cnt));
+ pr_err("\tfanout %u\n", le32_to_cpu(sup->fanout));
+ pr_err("\tlsave_cnt %u\n", le32_to_cpu(sup->lsave_cnt));
+ pr_err("\tdefault_compr %u\n",
(int)le16_to_cpu(sup->default_compr));
- printk(KERN_DEBUG "\trp_size %llu\n",
+ pr_err("\trp_size %llu\n",
(unsigned long long)le64_to_cpu(sup->rp_size));
- printk(KERN_DEBUG "\trp_uid %u\n",
- le32_to_cpu(sup->rp_uid));
- printk(KERN_DEBUG "\trp_gid %u\n",
- le32_to_cpu(sup->rp_gid));
- printk(KERN_DEBUG "\tfmt_version %u\n",
- le32_to_cpu(sup->fmt_version));
- printk(KERN_DEBUG "\ttime_gran %u\n",
- le32_to_cpu(sup->time_gran));
- printk(KERN_DEBUG "\tUUID %pUB\n",
- sup->uuid);
+ pr_err("\trp_uid %u\n", le32_to_cpu(sup->rp_uid));
+ pr_err("\trp_gid %u\n", le32_to_cpu(sup->rp_gid));
+ pr_err("\tfmt_version %u\n", le32_to_cpu(sup->fmt_version));
+ pr_err("\ttime_gran %u\n", le32_to_cpu(sup->time_gran));
+ pr_err("\tUUID %pUB\n", sup->uuid);
break;
}
case UBIFS_MST_NODE:
{
const struct ubifs_mst_node *mst = node;
- printk(KERN_DEBUG "\thighest_inum %llu\n",
+ pr_err("\thighest_inum %llu\n",
(unsigned long long)le64_to_cpu(mst->highest_inum));
- printk(KERN_DEBUG "\tcommit number %llu\n",
+ pr_err("\tcommit number %llu\n",
(unsigned long long)le64_to_cpu(mst->cmt_no));
- printk(KERN_DEBUG "\tflags %#x\n",
- le32_to_cpu(mst->flags));
- printk(KERN_DEBUG "\tlog_lnum %u\n",
- le32_to_cpu(mst->log_lnum));
- printk(KERN_DEBUG "\troot_lnum %u\n",
- le32_to_cpu(mst->root_lnum));
- printk(KERN_DEBUG "\troot_offs %u\n",
- le32_to_cpu(mst->root_offs));
- printk(KERN_DEBUG "\troot_len %u\n",
- le32_to_cpu(mst->root_len));
- printk(KERN_DEBUG "\tgc_lnum %u\n",
- le32_to_cpu(mst->gc_lnum));
- printk(KERN_DEBUG "\tihead_lnum %u\n",
- le32_to_cpu(mst->ihead_lnum));
- printk(KERN_DEBUG "\tihead_offs %u\n",
- le32_to_cpu(mst->ihead_offs));
- printk(KERN_DEBUG "\tindex_size %llu\n",
+ pr_err("\tflags %#x\n", le32_to_cpu(mst->flags));
+ pr_err("\tlog_lnum %u\n", le32_to_cpu(mst->log_lnum));
+ pr_err("\troot_lnum %u\n", le32_to_cpu(mst->root_lnum));
+ pr_err("\troot_offs %u\n", le32_to_cpu(mst->root_offs));
+ pr_err("\troot_len %u\n", le32_to_cpu(mst->root_len));
+ pr_err("\tgc_lnum %u\n", le32_to_cpu(mst->gc_lnum));
+ pr_err("\tihead_lnum %u\n", le32_to_cpu(mst->ihead_lnum));
+ pr_err("\tihead_offs %u\n", le32_to_cpu(mst->ihead_offs));
+ pr_err("\tindex_size %llu\n",
(unsigned long long)le64_to_cpu(mst->index_size));
- printk(KERN_DEBUG "\tlpt_lnum %u\n",
- le32_to_cpu(mst->lpt_lnum));
- printk(KERN_DEBUG "\tlpt_offs %u\n",
- le32_to_cpu(mst->lpt_offs));
- printk(KERN_DEBUG "\tnhead_lnum %u\n",
- le32_to_cpu(mst->nhead_lnum));
- printk(KERN_DEBUG "\tnhead_offs %u\n",
- le32_to_cpu(mst->nhead_offs));
- printk(KERN_DEBUG "\tltab_lnum %u\n",
- le32_to_cpu(mst->ltab_lnum));
- printk(KERN_DEBUG "\tltab_offs %u\n",
- le32_to_cpu(mst->ltab_offs));
- printk(KERN_DEBUG "\tlsave_lnum %u\n",
- le32_to_cpu(mst->lsave_lnum));
- printk(KERN_DEBUG "\tlsave_offs %u\n",
- le32_to_cpu(mst->lsave_offs));
- printk(KERN_DEBUG "\tlscan_lnum %u\n",
- le32_to_cpu(mst->lscan_lnum));
- printk(KERN_DEBUG "\tleb_cnt %u\n",
- le32_to_cpu(mst->leb_cnt));
- printk(KERN_DEBUG "\tempty_lebs %u\n",
- le32_to_cpu(mst->empty_lebs));
- printk(KERN_DEBUG "\tidx_lebs %u\n",
- le32_to_cpu(mst->idx_lebs));
- printk(KERN_DEBUG "\ttotal_free %llu\n",
+ pr_err("\tlpt_lnum %u\n", le32_to_cpu(mst->lpt_lnum));
+ pr_err("\tlpt_offs %u\n", le32_to_cpu(mst->lpt_offs));
+ pr_err("\tnhead_lnum %u\n", le32_to_cpu(mst->nhead_lnum));
+ pr_err("\tnhead_offs %u\n", le32_to_cpu(mst->nhead_offs));
+ pr_err("\tltab_lnum %u\n", le32_to_cpu(mst->ltab_lnum));
+ pr_err("\tltab_offs %u\n", le32_to_cpu(mst->ltab_offs));
+ pr_err("\tlsave_lnum %u\n", le32_to_cpu(mst->lsave_lnum));
+ pr_err("\tlsave_offs %u\n", le32_to_cpu(mst->lsave_offs));
+ pr_err("\tlscan_lnum %u\n", le32_to_cpu(mst->lscan_lnum));
+ pr_err("\tleb_cnt %u\n", le32_to_cpu(mst->leb_cnt));
+ pr_err("\tempty_lebs %u\n", le32_to_cpu(mst->empty_lebs));
+ pr_err("\tidx_lebs %u\n", le32_to_cpu(mst->idx_lebs));
+ pr_err("\ttotal_free %llu\n",
(unsigned long long)le64_to_cpu(mst->total_free));
- printk(KERN_DEBUG "\ttotal_dirty %llu\n",
+ pr_err("\ttotal_dirty %llu\n",
(unsigned long long)le64_to_cpu(mst->total_dirty));
- printk(KERN_DEBUG "\ttotal_used %llu\n",
+ pr_err("\ttotal_used %llu\n",
(unsigned long long)le64_to_cpu(mst->total_used));
- printk(KERN_DEBUG "\ttotal_dead %llu\n",
+ pr_err("\ttotal_dead %llu\n",
(unsigned long long)le64_to_cpu(mst->total_dead));
- printk(KERN_DEBUG "\ttotal_dark %llu\n",
+ pr_err("\ttotal_dark %llu\n",
(unsigned long long)le64_to_cpu(mst->total_dark));
break;
}
@@ -421,12 +407,9 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
{
const struct ubifs_ref_node *ref = node;
- printk(KERN_DEBUG "\tlnum %u\n",
- le32_to_cpu(ref->lnum));
- printk(KERN_DEBUG "\toffs %u\n",
- le32_to_cpu(ref->offs));
- printk(KERN_DEBUG "\tjhead %u\n",
- le32_to_cpu(ref->jhead));
+ pr_err("\tlnum %u\n", le32_to_cpu(ref->lnum));
+ pr_err("\toffs %u\n", le32_to_cpu(ref->offs));
+ pr_err("\tjhead %u\n", le32_to_cpu(ref->jhead));
break;
}
case UBIFS_INO_NODE:
@@ -434,40 +417,32 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
const struct ubifs_ino_node *ino = node;
key_read(c, &ino->key, &key);
- printk(KERN_DEBUG "\tkey %s\n", DBGKEY(&key));
- printk(KERN_DEBUG "\tcreat_sqnum %llu\n",
+ pr_err("\tkey %s\n",
+ dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN));
+ pr_err("\tcreat_sqnum %llu\n",
(unsigned long long)le64_to_cpu(ino->creat_sqnum));
- printk(KERN_DEBUG "\tsize %llu\n",
+ pr_err("\tsize %llu\n",
(unsigned long long)le64_to_cpu(ino->size));
- printk(KERN_DEBUG "\tnlink %u\n",
- le32_to_cpu(ino->nlink));
- printk(KERN_DEBUG "\tatime %lld.%u\n",
+ pr_err("\tnlink %u\n", le32_to_cpu(ino->nlink));
+ pr_err("\tatime %lld.%u\n",
(long long)le64_to_cpu(ino->atime_sec),
le32_to_cpu(ino->atime_nsec));
- printk(KERN_DEBUG "\tmtime %lld.%u\n",
+ pr_err("\tmtime %lld.%u\n",
(long long)le64_to_cpu(ino->mtime_sec),
le32_to_cpu(ino->mtime_nsec));
- printk(KERN_DEBUG "\tctime %lld.%u\n",
+ pr_err("\tctime %lld.%u\n",
(long long)le64_to_cpu(ino->ctime_sec),
le32_to_cpu(ino->ctime_nsec));
- printk(KERN_DEBUG "\tuid %u\n",
- le32_to_cpu(ino->uid));
- printk(KERN_DEBUG "\tgid %u\n",
- le32_to_cpu(ino->gid));
- printk(KERN_DEBUG "\tmode %u\n",
- le32_to_cpu(ino->mode));
- printk(KERN_DEBUG "\tflags %#x\n",
- le32_to_cpu(ino->flags));
- printk(KERN_DEBUG "\txattr_cnt %u\n",
- le32_to_cpu(ino->xattr_cnt));
- printk(KERN_DEBUG "\txattr_size %u\n",
- le32_to_cpu(ino->xattr_size));
- printk(KERN_DEBUG "\txattr_names %u\n",
- le32_to_cpu(ino->xattr_names));
- printk(KERN_DEBUG "\tcompr_type %#x\n",
+ pr_err("\tuid %u\n", le32_to_cpu(ino->uid));
+ pr_err("\tgid %u\n", le32_to_cpu(ino->gid));
+ pr_err("\tmode %u\n", le32_to_cpu(ino->mode));
+ pr_err("\tflags %#x\n", le32_to_cpu(ino->flags));
+ pr_err("\txattr_cnt %u\n", le32_to_cpu(ino->xattr_cnt));
+ pr_err("\txattr_size %u\n", le32_to_cpu(ino->xattr_size));
+ pr_err("\txattr_names %u\n", le32_to_cpu(ino->xattr_names));
+ pr_err("\tcompr_type %#x\n",
(int)le16_to_cpu(ino->compr_type));
- printk(KERN_DEBUG "\tdata len %u\n",
- le32_to_cpu(ino->data_len));
+ pr_err("\tdata len %u\n", le32_to_cpu(ino->data_len));
break;
}
case UBIFS_DENT_NODE:
@@ -477,21 +452,21 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
int nlen = le16_to_cpu(dent->nlen);
key_read(c, &dent->key, &key);
- printk(KERN_DEBUG "\tkey %s\n", DBGKEY(&key));
- printk(KERN_DEBUG "\tinum %llu\n",
+ pr_err("\tkey %s\n",
+ dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN));
+ pr_err("\tinum %llu\n",
(unsigned long long)le64_to_cpu(dent->inum));
- printk(KERN_DEBUG "\ttype %d\n", (int)dent->type);
- printk(KERN_DEBUG "\tnlen %d\n", nlen);
- printk(KERN_DEBUG "\tname ");
+ pr_err("\ttype %d\n", (int)dent->type);
+ pr_err("\tnlen %d\n", nlen);
+ pr_err("\tname ");
if (nlen > UBIFS_MAX_NLEN)
- printk(KERN_DEBUG "(bad name length, not printing, "
- "bad or corrupted node)");
+ pr_err("(bad name length, not printing, bad or corrupted node)");
else {
for (i = 0; i < nlen && dent->name[i]; i++)
- printk(KERN_CONT "%c", dent->name[i]);
+ pr_cont("%c", dent->name[i]);
}
- printk(KERN_CONT "\n");
+ pr_cont("\n");
break;
}
@@ -501,15 +476,14 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
int dlen = le32_to_cpu(ch->len) - UBIFS_DATA_NODE_SZ;
key_read(c, &dn->key, &key);
- printk(KERN_DEBUG "\tkey %s\n", DBGKEY(&key));
- printk(KERN_DEBUG "\tsize %u\n",
- le32_to_cpu(dn->size));
- printk(KERN_DEBUG "\tcompr_typ %d\n",
+ pr_err("\tkey %s\n",
+ dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN));
+ pr_err("\tsize %u\n", le32_to_cpu(dn->size));
+ pr_err("\tcompr_typ %d\n",
(int)le16_to_cpu(dn->compr_type));
- printk(KERN_DEBUG "\tdata size %d\n",
- dlen);
- printk(KERN_DEBUG "\tdata:\n");
- print_hex_dump(KERN_DEBUG, "\t", DUMP_PREFIX_OFFSET, 32, 1,
+ pr_err("\tdata size %d\n", dlen);
+ pr_err("\tdata:\n");
+ print_hex_dump(KERN_ERR, "\t", DUMP_PREFIX_OFFSET, 32, 1,
(void *)&dn->data, dlen, 0);
break;
}
@@ -517,11 +491,10 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
{
const struct ubifs_trun_node *trun = node;
- printk(KERN_DEBUG "\tinum %u\n",
- le32_to_cpu(trun->inum));
- printk(KERN_DEBUG "\told_size %llu\n",
+ pr_err("\tinum %u\n", le32_to_cpu(trun->inum));
+ pr_err("\told_size %llu\n",
(unsigned long long)le64_to_cpu(trun->old_size));
- printk(KERN_DEBUG "\tnew_size %llu\n",
+ pr_err("\tnew_size %llu\n",
(unsigned long long)le64_to_cpu(trun->new_size));
break;
}
@@ -530,19 +503,20 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
const struct ubifs_idx_node *idx = node;
n = le16_to_cpu(idx->child_cnt);
- printk(KERN_DEBUG "\tchild_cnt %d\n", n);
- printk(KERN_DEBUG "\tlevel %d\n",
- (int)le16_to_cpu(idx->level));
- printk(KERN_DEBUG "\tBranches:\n");
+ pr_err("\tchild_cnt %d\n", n);
+ pr_err("\tlevel %d\n", (int)le16_to_cpu(idx->level));
+ pr_err("\tBranches:\n");
for (i = 0; i < n && i < c->fanout - 1; i++) {
const struct ubifs_branch *br;
br = ubifs_idx_branch(c, idx, i);
key_read(c, &br->key, &key);
- printk(KERN_DEBUG "\t%d: LEB %d:%d len %d key %s\n",
+ pr_err("\t%d: LEB %d:%d len %d key %s\n",
i, le32_to_cpu(br->lnum), le32_to_cpu(br->offs),
- le32_to_cpu(br->len), DBGKEY(&key));
+ le32_to_cpu(br->len),
+ dbg_snprintf_key(c, &key, key_buf,
+ DBG_KEY_BUF_LEN));
}
break;
}
@@ -552,57 +526,55 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
{
const struct ubifs_orph_node *orph = node;
- printk(KERN_DEBUG "\tcommit number %llu\n",
+ pr_err("\tcommit number %llu\n",
(unsigned long long)
le64_to_cpu(orph->cmt_no) & LLONG_MAX);
- printk(KERN_DEBUG "\tlast node flag %llu\n",
+ pr_err("\tlast node flag %llu\n",
(unsigned long long)(le64_to_cpu(orph->cmt_no)) >> 63);
n = (le32_to_cpu(ch->len) - UBIFS_ORPH_NODE_SZ) >> 3;
- printk(KERN_DEBUG "\t%d orphan inode numbers:\n", n);
+ pr_err("\t%d orphan inode numbers:\n", n);
for (i = 0; i < n; i++)
- printk(KERN_DEBUG "\t ino %llu\n",
+ pr_err("\t ino %llu\n",
(unsigned long long)le64_to_cpu(orph->inos[i]));
break;
}
default:
- printk(KERN_DEBUG "node type %d was not recognized\n",
+ pr_err("node type %d was not recognized\n",
(int)ch->node_type);
}
spin_unlock(&dbg_lock);
}
-void dbg_dump_budget_req(const struct ubifs_budget_req *req)
+void ubifs_dump_budget_req(const struct ubifs_budget_req *req)
{
spin_lock(&dbg_lock);
- printk(KERN_DEBUG "Budgeting request: new_ino %d, dirtied_ino %d\n",
+ pr_err("Budgeting request: new_ino %d, dirtied_ino %d\n",
req->new_ino, req->dirtied_ino);
- printk(KERN_DEBUG "\tnew_ino_d %d, dirtied_ino_d %d\n",
+ pr_err("\tnew_ino_d %d, dirtied_ino_d %d\n",
req->new_ino_d, req->dirtied_ino_d);
- printk(KERN_DEBUG "\tnew_page %d, dirtied_page %d\n",
+ pr_err("\tnew_page %d, dirtied_page %d\n",
req->new_page, req->dirtied_page);
- printk(KERN_DEBUG "\tnew_dent %d, mod_dent %d\n",
+ pr_err("\tnew_dent %d, mod_dent %d\n",
req->new_dent, req->mod_dent);
- printk(KERN_DEBUG "\tidx_growth %d\n", req->idx_growth);
- printk(KERN_DEBUG "\tdata_growth %d dd_growth %d\n",
+ pr_err("\tidx_growth %d\n", req->idx_growth);
+ pr_err("\tdata_growth %d dd_growth %d\n",
req->data_growth, req->dd_growth);
spin_unlock(&dbg_lock);
}
-void dbg_dump_lstats(const struct ubifs_lp_stats *lst)
+void ubifs_dump_lstats(const struct ubifs_lp_stats *lst)
{
spin_lock(&dbg_lock);
- printk(KERN_DEBUG "(pid %d) Lprops statistics: empty_lebs %d, "
- "idx_lebs %d\n", current->pid, lst->empty_lebs, lst->idx_lebs);
- printk(KERN_DEBUG "\ttaken_empty_lebs %d, total_free %lld, "
- "total_dirty %lld\n", lst->taken_empty_lebs, lst->total_free,
- lst->total_dirty);
- printk(KERN_DEBUG "\ttotal_used %lld, total_dark %lld, "
- "total_dead %lld\n", lst->total_used, lst->total_dark,
- lst->total_dead);
+ pr_err("(pid %d) Lprops statistics: empty_lebs %d, idx_lebs %d\n",
+ current->pid, lst->empty_lebs, lst->idx_lebs);
+ pr_err("\ttaken_empty_lebs %d, total_free %lld, total_dirty %lld\n",
+ lst->taken_empty_lebs, lst->total_free, lst->total_dirty);
+ pr_err("\ttotal_used %lld, total_dark %lld, total_dead %lld\n",
+ lst->total_used, lst->total_dark, lst->total_dead);
spin_unlock(&dbg_lock);
}
-void dbg_dump_budg(struct ubifs_info *c)
+void ubifs_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi)
{
int i;
struct rb_node *rb;
@@ -610,54 +582,67 @@ 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 "\tdirty_pg_cnt %ld, dirty_zn_cnt %ld, "
- "clean_zn_cnt %ld\n", atomic_long_read(&c->dirty_pg_cnt),
+ pr_err("(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);
+ pr_err("\tbudg_data_growth %lld, budg_dd_growth %lld, budg_idx_growth %lld\n",
+ bi->data_growth, bi->dd_growth, bi->idx_growth);
+ pr_err("\tmin_idx_lebs %d, old_idx_sz %llu, uncommitted_idx %lld\n",
+ bi->min_idx_lebs, bi->old_idx_sz, bi->uncommitted_idx);
+ pr_err("\tpage_budget %d, inode_budget %d, dent_budget %d\n",
+ bi->page_budget, bi->inode_budget, bi->dent_budget);
+ pr_err("\tnospace %u, nospace_rp %u\n", bi->nospace, bi->nospace_rp);
+ pr_err("\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;
+
+ pr_err("\tfreeable_cnt %d, calc_idx_sz %lld, idx_gc_cnt %d\n",
+ c->freeable_cnt, c->calc_idx_sz, c->idx_gc_cnt);
+ pr_err("\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);
+ pr_err("\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++)
- printk(KERN_DEBUG "\tjhead %s\t LEB %d\n",
+ pr_err("\tjhead %s\t LEB %d\n",
dbg_jhead(c->jheads[i].wbuf.jhead),
c->jheads[i].wbuf.lnum);
for (rb = rb_first(&c->buds); rb; rb = rb_next(rb)) {
bud = rb_entry(rb, struct ubifs_bud, rb);
- printk(KERN_DEBUG "\tbud LEB %d\n", bud->lnum);
+ pr_err("\tbud LEB %d\n", bud->lnum);
}
list_for_each_entry(bud, &c->old_buds, list)
- printk(KERN_DEBUG "\told bud LEB %d\n", bud->lnum);
+ pr_err("\told bud LEB %d\n", bud->lnum);
list_for_each_entry(idx_gc, &c->idx_gc, list)
- printk(KERN_DEBUG "\tGC'ed idx LEB %d unmap %d\n",
+ pr_err("\tGC'ed idx LEB %d unmap %d\n",
idx_gc->lnum, idx_gc->unmap);
- printk(KERN_DEBUG "\tcommit state %d\n", c->cmt_state);
+ pr_err("\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",
+ pr_err("Budgeting predictions:\n");
+ pr_err("\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)
+void ubifs_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp)
{
int i, spc, dark = 0, dead = 0;
struct rb_node *rb;
@@ -670,21 +655,19 @@ void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp)
dark = ubifs_calc_dark(c, spc);
if (lp->flags & LPROPS_INDEX)
- printk(KERN_DEBUG "LEB %-7d free %-8d dirty %-8d used %-8d "
- "free + dirty %-8d flags %#x (", lp->lnum, lp->free,
- lp->dirty, c->leb_size - spc, spc, lp->flags);
+ pr_err("LEB %-7d free %-8d dirty %-8d used %-8d free + dirty %-8d flags %#x (",
+ lp->lnum, lp->free, lp->dirty, c->leb_size - spc, spc,
+ lp->flags);
else
- printk(KERN_DEBUG "LEB %-7d free %-8d dirty %-8d used %-8d "
- "free + dirty %-8d dark %-4d dead %-4d nodes fit %-3d "
- "flags %#-4x (", lp->lnum, lp->free, lp->dirty,
- c->leb_size - spc, spc, dark, dead,
- (int)(spc / UBIFS_MAX_NODE_SZ), lp->flags);
+ pr_err("LEB %-7d free %-8d dirty %-8d used %-8d free + dirty %-8d dark %-4d dead %-4d nodes fit %-3d flags %#-4x (",
+ lp->lnum, lp->free, lp->dirty, c->leb_size - spc, spc,
+ dark, dead, (int)(spc / UBIFS_MAX_NODE_SZ), lp->flags);
if (lp->flags & LPROPS_TAKEN) {
if (lp->flags & LPROPS_INDEX)
- printk(KERN_CONT "index, taken");
+ pr_cont("index, taken");
else
- printk(KERN_CONT "taken");
+ pr_cont("taken");
} else {
const char *s;
@@ -721,7 +704,7 @@ void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp)
break;
}
}
- printk(KERN_CONT "%s", s);
+ pr_cont("%s", s);
}
for (rb = rb_first((struct rb_root *)&c->buds); rb; rb = rb_next(rb)) {
@@ -729,120 +712,147 @@ 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) {
- printk(KERN_CONT ", jhead %s",
- dbg_jhead(i));
+ /*
+ * 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) {
+ pr_cont(", jhead %s", dbg_jhead(i));
head = 1;
}
}
if (!head)
- printk(KERN_CONT ", bud of jhead %s",
+ pr_cont(", bud of jhead %s",
dbg_jhead(bud->jhead));
}
}
if (lp->lnum == c->gc_lnum)
- printk(KERN_CONT ", GC LEB");
- printk(KERN_CONT ")\n");
+ pr_cont(", GC LEB");
+ pr_cont(")\n");
}
-void dbg_dump_lprops(struct ubifs_info *c)
+void ubifs_dump_lprops(struct ubifs_info *c)
{
int lnum, err;
struct ubifs_lprops lp;
struct ubifs_lp_stats lst;
- printk(KERN_DEBUG "(pid %d) start dumping LEB properties\n",
- current->pid);
+ pr_err("(pid %d) start dumping LEB properties\n", current->pid);
ubifs_get_lp_stats(c, &lst);
- dbg_dump_lstats(&lst);
+ ubifs_dump_lstats(&lst);
for (lnum = c->main_first; lnum < c->leb_cnt; lnum++) {
err = ubifs_read_one_lp(c, lnum, &lp);
- if (err)
+ if (err) {
ubifs_err("cannot read lprops for LEB %d", lnum);
+ continue;
+ }
- dbg_dump_lprop(c, &lp);
+ ubifs_dump_lprop(c, &lp);
}
- printk(KERN_DEBUG "(pid %d) finish dumping LEB properties\n",
- current->pid);
+ pr_err("(pid %d) finish dumping LEB properties\n", current->pid);
}
-void dbg_dump_lpt_info(struct ubifs_info *c)
+void ubifs_dump_lpt_info(struct ubifs_info *c)
{
int i;
spin_lock(&dbg_lock);
- printk(KERN_DEBUG "(pid %d) dumping LPT information\n", current->pid);
- printk(KERN_DEBUG "\tlpt_sz: %lld\n", c->lpt_sz);
- printk(KERN_DEBUG "\tpnode_sz: %d\n", c->pnode_sz);
- printk(KERN_DEBUG "\tnnode_sz: %d\n", c->nnode_sz);
- printk(KERN_DEBUG "\tltab_sz: %d\n", c->ltab_sz);
- printk(KERN_DEBUG "\tlsave_sz: %d\n", c->lsave_sz);
- printk(KERN_DEBUG "\tbig_lpt: %d\n", c->big_lpt);
- printk(KERN_DEBUG "\tlpt_hght: %d\n", c->lpt_hght);
- printk(KERN_DEBUG "\tpnode_cnt: %d\n", c->pnode_cnt);
- printk(KERN_DEBUG "\tnnode_cnt: %d\n", c->nnode_cnt);
- printk(KERN_DEBUG "\tdirty_pn_cnt: %d\n", c->dirty_pn_cnt);
- printk(KERN_DEBUG "\tdirty_nn_cnt: %d\n", c->dirty_nn_cnt);
- printk(KERN_DEBUG "\tlsave_cnt: %d\n", c->lsave_cnt);
- printk(KERN_DEBUG "\tspace_bits: %d\n", c->space_bits);
- printk(KERN_DEBUG "\tlpt_lnum_bits: %d\n", c->lpt_lnum_bits);
- printk(KERN_DEBUG "\tlpt_offs_bits: %d\n", c->lpt_offs_bits);
- printk(KERN_DEBUG "\tlpt_spc_bits: %d\n", c->lpt_spc_bits);
- printk(KERN_DEBUG "\tpcnt_bits: %d\n", c->pcnt_bits);
- printk(KERN_DEBUG "\tlnum_bits: %d\n", c->lnum_bits);
- printk(KERN_DEBUG "\tLPT root is at %d:%d\n", c->lpt_lnum, c->lpt_offs);
- printk(KERN_DEBUG "\tLPT head is at %d:%d\n",
+ pr_err("(pid %d) dumping LPT information\n", current->pid);
+ pr_err("\tlpt_sz: %lld\n", c->lpt_sz);
+ pr_err("\tpnode_sz: %d\n", c->pnode_sz);
+ pr_err("\tnnode_sz: %d\n", c->nnode_sz);
+ pr_err("\tltab_sz: %d\n", c->ltab_sz);
+ pr_err("\tlsave_sz: %d\n", c->lsave_sz);
+ pr_err("\tbig_lpt: %d\n", c->big_lpt);
+ pr_err("\tlpt_hght: %d\n", c->lpt_hght);
+ pr_err("\tpnode_cnt: %d\n", c->pnode_cnt);
+ pr_err("\tnnode_cnt: %d\n", c->nnode_cnt);
+ pr_err("\tdirty_pn_cnt: %d\n", c->dirty_pn_cnt);
+ pr_err("\tdirty_nn_cnt: %d\n", c->dirty_nn_cnt);
+ pr_err("\tlsave_cnt: %d\n", c->lsave_cnt);
+ pr_err("\tspace_bits: %d\n", c->space_bits);
+ pr_err("\tlpt_lnum_bits: %d\n", c->lpt_lnum_bits);
+ pr_err("\tlpt_offs_bits: %d\n", c->lpt_offs_bits);
+ pr_err("\tlpt_spc_bits: %d\n", c->lpt_spc_bits);
+ pr_err("\tpcnt_bits: %d\n", c->pcnt_bits);
+ pr_err("\tlnum_bits: %d\n", c->lnum_bits);
+ pr_err("\tLPT root is at %d:%d\n", c->lpt_lnum, c->lpt_offs);
+ pr_err("\tLPT head is at %d:%d\n",
c->nhead_lnum, c->nhead_offs);
- printk(KERN_DEBUG "\tLPT ltab is at %d:%d\n",
- c->ltab_lnum, c->ltab_offs);
+ pr_err("\tLPT ltab is at %d:%d\n", c->ltab_lnum, c->ltab_offs);
if (c->big_lpt)
- printk(KERN_DEBUG "\tLPT lsave is at %d:%d\n",
+ pr_err("\tLPT lsave is at %d:%d\n",
c->lsave_lnum, c->lsave_offs);
for (i = 0; i < c->lpt_lebs; i++)
- printk(KERN_DEBUG "\tLPT LEB %d free %d dirty %d tgc %d "
- "cmt %d\n", i + c->lpt_first, c->ltab[i].free,
- c->ltab[i].dirty, c->ltab[i].tgc, c->ltab[i].cmt);
+ pr_err("\tLPT LEB %d free %d dirty %d tgc %d cmt %d\n",
+ i + c->lpt_first, c->ltab[i].free, c->ltab[i].dirty,
+ c->ltab[i].tgc, c->ltab[i].cmt);
spin_unlock(&dbg_lock);
}
-void dbg_dump_leb(const struct ubifs_info *c, int lnum)
+void ubifs_dump_sleb(const struct ubifs_info *c,
+ const struct ubifs_scan_leb *sleb, int offs)
+{
+ struct ubifs_scan_node *snod;
+
+ pr_err("(pid %d) start dumping scanned data from LEB %d:%d\n",
+ current->pid, sleb->lnum, offs);
+
+ list_for_each_entry(snod, &sleb->nodes, list) {
+ cond_resched();
+ pr_err("Dumping node at LEB %d:%d len %d\n",
+ sleb->lnum, snod->offs, snod->len);
+ ubifs_dump_node(c, snod->node);
+ }
+}
+
+void ubifs_dump_leb(const struct ubifs_info *c, int lnum)
{
struct ubifs_scan_leb *sleb;
struct ubifs_scan_node *snod;
+ void *buf;
+
+ pr_err("(pid %d) start dumping LEB %d\n", current->pid, lnum);
- if (dbg_failure_mode)
+ buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+ if (!buf) {
+ ubifs_err("cannot allocate memory for dumping LEB %d", lnum);
return;
+ }
- printk(KERN_DEBUG "(pid %d) start dumping LEB %d\n",
- current->pid, lnum);
- sleb = ubifs_scan(c, lnum, 0, c->dbg->buf, 0);
+ 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,
+ pr_err("LEB %d has %d nodes ending at %d\n", lnum,
sleb->nodes_cnt, sleb->endpt);
list_for_each_entry(snod, &sleb->nodes, list) {
cond_resched();
- printk(KERN_DEBUG "Dumping node at LEB %d:%d len %d\n", lnum,
+ pr_err("Dumping node at LEB %d:%d len %d\n", lnum,
snod->offs, snod->len);
- dbg_dump_node(c, snod->node);
+ ubifs_dump_node(c, snod->node);
}
- printk(KERN_DEBUG "(pid %d) finish dumping LEB %d\n",
- current->pid, lnum);
+ pr_err("(pid %d) finish dumping LEB %d\n", current->pid, lnum);
ubifs_scan_destroy(sleb);
+
+out:
+ vfree(buf);
return;
}
-void dbg_dump_znode(const struct ubifs_info *c,
- const struct ubifs_znode *znode)
+void ubifs_dump_znode(const struct ubifs_info *c,
+ const struct ubifs_znode *znode)
{
int n;
const struct ubifs_zbranch *zbr;
+ char key_buf[DBG_KEY_BUF_LEN];
spin_lock(&dbg_lock);
if (znode->parent)
@@ -850,103 +860,102 @@ void dbg_dump_znode(const struct ubifs_info *c,
else
zbr = &c->zroot;
- printk(KERN_DEBUG "znode %p, LEB %d:%d len %d parent %p iip %d level %d"
- " child_cnt %d flags %lx\n", znode, zbr->lnum, zbr->offs,
- zbr->len, znode->parent, znode->iip, znode->level,
- znode->child_cnt, znode->flags);
+ pr_err("znode %p, LEB %d:%d len %d parent %p iip %d level %d child_cnt %d flags %lx\n",
+ znode, zbr->lnum, zbr->offs, zbr->len, znode->parent, znode->iip,
+ znode->level, znode->child_cnt, znode->flags);
if (znode->child_cnt <= 0 || znode->child_cnt > c->fanout) {
spin_unlock(&dbg_lock);
return;
}
- printk(KERN_DEBUG "zbranches:\n");
+ pr_err("zbranches:\n");
for (n = 0; n < znode->child_cnt; n++) {
zbr = &znode->zbranch[n];
if (znode->level > 0)
- printk(KERN_DEBUG "\t%d: znode %p LEB %d:%d len %d key "
- "%s\n", n, zbr->znode, zbr->lnum,
- zbr->offs, zbr->len,
- DBGKEY(&zbr->key));
+ pr_err("\t%d: znode %p LEB %d:%d len %d key %s\n",
+ n, zbr->znode, zbr->lnum, zbr->offs, zbr->len,
+ dbg_snprintf_key(c, &zbr->key, key_buf,
+ DBG_KEY_BUF_LEN));
else
- printk(KERN_DEBUG "\t%d: LNC %p LEB %d:%d len %d key "
- "%s\n", n, zbr->znode, zbr->lnum,
- zbr->offs, zbr->len,
- DBGKEY(&zbr->key));
+ pr_err("\t%d: LNC %p LEB %d:%d len %d key %s\n",
+ n, zbr->znode, zbr->lnum, zbr->offs, zbr->len,
+ dbg_snprintf_key(c, &zbr->key, key_buf,
+ DBG_KEY_BUF_LEN));
}
spin_unlock(&dbg_lock);
}
-void dbg_dump_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat)
+void ubifs_dump_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat)
{
int i;
- printk(KERN_DEBUG "(pid %d) start dumping heap cat %d (%d elements)\n",
+ pr_err("(pid %d) start dumping heap cat %d (%d elements)\n",
current->pid, cat, heap->cnt);
for (i = 0; i < heap->cnt; i++) {
struct ubifs_lprops *lprops = heap->arr[i];
- printk(KERN_DEBUG "\t%d. LEB %d hpos %d free %d dirty %d "
- "flags %d\n", i, lprops->lnum, lprops->hpos,
- lprops->free, lprops->dirty, lprops->flags);
+ pr_err("\t%d. LEB %d hpos %d free %d dirty %d flags %d\n",
+ i, lprops->lnum, lprops->hpos, lprops->free,
+ lprops->dirty, lprops->flags);
}
- printk(KERN_DEBUG "(pid %d) finish dumping heap\n", current->pid);
+ pr_err("(pid %d) finish dumping heap\n", current->pid);
}
-void dbg_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
- struct ubifs_nnode *parent, int iip)
+void ubifs_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
+ struct ubifs_nnode *parent, int iip)
{
int i;
- printk(KERN_DEBUG "(pid %d) dumping pnode:\n", current->pid);
- printk(KERN_DEBUG "\taddress %zx parent %zx cnext %zx\n",
+ pr_err("(pid %d) dumping pnode:\n", current->pid);
+ pr_err("\taddress %zx parent %zx cnext %zx\n",
(size_t)pnode, (size_t)parent, (size_t)pnode->cnext);
- printk(KERN_DEBUG "\tflags %lu iip %d level %d num %d\n",
+ pr_err("\tflags %lu iip %d level %d num %d\n",
pnode->flags, iip, pnode->level, pnode->num);
for (i = 0; i < UBIFS_LPT_FANOUT; i++) {
struct ubifs_lprops *lp = &pnode->lprops[i];
- printk(KERN_DEBUG "\t%d: free %d dirty %d flags %d lnum %d\n",
+ pr_err("\t%d: free %d dirty %d flags %d lnum %d\n",
i, lp->free, lp->dirty, lp->flags, lp->lnum);
}
}
-void dbg_dump_tnc(struct ubifs_info *c)
+void ubifs_dump_tnc(struct ubifs_info *c)
{
struct ubifs_znode *znode;
int level;
- printk(KERN_DEBUG "\n");
- printk(KERN_DEBUG "(pid %d) start dumping TNC tree\n", current->pid);
+ pr_err("\n");
+ pr_err("(pid %d) start dumping TNC tree\n", current->pid);
znode = ubifs_tnc_levelorder_next(c->zroot.znode, NULL);
level = znode->level;
- printk(KERN_DEBUG "== Level %d ==\n", level);
+ pr_err("== Level %d ==\n", level);
while (znode) {
if (level != znode->level) {
level = znode->level;
- printk(KERN_DEBUG "== Level %d ==\n", level);
+ pr_err("== Level %d ==\n", level);
}
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
znode = ubifs_tnc_levelorder_next(c->zroot.znode, znode);
}
- printk(KERN_DEBUG "(pid %d) finish dumping TNC tree\n", current->pid);
+ pr_err("(pid %d) finish dumping TNC tree\n", current->pid);
}
static int dump_znode(struct ubifs_info *c, struct ubifs_znode *znode,
void *priv)
{
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
return 0;
}
/**
- * dbg_dump_index - dump the on-flash index.
+ * ubifs_dump_index - dump the on-flash index.
* @c: UBIFS file-system description object
*
- * This function dumps whole UBIFS indexing B-tree, unlike 'dbg_dump_tnc()'
+ * This function dumps whole UBIFS indexing B-tree, unlike 'ubifs_dump_tnc()'
* which dumps only in-memory znodes and does not read znodes which from flash.
*/
-void dbg_dump_index(struct ubifs_info *c)
+void ubifs_dump_index(struct ubifs_info *c)
{
dbg_walk_index(c, NULL, dump_znode, NULL);
}
@@ -961,11 +970,41 @@ 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 +1021,15 @@ 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",
@@ -999,21 +1041,22 @@ 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_dump_lstats(&d->saved_lst);
+ ubifs_msg("saved budgeting info dump");
+ ubifs_dump_budg(c, &d->saved_bi);
+ ubifs_msg("saved idx_gc_cnt %d", d->saved_idx_gc_cnt);
ubifs_msg("current lprops statistics dump");
- dbg_dump_lstats(&lst);
-
- spin_lock(&c->space_lock);
- dbg_dump_budg(c);
- spin_unlock(&c->space_lock);
+ ubifs_get_lp_stats(c, &lst);
+ ubifs_dump_lstats(&lst);
+ ubifs_msg("current budgeting info dump");
+ ubifs_dump_budg(c, &c->bi);
dump_stack();
return -EINVAL;
}
/**
* dbg_check_synced_i_size - check synchronized inode size.
+ * @c: UBIFS file-system description object
* @inode: inode to check
*
* If inode is clean, synchronized inode size has to be equivalent to current
@@ -1021,12 +1064,12 @@ out:
* has to be locked). Returns %0 if synchronized inode size if correct, and
* %-EINVAL if not.
*/
-int dbg_check_synced_i_size(struct inode *inode)
+int dbg_check_synced_i_size(const struct ubifs_info *c, struct inode *inode)
{
int err = 0;
struct ubifs_inode *ui = ubifs_inode(inode);
- if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+ if (!dbg_is_chk_gen(c))
return 0;
if (!S_ISREG(inode->i_mode))
return 0;
@@ -1034,11 +1077,11 @@ int dbg_check_synced_i_size(struct inode *inode)
mutex_lock(&ui->ui_mutex);
spin_lock(&ui->ui_lock);
if (ui->ui_size != ui->synced_i_size && !ui->dirty) {
- ubifs_err("ui_size is %lld, synced_i_size is %lld, but inode "
- "is clean", ui->ui_size, ui->synced_i_size);
+ ubifs_err("ui_size is %lld, synced_i_size is %lld, but inode is clean",
+ ui->ui_size, ui->synced_i_size);
ubifs_err("i_ino %lu, i_mode %#x, i_size %lld", inode->i_ino,
inode->i_mode, i_size_read(inode));
- dbg_dump_stack();
+ dump_stack();
err = -EINVAL;
}
spin_unlock(&ui->ui_lock);
@@ -1059,7 +1102,7 @@ int dbg_check_synced_i_size(struct inode *inode)
* Note, it is good idea to make sure the @dir->i_mutex is locked before
* calling this function.
*/
-int dbg_check_dir_size(struct ubifs_info *c, const struct inode *dir)
+int dbg_check_dir(struct ubifs_info *c, const struct inode *dir)
{
unsigned int nlink = 2;
union ubifs_key key;
@@ -1067,7 +1110,7 @@ int dbg_check_dir_size(struct ubifs_info *c, const struct inode *dir)
struct qstr nm = { .name = NULL };
loff_t size = UBIFS_INO_NODE_SZ;
- if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+ if (!dbg_is_chk_gen(c))
return 0;
if (!S_ISDIR(dir->i_mode))
@@ -1097,16 +1140,17 @@ int dbg_check_dir_size(struct ubifs_info *c, const struct inode *dir)
kfree(pdent);
if (i_size_read(dir) != size) {
- ubifs_err("directory inode %lu has size %llu, "
- "but calculated size is %llu", dir->i_ino,
- (unsigned long long)i_size_read(dir),
+ ubifs_err("directory inode %lu has size %llu, but calculated size is %llu",
+ dir->i_ino, (unsigned long long)i_size_read(dir),
(unsigned long long)size);
+ ubifs_dump_inode(c, dir);
dump_stack();
return -EINVAL;
}
if (dir->i_nlink != nlink) {
- ubifs_err("directory inode %lu has nlink %u, but calculated "
- "nlink is %u", dir->i_ino, dir->i_nlink, nlink);
+ ubifs_err("directory inode %lu has nlink %u, but calculated nlink is %u",
+ dir->i_ino, dir->i_nlink, nlink);
+ ubifs_dump_inode(c, dir);
dump_stack();
return -EINVAL;
}
@@ -1133,6 +1177,7 @@ static int dbg_check_key_order(struct ubifs_info *c, struct ubifs_zbranch *zbr1,
int err, nlen1, nlen2, cmp;
struct ubifs_dent_node *dent1, *dent2;
union ubifs_key key;
+ char key_buf[DBG_KEY_BUF_LEN];
ubifs_assert(!keys_cmp(c, &zbr1->key, &zbr2->key));
dent1 = kmalloc(UBIFS_MAX_DENT_NODE_SZ, GFP_NOFS);
@@ -1162,21 +1207,25 @@ static int dbg_check_key_order(struct ubifs_info *c, struct ubifs_zbranch *zbr1,
err = 1;
key_read(c, &dent1->key, &key);
if (keys_cmp(c, &zbr1->key, &key)) {
- dbg_err("1st entry at %d:%d has key %s", zbr1->lnum,
- zbr1->offs, DBGKEY(&key));
- dbg_err("but it should have key %s according to tnc",
- DBGKEY(&zbr1->key));
- dbg_dump_node(c, dent1);
+ ubifs_err("1st entry at %d:%d has key %s", zbr1->lnum,
+ zbr1->offs, dbg_snprintf_key(c, &key, key_buf,
+ DBG_KEY_BUF_LEN));
+ ubifs_err("but it should have key %s according to tnc",
+ dbg_snprintf_key(c, &zbr1->key, key_buf,
+ DBG_KEY_BUF_LEN));
+ ubifs_dump_node(c, dent1);
goto out_free;
}
key_read(c, &dent2->key, &key);
if (keys_cmp(c, &zbr2->key, &key)) {
- dbg_err("2nd entry at %d:%d has key %s", zbr1->lnum,
- zbr1->offs, DBGKEY(&key));
- dbg_err("but it should have key %s according to tnc",
- DBGKEY(&zbr2->key));
- dbg_dump_node(c, dent2);
+ ubifs_err("2nd entry at %d:%d has key %s", zbr1->lnum,
+ zbr1->offs, dbg_snprintf_key(c, &key, key_buf,
+ DBG_KEY_BUF_LEN));
+ ubifs_err("but it should have key %s according to tnc",
+ dbg_snprintf_key(c, &zbr2->key, key_buf,
+ DBG_KEY_BUF_LEN));
+ ubifs_dump_node(c, dent2);
goto out_free;
}
@@ -1189,15 +1238,15 @@ static int dbg_check_key_order(struct ubifs_info *c, struct ubifs_zbranch *zbr1,
goto out_free;
}
if (cmp == 0 && nlen1 == nlen2)
- dbg_err("2 xent/dent nodes with the same name");
+ ubifs_err("2 xent/dent nodes with the same name");
else
- dbg_err("bad order of colliding key %s",
- DBGKEY(&key));
+ ubifs_err("bad order of colliding key %s",
+ dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN));
ubifs_msg("first node at %d:%d\n", zbr1->lnum, zbr1->offs);
- dbg_dump_node(c, dent1);
+ ubifs_dump_node(c, dent1);
ubifs_msg("second node at %d:%d\n", zbr2->lnum, zbr2->offs);
- dbg_dump_node(c, dent2);
+ ubifs_dump_node(c, dent2);
out_free:
kfree(dent2);
@@ -1400,10 +1449,10 @@ static int dbg_check_znode(struct ubifs_info *c, struct ubifs_zbranch *zbr)
out:
ubifs_err("failed, error %d", err);
ubifs_msg("dump of the znode");
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
if (zp) {
ubifs_msg("dump of the parent znode");
- dbg_dump_znode(c, zp);
+ ubifs_dump_znode(c, zp);
}
dump_stack();
return -EINVAL;
@@ -1423,7 +1472,7 @@ int dbg_check_tnc(struct ubifs_info *c, int extra)
long clean_cnt = 0, dirty_cnt = 0;
int err, last;
- if (!(ubifs_chk_flags & UBIFS_CHK_TNC))
+ if (!dbg_is_chk_index(c))
return 0;
ubifs_assert(mutex_is_locked(&c->tnc_mutex));
@@ -1470,9 +1519,9 @@ int dbg_check_tnc(struct ubifs_info *c, int extra)
return err;
if (err) {
ubifs_msg("first znode");
- dbg_dump_znode(c, prev);
+ ubifs_dump_znode(c, prev);
ubifs_msg("second znode");
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
return -EINVAL;
}
}
@@ -1559,9 +1608,9 @@ int dbg_walk_index(struct ubifs_info *c, dbg_leaf_callback leaf_cb,
if (znode_cb) {
err = znode_cb(c, znode, priv);
if (err) {
- ubifs_err("znode checking function returned "
- "error %d", err);
- dbg_dump_znode(c, znode);
+ ubifs_err("znode checking function returned error %d",
+ err);
+ ubifs_dump_znode(c, znode);
goto out_dump;
}
}
@@ -1570,9 +1619,7 @@ int dbg_walk_index(struct ubifs_info *c, dbg_leaf_callback leaf_cb,
zbr = &znode->zbranch[idx];
err = leaf_cb(c, zbr, priv);
if (err) {
- ubifs_err("leaf checking function "
- "returned error %d, for leaf "
- "at LEB %d:%d",
+ ubifs_err("leaf checking function returned error %d, for leaf at LEB %d:%d",
err, zbr->lnum, zbr->offs);
goto out_dump;
}
@@ -1629,7 +1676,7 @@ out_dump:
else
zbr = &c->zroot;
ubifs_msg("dump of znode at LEB %d:%d", zbr->lnum, zbr->offs);
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
out_unlock:
mutex_unlock(&c->tnc_mutex);
return err;
@@ -1670,7 +1717,7 @@ int dbg_check_idx_size(struct ubifs_info *c, long long idx_size)
int err;
long long calc = 0;
- if (!(ubifs_chk_flags & UBIFS_CHK_IDX_SZ))
+ if (!dbg_is_chk_index(c))
return 0;
err = dbg_walk_index(c, NULL, add_size, &calc);
@@ -1680,8 +1727,8 @@ int dbg_check_idx_size(struct ubifs_info *c, long long idx_size)
}
if (calc != idx_size) {
- ubifs_err("index size check failed: calculated size is %lld, "
- "should be %lld", calc, idx_size);
+ ubifs_err("index size check failed: calculated size is %lld, should be %lld",
+ calc, idx_size);
dump_stack();
return -EINVAL;
}
@@ -1751,6 +1798,8 @@ 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 +1823,46 @@ 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;
}
@@ -1964,8 +2040,7 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr,
fscki = read_add_inode(c, priv, inum);
if (IS_ERR(fscki)) {
err = PTR_ERR(fscki);
- ubifs_err("error %d while processing data node and "
- "trying to find inode node %lu",
+ ubifs_err("error %d while processing data node and trying to find inode node %lu",
err, (unsigned long)inum);
goto out_dump;
}
@@ -1975,9 +2050,8 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr,
blk_offs <<= UBIFS_BLOCK_SHIFT;
blk_offs += le32_to_cpu(dn->size);
if (blk_offs > fscki->size) {
- ubifs_err("data node at LEB %d:%d is not within inode "
- "size %lld", zbr->lnum, zbr->offs,
- fscki->size);
+ ubifs_err("data node at LEB %d:%d is not within inode size %lld",
+ zbr->lnum, zbr->offs, fscki->size);
err = -EINVAL;
goto out_dump;
}
@@ -1998,8 +2072,7 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr,
fscki = read_add_inode(c, priv, inum);
if (IS_ERR(fscki)) {
err = PTR_ERR(fscki);
- ubifs_err("error %d while processing entry node and "
- "trying to find inode node %lu",
+ ubifs_err("error %d while processing entry node and trying to find inode node %lu",
err, (unsigned long)inum);
goto out_dump;
}
@@ -2011,8 +2084,7 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr,
fscki1 = read_add_inode(c, priv, inum);
if (IS_ERR(fscki1)) {
err = PTR_ERR(fscki1);
- ubifs_err("error %d while processing entry node and "
- "trying to find parent inode node %lu",
+ ubifs_err("error %d while processing entry node and trying to find parent inode node %lu",
err, (unsigned long)inum);
goto out_dump;
}
@@ -2036,7 +2108,7 @@ out:
out_dump:
ubifs_msg("dump of node at LEB %d:%d", zbr->lnum, zbr->offs);
- dbg_dump_node(c, node);
+ ubifs_dump_node(c, node);
out_free:
kfree(node);
return err;
@@ -2048,26 +2120,10 @@ out_free:
*/
static void free_inodes(struct fsck_data *fsckd)
{
- struct rb_node *this = fsckd->inodes.rb_node;
- struct fsck_inode *fscki;
+ struct fsck_inode *fscki, *n;
- while (this) {
- if (this->rb_left)
- this = this->rb_left;
- else if (this->rb_right)
- this = this->rb_right;
- else {
- fscki = rb_entry(this, struct fsck_inode, rb);
- this = rb_parent(this);
- if (this) {
- if (this->rb_left == &fscki->rb)
- this->rb_left = NULL;
- else
- this->rb_right = NULL;
- }
- kfree(fscki);
- }
- }
+ rbtree_postorder_for_each_entry_safe(fscki, n, &fsckd->inodes, rb)
+ kfree(fscki);
}
/**
@@ -2102,61 +2158,52 @@ static int check_inodes(struct ubifs_info *c, struct fsck_data *fsckd)
*/
if (fscki->inum != UBIFS_ROOT_INO &&
fscki->references != 1) {
- ubifs_err("directory inode %lu has %d "
- "direntries which refer it, but "
- "should be 1",
+ ubifs_err("directory inode %lu has %d direntries which refer it, but should be 1",
(unsigned long)fscki->inum,
fscki->references);
goto out_dump;
}
if (fscki->inum == UBIFS_ROOT_INO &&
fscki->references != 0) {
- ubifs_err("root inode %lu has non-zero (%d) "
- "direntries which refer it",
+ ubifs_err("root inode %lu has non-zero (%d) direntries which refer it",
(unsigned long)fscki->inum,
fscki->references);
goto out_dump;
}
if (fscki->calc_sz != fscki->size) {
- ubifs_err("directory inode %lu size is %lld, "
- "but calculated size is %lld",
+ ubifs_err("directory inode %lu size is %lld, but calculated size is %lld",
(unsigned long)fscki->inum,
fscki->size, fscki->calc_sz);
goto out_dump;
}
if (fscki->calc_cnt != fscki->nlink) {
- ubifs_err("directory inode %lu nlink is %d, "
- "but calculated nlink is %d",
+ ubifs_err("directory inode %lu nlink is %d, but calculated nlink is %d",
(unsigned long)fscki->inum,
fscki->nlink, fscki->calc_cnt);
goto out_dump;
}
} else {
if (fscki->references != fscki->nlink) {
- ubifs_err("inode %lu nlink is %d, but "
- "calculated nlink is %d",
+ ubifs_err("inode %lu nlink is %d, but calculated nlink is %d",
(unsigned long)fscki->inum,
fscki->nlink, fscki->references);
goto out_dump;
}
}
if (fscki->xattr_sz != fscki->calc_xsz) {
- ubifs_err("inode %lu has xattr size %u, but "
- "calculated size is %lld",
+ ubifs_err("inode %lu has xattr size %u, but calculated size is %lld",
(unsigned long)fscki->inum, fscki->xattr_sz,
fscki->calc_xsz);
goto out_dump;
}
if (fscki->xattr_cnt != fscki->calc_xcnt) {
- ubifs_err("inode %lu has %u xattrs, but "
- "calculated count is %lld",
+ ubifs_err("inode %lu has %u xattrs, but calculated count is %lld",
(unsigned long)fscki->inum,
fscki->xattr_cnt, fscki->calc_xcnt);
goto out_dump;
}
if (fscki->xattr_nms != fscki->calc_xnms) {
- ubifs_err("inode %lu has xattr names' size %u, but "
- "calculated names' size is %lld",
+ ubifs_err("inode %lu has xattr names' size %u, but calculated names' size is %lld",
(unsigned long)fscki->inum, fscki->xattr_nms,
fscki->calc_xnms);
goto out_dump;
@@ -2194,7 +2241,7 @@ out_dump:
ubifs_msg("dump of the inode %lu sitting in LEB %d:%d",
(unsigned long)fscki->inum, zbr->lnum, zbr->offs);
- dbg_dump_node(c, ino);
+ ubifs_dump_node(c, ino);
kfree(ino);
return -EINVAL;
}
@@ -2217,7 +2264,7 @@ int dbg_check_filesystem(struct ubifs_info *c)
int err;
struct fsck_data fsckd;
- if (!(ubifs_chk_flags & UBIFS_CHK_FS))
+ if (!dbg_is_chk_fs(c))
return 0;
fsckd.inodes = RB_ROOT;
@@ -2252,7 +2299,7 @@ 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))
+ if (!dbg_is_chk_gen(c))
return 0;
for (cur = head->next; cur->next != head; cur = cur->next) {
@@ -2265,12 +2312,12 @@ int dbg_check_data_nodes_order(struct ubifs_info *c, struct list_head *head)
if (sa->type != UBIFS_DATA_NODE) {
ubifs_err("bad node type %d", sa->type);
- dbg_dump_node(c, sa->node);
+ ubifs_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);
+ ubifs_dump_node(c, sb->node);
return -EINVAL;
}
@@ -2301,8 +2348,8 @@ int dbg_check_data_nodes_order(struct ubifs_info *c, struct list_head *head)
return 0;
error_dump:
- dbg_dump_node(c, sa->node);
- dbg_dump_node(c, sb->node);
+ ubifs_dump_node(c, sa->node);
+ ubifs_dump_node(c, sb->node);
return -EINVAL;
}
@@ -2319,7 +2366,7 @@ 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))
+ if (!dbg_is_chk_gen(c))
return 0;
for (cur = head->next; cur->next != head; cur = cur->next) {
@@ -2333,13 +2380,13 @@ int dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head)
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);
+ ubifs_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);
+ ubifs_dump_node(c, sb->node);
return -EINVAL;
}
@@ -2379,7 +2426,8 @@ int dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head)
hashb = key_block(c, &sb->key);
if (hasha > hashb) {
- ubifs_err("larger hash %u goes before %u", hasha, hashb);
+ ubifs_err("larger hash %u goes before %u",
+ hasha, hashb);
goto error_dump;
}
}
@@ -2388,400 +2436,363 @@ int dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head)
error_dump:
ubifs_msg("dumping first node");
- dbg_dump_node(c, sa->node);
+ ubifs_dump_node(c, sa->node);
ubifs_msg("dumping second node");
- dbg_dump_node(c, sb->node);
+ ubifs_dump_node(c, sb->node);
return -EINVAL;
return 0;
}
-static int invocation_cnt;
-
-int dbg_force_in_the_gaps(void)
-{
- if (!dbg_force_in_the_gaps_enabled)
- return 0;
- /* Force in-the-gaps every 8th commit */
- return !((invocation_cnt++) & 0x7);
-}
-
-/* Failure mode for recovery testing */
-
-#define chance(n, d) (simple_rand() <= (n) * 32768LL / (d))
-
-struct failure_mode_info {
- struct list_head list;
- struct ubifs_info *c;
-};
-
-static LIST_HEAD(fmi_list);
-static DEFINE_SPINLOCK(fmi_lock);
-
-static unsigned int next;
-
-static int simple_rand(void)
-{
- if (next == 0)
- next = current->pid;
- next = next * 1103515245 + 12345;
- return (next >> 16) & 32767;
-}
-
-static void failure_mode_init(struct ubifs_info *c)
+static inline int chance(unsigned int n, unsigned int out_of)
{
- struct failure_mode_info *fmi;
+ return !!((prandom_u32() % out_of) + 1 <= n);
- fmi = kmalloc(sizeof(struct failure_mode_info), GFP_NOFS);
- if (!fmi) {
- ubifs_err("Failed to register failure mode - no memory");
- return;
- }
- fmi->c = c;
- spin_lock(&fmi_lock);
- list_add_tail(&fmi->list, &fmi_list);
- spin_unlock(&fmi_lock);
}
-static void failure_mode_exit(struct ubifs_info *c)
+static int power_cut_emulated(struct ubifs_info *c, int lnum, int write)
{
- struct failure_mode_info *fmi, *tmp;
-
- spin_lock(&fmi_lock);
- list_for_each_entry_safe(fmi, tmp, &fmi_list, list)
- if (fmi->c == c) {
- list_del(&fmi->list);
- kfree(fmi);
- }
- spin_unlock(&fmi_lock);
-}
-
-static struct ubifs_info *dbg_find_info(struct ubi_volume_desc *desc)
-{
- struct failure_mode_info *fmi;
-
- spin_lock(&fmi_lock);
- list_for_each_entry(fmi, &fmi_list, list)
- if (fmi->c->ubi == desc) {
- struct ubifs_info *c = fmi->c;
-
- spin_unlock(&fmi_lock);
- return c;
- }
- spin_unlock(&fmi_lock);
- return NULL;
-}
-
-static int in_failure_mode(struct ubi_volume_desc *desc)
-{
- struct ubifs_info *c = dbg_find_info(desc);
-
- if (c && dbg_failure_mode)
- return c->dbg->failure_mode;
- return 0;
-}
+ struct ubifs_debug_info *d = c->dbg;
-static int do_fail(struct ubi_volume_desc *desc, int lnum, int write)
-{
- struct ubifs_info *c = dbg_find_info(desc);
- struct ubifs_debug_info *d;
+ ubifs_assert(dbg_is_tst_rcvry(c));
- if (!c || !dbg_failure_mode)
- return 0;
- d = c->dbg;
- if (d->failure_mode)
- return 1;
- if (!d->fail_cnt) {
- /* First call - decide delay to failure */
+ if (!d->pc_cnt) {
+ /* First call - decide delay to the power cut */
if (chance(1, 2)) {
- unsigned int delay = 1 << (simple_rand() >> 11);
+ unsigned long delay;
if (chance(1, 2)) {
- d->fail_delay = 1;
- d->fail_timeout = jiffies +
- msecs_to_jiffies(delay);
- dbg_rcvry("failing after %ums", delay);
+ d->pc_delay = 1;
+ /* Fail withing 1 minute */
+ delay = prandom_u32() % 60000;
+ d->pc_timeout = jiffies;
+ d->pc_timeout += msecs_to_jiffies(delay);
+ ubifs_warn("failing after %lums", delay);
} else {
- d->fail_delay = 2;
- d->fail_cnt_max = delay;
- dbg_rcvry("failing after %u calls", delay);
+ d->pc_delay = 2;
+ delay = prandom_u32() % 10000;
+ /* Fail within 10000 operations */
+ d->pc_cnt_max = delay;
+ ubifs_warn("failing after %lu calls", delay);
}
}
- d->fail_cnt += 1;
+
+ d->pc_cnt += 1;
}
+
/* Determine if failure delay has expired */
- if (d->fail_delay == 1) {
- if (time_before(jiffies, d->fail_timeout))
+ if (d->pc_delay == 1 && time_before(jiffies, d->pc_timeout))
return 0;
- } else if (d->fail_delay == 2)
- if (d->fail_cnt++ < d->fail_cnt_max)
+ if (d->pc_delay == 2 && d->pc_cnt++ < d->pc_cnt_max)
return 0;
+
if (lnum == UBIFS_SB_LNUM) {
- if (write) {
- if (chance(1, 2))
- return 0;
- } else if (chance(19, 20))
+ if (write && chance(1, 2))
return 0;
- dbg_rcvry("failing in super block LEB %d", lnum);
+ if (chance(19, 20))
+ return 0;
+ ubifs_warn("failing in super block LEB %d", lnum);
} else if (lnum == UBIFS_MST_LNUM || lnum == UBIFS_MST_LNUM + 1) {
if (chance(19, 20))
return 0;
- dbg_rcvry("failing in master LEB %d", lnum);
+ ubifs_warn("failing in master LEB %d", lnum);
} else if (lnum >= UBIFS_LOG_LNUM && lnum <= c->log_last) {
- if (write) {
- if (chance(99, 100))
- return 0;
- } else if (chance(399, 400))
+ if (write && chance(99, 100))
+ return 0;
+ if (chance(399, 400))
return 0;
- dbg_rcvry("failing in log LEB %d", lnum);
+ ubifs_warn("failing in log LEB %d", lnum);
} else if (lnum >= c->lpt_first && lnum <= c->lpt_last) {
- if (write) {
- if (chance(7, 8))
- return 0;
- } else if (chance(19, 20))
+ if (write && chance(7, 8))
+ return 0;
+ if (chance(19, 20))
return 0;
- dbg_rcvry("failing in LPT LEB %d", lnum);
+ ubifs_warn("failing in LPT LEB %d", lnum);
} else if (lnum >= c->orph_first && lnum <= c->orph_last) {
- if (write) {
- if (chance(1, 2))
- return 0;
- } else if (chance(9, 10))
+ if (write && chance(1, 2))
return 0;
- dbg_rcvry("failing in orphan LEB %d", lnum);
+ if (chance(9, 10))
+ return 0;
+ ubifs_warn("failing in orphan LEB %d", lnum);
} else if (lnum == c->ihead_lnum) {
if (chance(99, 100))
return 0;
- dbg_rcvry("failing in index head LEB %d", lnum);
+ ubifs_warn("failing in index head LEB %d", lnum);
} else if (c->jheads && lnum == c->jheads[GCHD].wbuf.lnum) {
if (chance(9, 10))
return 0;
- dbg_rcvry("failing in GC head LEB %d", lnum);
+ ubifs_warn("failing in GC head LEB %d", lnum);
} else if (write && !RB_EMPTY_ROOT(&c->buds) &&
!ubifs_search_bud(c, lnum)) {
if (chance(19, 20))
return 0;
- dbg_rcvry("failing in non-bud LEB %d", lnum);
+ ubifs_warn("failing in non-bud LEB %d", lnum);
} else if (c->cmt_state == COMMIT_RUNNING_BACKGROUND ||
c->cmt_state == COMMIT_RUNNING_REQUIRED) {
if (chance(999, 1000))
return 0;
- dbg_rcvry("failing in bud LEB %d commit running", lnum);
+ ubifs_warn("failing in bud LEB %d commit running", lnum);
} else {
if (chance(9999, 10000))
return 0;
- dbg_rcvry("failing in bud LEB %d commit not running", lnum);
+ ubifs_warn("failing in bud LEB %d commit not running", lnum);
}
- ubifs_err("*** SETTING FAILURE MODE ON (LEB %d) ***", lnum);
- d->failure_mode = 1;
+
+ d->pc_happened = 1;
+ ubifs_warn("========== Power cut emulated ==========");
dump_stack();
return 1;
}
-static void cut_data(const void *buf, int len)
+static int corrupt_data(const struct ubifs_info *c, const void *buf,
+ unsigned int len)
{
- int flen, i;
+ unsigned int from, to, ffs = chance(1, 2);
unsigned char *p = (void *)buf;
- flen = (len * (long long)simple_rand()) >> 15;
- for (i = flen; i < len; i++)
- p[i] = 0xff;
-}
+ from = prandom_u32() % len;
+ /* Corruption span max to end of write unit */
+ to = min(len, ALIGN(from + 1, c->max_write_size));
-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 ubi_leb_read(desc, lnum, buf, offset, len, check);
+ ubifs_warn("filled bytes %u-%u with %s", from, to - 1,
+ ffs ? "0xFFs" : "random data");
+
+ if (ffs)
+ memset(p + from, 0xFF, to - from);
+ else
+ prandom_bytes(p + from, to - from);
+
+ return to;
}
-int dbg_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
- int offset, int len, int dtype)
+int dbg_leb_write(struct ubifs_info *c, int lnum, const void *buf,
+ int offs, int len)
{
int err, failing;
- if (in_failure_mode(desc))
- return -EIO;
- failing = do_fail(desc, lnum, 1);
- if (failing)
- cut_data(buf, len);
- err = ubi_leb_write(desc, lnum, buf, offset, len, dtype);
+ if (c->dbg->pc_happened)
+ return -EROFS;
+
+ failing = power_cut_emulated(c, lnum, 1);
+ if (failing) {
+ len = corrupt_data(c, buf, len);
+ ubifs_warn("actually write %d bytes to LEB %d:%d (the buffer was corrupted)",
+ len, lnum, offs);
+ }
+ err = ubi_leb_write(c->ubi, lnum, buf, offs, len);
if (err)
return err;
if (failing)
- return -EIO;
+ return -EROFS;
return 0;
}
-int dbg_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf,
- int len, int dtype)
+int dbg_leb_change(struct ubifs_info *c, int lnum, const void *buf,
+ int len)
{
int err;
- if (do_fail(desc, lnum, 1))
- return -EIO;
- err = ubi_leb_change(desc, lnum, buf, len, dtype);
+ if (c->dbg->pc_happened)
+ return -EROFS;
+ if (power_cut_emulated(c, lnum, 1))
+ return -EROFS;
+ err = ubi_leb_change(c->ubi, lnum, buf, len);
if (err)
return err;
- if (do_fail(desc, lnum, 1))
- return -EIO;
+ if (power_cut_emulated(c, lnum, 1))
+ return -EROFS;
return 0;
}
-int dbg_leb_erase(struct ubi_volume_desc *desc, int lnum)
+int dbg_leb_unmap(struct ubifs_info *c, int lnum)
{
int err;
- if (do_fail(desc, lnum, 0))
- return -EIO;
- err = ubi_leb_erase(desc, lnum);
+ if (c->dbg->pc_happened)
+ return -EROFS;
+ if (power_cut_emulated(c, lnum, 0))
+ return -EROFS;
+ err = ubi_leb_unmap(c->ubi, lnum);
if (err)
return err;
- if (do_fail(desc, lnum, 0))
- return -EIO;
+ if (power_cut_emulated(c, lnum, 0))
+ return -EROFS;
return 0;
}
-int dbg_leb_unmap(struct ubi_volume_desc *desc, int lnum)
+int dbg_leb_map(struct ubifs_info *c, int lnum)
{
int err;
- if (do_fail(desc, lnum, 0))
- return -EIO;
- err = ubi_leb_unmap(desc, lnum);
+ if (c->dbg->pc_happened)
+ return -EROFS;
+ if (power_cut_emulated(c, lnum, 0))
+ return -EROFS;
+ err = ubi_leb_map(c->ubi, lnum);
if (err)
return err;
- if (do_fail(desc, lnum, 0))
- return -EIO;
+ if (power_cut_emulated(c, lnum, 0))
+ return -EROFS;
return 0;
}
-int dbg_is_mapped(struct ubi_volume_desc *desc, int lnum)
-{
- if (in_failure_mode(desc))
- return -EIO;
- return ubi_is_mapped(desc, lnum);
-}
+/*
+ * Root directory for UBIFS stuff in debugfs. Contains sub-directories which
+ * contain the stuff specific to particular file-system mounts.
+ */
+static struct dentry *dfs_rootdir;
-int dbg_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype)
+static int dfs_file_open(struct inode *inode, struct file *file)
{
- int err;
-
- if (do_fail(desc, lnum, 0))
- return -EIO;
- err = ubi_leb_map(desc, lnum, dtype);
- if (err)
- return err;
- if (do_fail(desc, lnum, 0))
- return -EIO;
- return 0;
+ file->private_data = inode->i_private;
+ return nonseekable_open(inode, file);
}
/**
- * ubifs_debugging_init - initialize UBIFS debugging.
- * @c: UBIFS file-system description object
+ * provide_user_output - provide output to the user reading a debugfs file.
+ * @val: boolean value for the answer
+ * @u: the buffer to store the answer at
+ * @count: size of the buffer
+ * @ppos: position in the @u output buffer
*
- * This function initializes debugging-related data for the file system.
- * Returns zero in case of success and a negative error code in case of
+ * This is a simple helper function which stores @val boolean value in the user
+ * buffer when the user reads one of UBIFS debugfs files. Returns amount of
+ * bytes written to @u in case of success and a negative error code in case of
* failure.
*/
-int ubifs_debugging_init(struct ubifs_info *c)
+static int provide_user_output(int val, char __user *u, size_t count,
+ loff_t *ppos)
{
- c->dbg = kzalloc(sizeof(struct ubifs_debug_info), GFP_KERNEL);
- if (!c->dbg)
- return -ENOMEM;
-
- c->dbg->buf = vmalloc(c->leb_size);
- if (!c->dbg->buf)
- goto out;
+ char buf[3];
- failure_mode_init(c);
- return 0;
+ if (val)
+ buf[0] = '1';
+ else
+ buf[0] = '0';
+ buf[1] = '\n';
+ buf[2] = 0x00;
-out:
- kfree(c->dbg);
- return -ENOMEM;
+ return simple_read_from_buffer(u, count, ppos, buf, 2);
}
-/**
- * ubifs_debugging_exit - free debugging data.
- * @c: UBIFS file-system description object
- */
-void ubifs_debugging_exit(struct ubifs_info *c)
+static ssize_t dfs_file_read(struct file *file, char __user *u, size_t count,
+ loff_t *ppos)
{
- failure_mode_exit(c);
- vfree(c->dbg->buf);
- kfree(c->dbg);
-}
+ struct dentry *dent = file->f_path.dentry;
+ struct ubifs_info *c = file->private_data;
+ struct ubifs_debug_info *d = c->dbg;
+ int val;
+
+ if (dent == d->dfs_chk_gen)
+ val = d->chk_gen;
+ else if (dent == d->dfs_chk_index)
+ val = d->chk_index;
+ else if (dent == d->dfs_chk_orph)
+ val = d->chk_orph;
+ else if (dent == d->dfs_chk_lprops)
+ val = d->chk_lprops;
+ else if (dent == d->dfs_chk_fs)
+ val = d->chk_fs;
+ else if (dent == d->dfs_tst_rcvry)
+ val = d->tst_rcvry;
+ else if (dent == d->dfs_ro_error)
+ val = c->ro_error;
+ else
+ return -EINVAL;
-/*
- * Root directory for UBIFS stuff in debugfs. Contains sub-directories which
- * contain the stuff specific to particular file-system mounts.
- */
-static struct dentry *dfs_rootdir;
+ return provide_user_output(val, u, count, ppos);
+}
/**
- * dbg_debugfs_init - initialize debugfs file-system.
+ * interpret_user_input - interpret user debugfs file input.
+ * @u: user-provided buffer with the input
+ * @count: buffer size
*
- * UBIFS uses debugfs file-system to expose various debugging knobs to
- * user-space. This function creates "ubifs" directory in the debugfs
- * file-system. Returns zero in case of success and a negative error code in
- * case of failure.
+ * This is a helper function which interpret user input to a boolean UBIFS
+ * debugfs file. Returns %0 or %1 in case of success and a negative error code
+ * in case of failure.
*/
-int dbg_debugfs_init(void)
+static int interpret_user_input(const char __user *u, size_t count)
{
- dfs_rootdir = debugfs_create_dir("ubifs", NULL);
- if (IS_ERR(dfs_rootdir)) {
- int err = PTR_ERR(dfs_rootdir);
- ubifs_err("cannot create \"ubifs\" debugfs directory, "
- "error %d\n", err);
- return err;
- }
+ size_t buf_size;
+ char buf[8];
- return 0;
-}
+ buf_size = min_t(size_t, count, (sizeof(buf) - 1));
+ if (copy_from_user(buf, u, buf_size))
+ return -EFAULT;
-/**
- * dbg_debugfs_exit - remove the "ubifs" directory from debugfs file-system.
- */
-void dbg_debugfs_exit(void)
-{
- debugfs_remove(dfs_rootdir);
-}
+ if (buf[0] == '1')
+ return 1;
+ else if (buf[0] == '0')
+ return 0;
-static int open_debugfs_file(struct inode *inode, struct file *file)
-{
- file->private_data = inode->i_private;
- return 0;
+ return -EINVAL;
}
-static ssize_t write_debugfs_file(struct file *file, const char __user *buf,
- size_t count, loff_t *ppos)
+static ssize_t dfs_file_write(struct file *file, const char __user *u,
+ size_t count, loff_t *ppos)
{
struct ubifs_info *c = file->private_data;
struct ubifs_debug_info *d = c->dbg;
+ struct dentry *dent = file->f_path.dentry;
+ int val;
- 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) {
+ /*
+ * TODO: this is racy - the file-system might have already been
+ * unmounted and we'd oops in this case. The plan is to fix it with
+ * help of 'iterate_supers_type()' which we should have in v3.0: when
+ * a debugfs opened, we rember FS's UUID in file->private_data. Then
+ * whenever we access the FS via a debugfs file, we iterate all UBIFS
+ * superblocks and fine the one with the same UUID, and take the
+ * locking right.
+ *
+ * The other way to go suggested by Al Viro is to create a separate
+ * 'ubifs-debug' file-system instead.
+ */
+ if (file->f_path.dentry == d->dfs_dump_lprops) {
+ ubifs_dump_lprops(c);
+ return count;
+ }
+ if (file->f_path.dentry == d->dfs_dump_budg) {
+ ubifs_dump_budg(c, &c->bi);
+ return count;
+ }
+ if (file->f_path.dentry == d->dfs_dump_tnc) {
mutex_lock(&c->tnc_mutex);
- dbg_dump_tnc(c);
+ ubifs_dump_tnc(c);
mutex_unlock(&c->tnc_mutex);
- } else
+ return count;
+ }
+
+ val = interpret_user_input(u, count);
+ if (val < 0)
+ return val;
+
+ if (dent == d->dfs_chk_gen)
+ d->chk_gen = val;
+ else if (dent == d->dfs_chk_index)
+ d->chk_index = val;
+ else if (dent == d->dfs_chk_orph)
+ d->chk_orph = val;
+ else if (dent == d->dfs_chk_lprops)
+ d->chk_lprops = val;
+ else if (dent == d->dfs_chk_fs)
+ d->chk_fs = val;
+ else if (dent == d->dfs_tst_rcvry)
+ d->tst_rcvry = val;
+ else if (dent == d->dfs_ro_error)
+ c->ro_error = !!val;
+ else
return -EINVAL;
- *ppos += count;
return count;
}
static const struct file_operations dfs_fops = {
- .open = open_debugfs_file,
- .write = write_debugfs_file,
+ .open = dfs_file_open,
+ .read = dfs_file_read,
+ .write = dfs_file_write,
.owner = THIS_MODULE,
- .llseek = default_llseek,
+ .llseek = no_llseek,
};
/**
@@ -2798,46 +2809,104 @@ static const struct file_operations dfs_fops = {
*/
int dbg_debugfs_init_fs(struct ubifs_info *c)
{
- int err;
+ int err, n;
const char *fname;
struct dentry *dent;
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);
+ if (!IS_ENABLED(CONFIG_DEBUG_FS))
+ return 0;
+
+ n = snprintf(d->dfs_dir_name, UBIFS_DFS_DIR_LEN + 1, UBIFS_DFS_DIR_NAME,
+ c->vi.ubi_num, c->vi.vol_id);
+ if (n == UBIFS_DFS_DIR_LEN) {
+ /* The array size is too small */
+ fname = UBIFS_DFS_DIR_NAME;
+ dent = ERR_PTR(-EINVAL);
goto out;
}
+ 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;
+ fname = "chk_general";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, d->dfs_dir, c,
+ &dfs_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ d->dfs_chk_gen = dent;
+
+ fname = "chk_index";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, d->dfs_dir, c,
+ &dfs_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ d->dfs_chk_index = dent;
+
+ fname = "chk_orphans";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, d->dfs_dir, c,
+ &dfs_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ d->dfs_chk_orph = dent;
+
+ fname = "chk_lprops";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, d->dfs_dir, c,
+ &dfs_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ d->dfs_chk_lprops = dent;
+
+ fname = "chk_fs";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, d->dfs_dir, c,
+ &dfs_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ d->dfs_chk_fs = dent;
+
+ fname = "tst_recovery";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, d->dfs_dir, c,
+ &dfs_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ d->dfs_tst_rcvry = dent;
+
+ fname = "ro_error";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, d->dfs_dir, c,
+ &dfs_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ d->dfs_ro_error = 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 file or directory, error %d\n",
+ fname, err);
return err;
}
@@ -2847,7 +2916,185 @@ out:
*/
void dbg_debugfs_exit_fs(struct ubifs_info *c)
{
- debugfs_remove_recursive(c->dbg->dfs_dir);
+ if (IS_ENABLED(CONFIG_DEBUG_FS))
+ debugfs_remove_recursive(c->dbg->dfs_dir);
+}
+
+struct ubifs_global_debug_info ubifs_dbg;
+
+static struct dentry *dfs_chk_gen;
+static struct dentry *dfs_chk_index;
+static struct dentry *dfs_chk_orph;
+static struct dentry *dfs_chk_lprops;
+static struct dentry *dfs_chk_fs;
+static struct dentry *dfs_tst_rcvry;
+
+static ssize_t dfs_global_file_read(struct file *file, char __user *u,
+ size_t count, loff_t *ppos)
+{
+ struct dentry *dent = file->f_path.dentry;
+ int val;
+
+ if (dent == dfs_chk_gen)
+ val = ubifs_dbg.chk_gen;
+ else if (dent == dfs_chk_index)
+ val = ubifs_dbg.chk_index;
+ else if (dent == dfs_chk_orph)
+ val = ubifs_dbg.chk_orph;
+ else if (dent == dfs_chk_lprops)
+ val = ubifs_dbg.chk_lprops;
+ else if (dent == dfs_chk_fs)
+ val = ubifs_dbg.chk_fs;
+ else if (dent == dfs_tst_rcvry)
+ val = ubifs_dbg.tst_rcvry;
+ else
+ return -EINVAL;
+
+ return provide_user_output(val, u, count, ppos);
}
-#endif /* CONFIG_UBIFS_FS_DEBUG */
+static ssize_t dfs_global_file_write(struct file *file, const char __user *u,
+ size_t count, loff_t *ppos)
+{
+ struct dentry *dent = file->f_path.dentry;
+ int val;
+
+ val = interpret_user_input(u, count);
+ if (val < 0)
+ return val;
+
+ if (dent == dfs_chk_gen)
+ ubifs_dbg.chk_gen = val;
+ else if (dent == dfs_chk_index)
+ ubifs_dbg.chk_index = val;
+ else if (dent == dfs_chk_orph)
+ ubifs_dbg.chk_orph = val;
+ else if (dent == dfs_chk_lprops)
+ ubifs_dbg.chk_lprops = val;
+ else if (dent == dfs_chk_fs)
+ ubifs_dbg.chk_fs = val;
+ else if (dent == dfs_tst_rcvry)
+ ubifs_dbg.tst_rcvry = val;
+ else
+ return -EINVAL;
+
+ return count;
+}
+
+static const struct file_operations dfs_global_fops = {
+ .read = dfs_global_file_read,
+ .write = dfs_global_file_write,
+ .owner = THIS_MODULE,
+ .llseek = no_llseek,
+};
+
+/**
+ * dbg_debugfs_init - initialize debugfs file-system.
+ *
+ * UBIFS uses debugfs file-system to expose various debugging knobs to
+ * user-space. This function creates "ubifs" directory in the debugfs
+ * file-system. Returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+int dbg_debugfs_init(void)
+{
+ int err;
+ const char *fname;
+ struct dentry *dent;
+
+ if (!IS_ENABLED(CONFIG_DEBUG_FS))
+ return 0;
+
+ fname = "ubifs";
+ dent = debugfs_create_dir(fname, NULL);
+ if (IS_ERR_OR_NULL(dent))
+ goto out;
+ dfs_rootdir = dent;
+
+ fname = "chk_general";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, dfs_rootdir, NULL,
+ &dfs_global_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ dfs_chk_gen = dent;
+
+ fname = "chk_index";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, dfs_rootdir, NULL,
+ &dfs_global_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ dfs_chk_index = dent;
+
+ fname = "chk_orphans";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, dfs_rootdir, NULL,
+ &dfs_global_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ dfs_chk_orph = dent;
+
+ fname = "chk_lprops";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, dfs_rootdir, NULL,
+ &dfs_global_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ dfs_chk_lprops = dent;
+
+ fname = "chk_fs";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, dfs_rootdir, NULL,
+ &dfs_global_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ dfs_chk_fs = dent;
+
+ fname = "tst_recovery";
+ dent = debugfs_create_file(fname, S_IRUSR | S_IWUSR, dfs_rootdir, NULL,
+ &dfs_global_fops);
+ if (IS_ERR_OR_NULL(dent))
+ goto out_remove;
+ dfs_tst_rcvry = dent;
+
+ return 0;
+
+out_remove:
+ debugfs_remove_recursive(dfs_rootdir);
+out:
+ err = dent ? PTR_ERR(dent) : -ENODEV;
+ ubifs_err("cannot create \"%s\" debugfs file or directory, error %d\n",
+ fname, err);
+ return err;
+}
+
+/**
+ * dbg_debugfs_exit - remove the "ubifs" directory from debugfs file-system.
+ */
+void dbg_debugfs_exit(void)
+{
+ if (IS_ENABLED(CONFIG_DEBUG_FS))
+ debugfs_remove_recursive(dfs_rootdir);
+}
+
+/**
+ * ubifs_debugging_init - initialize UBIFS debugging.
+ * @c: UBIFS file-system description object
+ *
+ * This function initializes debugging-related data for the file system.
+ * Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+int ubifs_debugging_init(struct ubifs_info *c)
+{
+ c->dbg = kzalloc(sizeof(struct ubifs_debug_info), GFP_KERNEL);
+ if (!c->dbg)
+ return -ENOMEM;
+
+ return 0;
+}
+
+/**
+ * ubifs_debugging_exit - free debugging data.
+ * @c: UBIFS file-system description object
+ */
+void ubifs_debugging_exit(struct ubifs_info *c)
+{
+ kfree(c->dbg);
+}
diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h
index 69ebe472915..e03d5179769 100644
--- a/fs/ubifs/debug.h
+++ b/fs/ubifs/debug.h
@@ -23,19 +23,31 @@
#ifndef __UBIFS_DEBUG_H__
#define __UBIFS_DEBUG_H__
-#ifdef CONFIG_UBIFS_FS_DEBUG
+/* 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);
+
+/*
+ * The UBIFS debugfs directory name pattern and maximum name length (3 for "ubi"
+ * + 1 for "_" and plus 2x2 for 2 UBI numbers and 1 for the trailing zero byte.
+ */
+#define UBIFS_DFS_DIR_NAME "ubi%d_%d"
+#define UBIFS_DFS_DIR_LEN (3 + 1 + 2*2 + 1)
/**
* 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()'
- * @failure_mode: failure mode for recovery testing
- * @fail_delay: 0=>don't delay, 1=>delay a time, 2=>delay a number of calls
- * @fail_timeout: time in jiffies when delay of failure mode expires
- * @fail_cnt: current number of calls to failure mode I/O functions
- * @fail_cnt_max: number of calls by which to delay failure mode
+ *
+ * @pc_happened: non-zero if an emulated power cut happened
+ * @pc_delay: 0=>don't delay, 1=>delay a time, 2=>delay a number of calls
+ * @pc_timeout: time in jiffies when delay of failure mode expires
+ * @pc_cnt: current number of calls to failure mode I/O functions
+ * @pc_cnt_max: number of calls by which to delay failure mode
+ *
* @chk_lpt_sz: used by LPT tree size checker
* @chk_lpt_sz2: used by LPT tree size checker
* @chk_lpt_wastage: used by LPT tree size checker
@@ -45,24 +57,44 @@
* @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
+ *
+ * @chk_gen: if general extra checks are enabled
+ * @chk_index: if index xtra checks are enabled
+ * @chk_orph: if orphans extra checks are enabled
+ * @chk_lprops: if lprops extra checks are enabled
+ * @chk_fs: if UBIFS contents extra checks are enabled
+ * @tst_rcvry: if UBIFS recovery testing mode enabled
*
- * 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
+ * @dfs_chk_gen: debugfs knob to enable UBIFS general extra checks
+ * @dfs_chk_index: debugfs knob to enable UBIFS index extra checks
+ * @dfs_chk_orph: debugfs knob to enable UBIFS orphans extra checks
+ * @dfs_chk_lprops: debugfs knob to enable UBIFS LEP properties extra checks
+ * @dfs_chk_fs: debugfs knob to enable UBIFS contents extra checks
+ * @dfs_tst_rcvry: debugfs knob to enable UBIFS recovery testing
+ * @dfs_ro_error: debugfs knob to switch UBIFS to R/O mode (different to
+ * re-mounting to R/O mode because it does not flush any buffers
+ * and UBIFS just starts returning -EROFS on all write
+ * operations)
*/
struct ubifs_debug_info {
- void *buf;
struct ubifs_zbranch old_zroot;
int old_zroot_level;
unsigned long long old_zroot_sqnum;
- int failure_mode;
- int fail_delay;
- unsigned long fail_timeout;
- unsigned int fail_cnt;
- unsigned int fail_cnt_max;
+
+ int pc_happened;
+ int pc_delay;
+ unsigned long pc_timeout;
+ unsigned int pc_cnt;
+ unsigned int pc_cnt_max;
+
long long chk_lpt_sz;
long long chk_lpt_sz2;
long long chk_lpt_wastage;
@@ -72,198 +104,141 @@ 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;
+
+ unsigned int chk_gen:1;
+ unsigned int chk_index:1;
+ unsigned int chk_orph:1;
+ unsigned int chk_lprops:1;
+ unsigned int chk_fs:1;
+ unsigned int tst_rcvry:1;
- char dfs_dir_name[100];
+ char dfs_dir_name[UBIFS_DFS_DIR_LEN + 1];
struct dentry *dfs_dir;
struct dentry *dfs_dump_lprops;
struct dentry *dfs_dump_budg;
struct dentry *dfs_dump_tnc;
+ struct dentry *dfs_chk_gen;
+ struct dentry *dfs_chk_index;
+ struct dentry *dfs_chk_orph;
+ struct dentry *dfs_chk_lprops;
+ struct dentry *dfs_chk_fs;
+ struct dentry *dfs_tst_rcvry;
+ struct dentry *dfs_ro_error;
+};
+
+/**
+ * ubifs_global_debug_info - global (not per-FS) UBIFS debugging information.
+ *
+ * @chk_gen: if general extra checks are enabled
+ * @chk_index: if index xtra checks are enabled
+ * @chk_orph: if orphans extra checks are enabled
+ * @chk_lprops: if lprops extra checks are enabled
+ * @chk_fs: if UBIFS contents extra checks are enabled
+ * @tst_rcvry: if UBIFS recovery testing mode enabled
+ */
+struct ubifs_global_debug_info {
+ unsigned int chk_gen:1;
+ unsigned int chk_index:1;
+ unsigned int chk_orph:1;
+ unsigned int chk_lprops:1;
+ unsigned int chk_fs:1;
+ unsigned int tst_rcvry:1;
};
#define ubifs_assert(expr) do { \
if (unlikely(!(expr))) { \
- printk(KERN_CRIT "UBIFS assert failed in %s at %u (pid %d)\n", \
+ pr_crit("UBIFS assert failed in %s at %u (pid %d)\n", \
__func__, __LINE__, current->pid); \
- dbg_dump_stack(); \
+ dump_stack(); \
} \
} while (0)
#define ubifs_assert_cmt_locked(c) do { \
if (unlikely(down_write_trylock(&(c)->commit_sem))) { \
up_write(&(c)->commit_sem); \
- printk(KERN_CRIT "commit lock is not locked!\n"); \
+ pr_crit("commit lock is not locked!\n"); \
ubifs_assert(0); \
} \
} 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 ubifs_dbg_msg(type, fmt, ...) \
+ pr_debug("UBIFS DBG " type " (pid %d): " fmt "\n", current->pid, \
+ ##__VA_ARGS__)
-#define dbg_do_msg(typ, fmt, ...) do { \
- if (ubifs_msg_flags & typ) \
- dbg_msg(fmt, ##__VA_ARGS__); \
+#define DBG_KEY_BUF_LEN 48
+#define ubifs_dbg_msg_key(type, key, fmt, ...) do { \
+ char __tmp_key_buf[DBG_KEY_BUF_LEN]; \
+ pr_debug("UBIFS DBG " type " (pid %d): " fmt "%s\n", current->pid, \
+ ##__VA_ARGS__, \
+ dbg_snprintf_key(c, key, __tmp_key_buf, DBG_KEY_BUF_LEN)); \
} while (0)
-#define dbg_err(fmt, ...) do { \
- spin_lock(&dbg_lock); \
- ubifs_err(fmt, ##__VA_ARGS__); \
- spin_unlock(&dbg_lock); \
-} while (0)
-
-const char *dbg_key_str0(const struct ubifs_info *c,
- const union ubifs_key *key);
-const char *dbg_key_str1(const struct ubifs_info *c,
- const union ubifs_key *key);
-
-/*
- * DBGKEY macros require @dbg_lock to be held, which it is in the dbg message
- * macros.
- */
-#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 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__)
+#define dbg_jnlk(key, fmt, ...) \
+ ubifs_dbg_msg_key("jnl", key, 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__)
+#define dbg_tnck(key, fmt, ...) \
+ ubifs_dbg_msg_key("tnc", key, 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__)
+#define dbg_mntk(key, fmt, ...) \
+ ubifs_dbg_msg_key("mnt", key, 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__)
-
-/*
- * 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).
- *
- * UBIFS_CHK_GEN: general checks
- * UBIFS_CHK_TNC: check TNC
- * UBIFS_CHK_IDX_SZ: check index size
- * UBIFS_CHK_ORPH: check orphans
- * UBIFS_CHK_OLD_IDX: check the old index
- * UBIFS_CHK_LPROPS: check lprops
- * UBIFS_CHK_FS: check the file-system
- */
-enum {
- UBIFS_CHK_GEN = 0x1,
- UBIFS_CHK_TNC = 0x2,
- UBIFS_CHK_IDX_SZ = 0x4,
- UBIFS_CHK_ORPH = 0x8,
- UBIFS_CHK_OLD_IDX = 0x10,
- UBIFS_CHK_LPROPS = 0x20,
- UBIFS_CHK_FS = 0x40,
-};
-
-/*
- * Special testing flags (must match tst_names in debug.c).
- *
- * 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
+#define dbg_rcvry(fmt, ...) ubifs_dbg_msg("rcvry", fmt, ##__VA_ARGS__)
-extern spinlock_t dbg_lock;
+extern struct ubifs_global_debug_info ubifs_dbg;
-extern unsigned int ubifs_msg_flags;
-extern unsigned int ubifs_chk_flags;
-extern unsigned int ubifs_tst_flags;
+static inline int dbg_is_chk_gen(const struct ubifs_info *c)
+{
+ return !!(ubifs_dbg.chk_gen || c->dbg->chk_gen);
+}
+static inline int dbg_is_chk_index(const struct ubifs_info *c)
+{
+ return !!(ubifs_dbg.chk_index || c->dbg->chk_index);
+}
+static inline int dbg_is_chk_orph(const struct ubifs_info *c)
+{
+ return !!(ubifs_dbg.chk_orph || c->dbg->chk_orph);
+}
+static inline int dbg_is_chk_lprops(const struct ubifs_info *c)
+{
+ return !!(ubifs_dbg.chk_lprops || c->dbg->chk_lprops);
+}
+static inline int dbg_is_chk_fs(const struct ubifs_info *c)
+{
+ return !!(ubifs_dbg.chk_fs || c->dbg->chk_fs);
+}
+static inline int dbg_is_tst_rcvry(const struct ubifs_info *c)
+{
+ return !!(ubifs_dbg.tst_rcvry || c->dbg->tst_rcvry);
+}
+static inline int dbg_is_power_cut(const struct ubifs_info *c)
+{
+ return !!c->dbg->pc_happened;
+}
int ubifs_debugging_init(struct ubifs_info *c);
void ubifs_debugging_exit(struct ubifs_info *c);
@@ -274,31 +249,30 @@ const char *dbg_cstate(int cmt_state);
const char *dbg_jhead(int jhead);
const char *dbg_get_key_dump(const struct ubifs_info *c,
const union ubifs_key *key);
-void dbg_dump_inode(const struct ubifs_info *c, const struct inode *inode);
-void dbg_dump_node(const struct ubifs_info *c, const void *node);
-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_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);
-void dbg_dump_leb(const struct ubifs_info *c, int lnum);
-void dbg_dump_znode(const struct ubifs_info *c,
- const struct ubifs_znode *znode);
-void dbg_dump_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat);
-void dbg_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
- struct ubifs_nnode *parent, int iip);
-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);
+const char *dbg_snprintf_key(const struct ubifs_info *c,
+ const union ubifs_key *key, char *buffer, int len);
+void ubifs_dump_inode(struct ubifs_info *c, const struct inode *inode);
+void ubifs_dump_node(const struct ubifs_info *c, const void *node);
+void ubifs_dump_budget_req(const struct ubifs_budget_req *req);
+void ubifs_dump_lstats(const struct ubifs_lp_stats *lst);
+void ubifs_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi);
+void ubifs_dump_lprop(const struct ubifs_info *c,
+ const struct ubifs_lprops *lp);
+void ubifs_dump_lprops(struct ubifs_info *c);
+void ubifs_dump_lpt_info(struct ubifs_info *c);
+void ubifs_dump_leb(const struct ubifs_info *c, int lnum);
+void ubifs_dump_sleb(const struct ubifs_info *c,
+ const struct ubifs_scan_leb *sleb, int offs);
+void ubifs_dump_znode(const struct ubifs_info *c,
+ const struct ubifs_znode *znode);
+void ubifs_dump_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap,
+ int cat);
+void ubifs_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
+ struct ubifs_nnode *parent, int iip);
+void ubifs_dump_tnc(struct ubifs_info *c);
+void ubifs_dump_index(struct ubifs_info *c);
+void ubifs_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);
@@ -312,14 +286,13 @@ int dbg_check_cats(struct ubifs_info *c);
int dbg_check_ltab(struct ubifs_info *c);
int dbg_chk_lpt_free_spc(struct ubifs_info *c);
int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len);
-int dbg_check_synced_i_size(struct inode *inode);
-int dbg_check_dir_size(struct ubifs_info *c, const struct inode *dir);
+int dbg_check_synced_i_size(const struct ubifs_info *c, struct inode *inode);
+int dbg_check_dir(struct ubifs_info *c, const struct inode *dir);
int dbg_check_tnc(struct ubifs_info *c, int extra);
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,
@@ -327,57 +300,11 @@ int dbg_check_inode_size(struct ubifs_info *c, const struct inode *inode,
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)
-
-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
-#define ubi_leb_erase dbg_leb_erase
-#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,
- int len, int check);
-int dbg_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
- int offset, int len, int dtype);
-int dbg_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf,
- int len, int dtype);
-int dbg_leb_erase(struct ubi_volume_desc *desc, int lnum);
-int dbg_leb_unmap(struct ubi_volume_desc *desc, int lnum);
-int dbg_is_mapped(struct ubi_volume_desc *desc, int lnum);
-int dbg_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype);
-
-static inline int dbg_read(struct ubi_volume_desc *desc, int lnum, char *buf,
- int offset, int len)
-{
- return dbg_leb_read(desc, lnum, buf, offset, len, 0);
-}
-
-static inline int dbg_write(struct ubi_volume_desc *desc, int lnum,
- const void *buf, int offset, int len)
-{
- return dbg_leb_write(desc, lnum, buf, offset, len, UBI_UNKNOWN);
-}
-
-static inline int dbg_change(struct ubi_volume_desc *desc, int lnum,
- const void *buf, int len)
-{
- return dbg_leb_change(desc, lnum, buf, len, UBI_UNKNOWN);
-}
+int dbg_leb_write(struct ubifs_info *c, int lnum, const void *buf, int offs,
+ int len);
+int dbg_leb_change(struct ubifs_info *c, int lnum, const void *buf, int len);
+int dbg_leb_unmap(struct ubifs_info *c, int lnum);
+int dbg_leb_map(struct ubifs_info *c, int lnum);
/* Debugfs-related stuff */
int dbg_debugfs_init(void);
@@ -385,98 +312,4 @@ void dbg_debugfs_exit(void);
int dbg_debugfs_init_fs(struct ubifs_info *c);
void dbg_debugfs_exit_fs(struct ubifs_info *c);
-#else /* !CONFIG_UBIFS_FS_DEBUG */
-
-/* Use "if (0)" to make compiler check arguments even if debugging is off */
-#define ubifs_assert(expr) do { \
- if (0 && (expr)) \
- printk(KERN_CRIT "UBIFS assert failed in %s at %u (pid %d)\n", \
- __func__, __LINE__, current->pid); \
-} while (0)
-
-#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__); \
-} 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 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_check_data_nodes_order(c, head) 0
-#define dbg_check_nondata_nodes_order(c, head) 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
-
-#endif /* !CONFIG_UBIFS_FS_DEBUG */
#endif /* !__UBIFS_DEBUG_H__ */
diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c
index 14f64b689d7..ea41649e4ca 100644
--- a/fs/ubifs/dir.c
+++ b/fs/ubifs/dir.c
@@ -56,7 +56,7 @@
*
* This function returns the inherited flags.
*/
-static int inherit_flags(const struct inode *dir, int mode)
+static int inherit_flags(const struct inode *dir, umode_t mode)
{
int flags;
const struct ubifs_inode *ui = ubifs_inode(dir);
@@ -86,7 +86,7 @@ static int inherit_flags(const struct inode *dir, int mode)
* case of failure.
*/
struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir,
- int mode)
+ umode_t mode)
{
struct inode *inode;
struct ubifs_inode *ui;
@@ -102,7 +102,7 @@ struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir,
* UBIFS has to fully control "clean <-> dirty" transitions of inodes
* to make budgeting work.
*/
- inode->i_flags |= (S_NOCMTIME);
+ inode->i_flags |= S_NOCMTIME;
inode_init_owner(inode, dir, mode);
inode->i_mtime = inode->i_atime = inode->i_ctime =
@@ -170,11 +170,11 @@ struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir,
return inode;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
-
-static int dbg_check_name(struct ubifs_dent_node *dent, struct qstr *nm)
+static int dbg_check_name(const struct ubifs_info *c,
+ const struct ubifs_dent_node *dent,
+ const struct qstr *nm)
{
- if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+ if (!dbg_is_chk_gen(c))
return 0;
if (le16_to_cpu(dent->nlen) != nm->len)
return -EINVAL;
@@ -183,14 +183,8 @@ static int dbg_check_name(struct ubifs_dent_node *dent, struct qstr *nm)
return 0;
}
-#else
-
-#define dbg_check_name(dent, nm) 0
-
-#endif
-
static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
- struct nameidata *nd)
+ unsigned int flags)
{
int err;
union ubifs_key key;
@@ -198,8 +192,7 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
struct ubifs_dent_node *dent;
struct ubifs_info *c = dir->i_sb->s_fs_info;
- dbg_gen("'%.*s' in dir ino %lu",
- dentry->d_name.len, dentry->d_name.name, dir->i_ino);
+ dbg_gen("'%pd' in dir ino %lu", dentry, dir->i_ino);
if (dentry->d_name.len > UBIFS_MAX_NLEN)
return ERR_PTR(-ENAMETOOLONG);
@@ -219,7 +212,7 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
goto out;
}
- if (dbg_check_name(dent, &dentry->d_name)) {
+ if (dbg_check_name(c, dent, &dentry->d_name)) {
err = -EINVAL;
goto out;
}
@@ -231,8 +224,8 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
* checking.
*/
err = PTR_ERR(inode);
- ubifs_err("dead directory entry '%.*s', error %d",
- dentry->d_name.len, dentry->d_name.name, err);
+ ubifs_err("dead directory entry '%pd', error %d",
+ dentry, err);
ubifs_ro_mode(c, err);
goto out;
}
@@ -251,8 +244,8 @@ out:
return ERR_PTR(err);
}
-static int ubifs_create(struct inode *dir, struct dentry *dentry, int mode,
- struct nameidata *nd)
+static int ubifs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
+ bool excl)
{
struct inode *inode;
struct ubifs_info *c = dir->i_sb->s_fs_info;
@@ -266,8 +259,8 @@ static int ubifs_create(struct inode *dir, struct dentry *dentry, int mode,
* parent directory inode.
*/
- dbg_gen("dent '%.*s', mode %#x in dir ino %lu",
- dentry->d_name.len, dentry->d_name.name, mode, dir->i_ino);
+ dbg_gen("dent '%pd', mode %#hx in dir ino %lu",
+ dentry, mode, dir->i_ino);
err = ubifs_budget_space(c, &req);
if (err)
@@ -352,38 +345,46 @@ static unsigned int vfs_dent_type(uint8_t type)
* This means that UBIFS cannot support NFS which requires full
* 'seekdir()'/'telldir()' support.
*/
-static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir)
+static int ubifs_readdir(struct file *file, struct dir_context *ctx)
{
- int err, over = 0;
+ int err;
struct qstr nm;
union ubifs_key key;
struct ubifs_dent_node *dent;
- struct inode *dir = file->f_path.dentry->d_inode;
+ struct inode *dir = file_inode(file);
struct ubifs_info *c = dir->i_sb->s_fs_info;
- dbg_gen("dir ino %lu, f_pos %#llx", dir->i_ino, file->f_pos);
+ dbg_gen("dir ino %lu, f_pos %#llx", dir->i_ino, ctx->pos);
- if (file->f_pos > UBIFS_S_KEY_HASH_MASK || file->f_pos == 2)
+ if (ctx->pos > UBIFS_S_KEY_HASH_MASK || ctx->pos == 2)
/*
* The directory was seek'ed to a senseless position or there
* are no more entries.
*/
return 0;
- /* File positions 0 and 1 correspond to "." and ".." */
- if (file->f_pos == 0) {
- ubifs_assert(!file->private_data);
- over = filldir(dirent, ".", 1, 0, dir->i_ino, DT_DIR);
- if (over)
- return 0;
- file->f_pos = 1;
+ if (file->f_version == 0) {
+ /*
+ * The file was seek'ed, which means that @file->private_data
+ * is now invalid. This may also be just the first
+ * 'ubifs_readdir()' invocation, in which case
+ * @file->private_data is NULL, and the below code is
+ * basically a no-op.
+ */
+ kfree(file->private_data);
+ file->private_data = NULL;
}
- if (file->f_pos == 1) {
+ /*
+ * 'generic_file_llseek()' unconditionally sets @file->f_version to
+ * zero, and we use this for detecting whether the file was seek'ed.
+ */
+ file->f_version = 1;
+
+ /* File positions 0 and 1 correspond to "." and ".." */
+ if (ctx->pos < 2) {
ubifs_assert(!file->private_data);
- over = filldir(dirent, "..", 2, 1,
- parent_ino(file->f_path.dentry), DT_DIR);
- if (over)
+ if (!dir_emit_dots(file, ctx))
return 0;
/* Find the first entry in TNC and save it */
@@ -395,7 +396,7 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir)
goto out;
}
- file->f_pos = key_hash_flash(c, &dent->key);
+ ctx->pos = key_hash_flash(c, &dent->key);
file->private_data = dent;
}
@@ -403,17 +404,16 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir)
if (!dent) {
/*
* The directory was seek'ed to and is now readdir'ed.
- * Find the entry corresponding to @file->f_pos or the
- * closest one.
+ * Find the entry corresponding to @ctx->pos or the closest one.
*/
- dent_key_init_hash(c, &key, dir->i_ino, file->f_pos);
+ dent_key_init_hash(c, &key, dir->i_ino, ctx->pos);
nm.name = NULL;
dent = ubifs_tnc_next_ent(c, &key, &nm);
if (IS_ERR(dent)) {
err = PTR_ERR(dent);
goto out;
}
- file->f_pos = key_hash_flash(c, &dent->key);
+ ctx->pos = key_hash_flash(c, &dent->key);
file->private_data = dent;
}
@@ -425,10 +425,9 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir)
ubifs_inode(dir)->creat_sqnum);
nm.len = le16_to_cpu(dent->nlen);
- over = filldir(dirent, dent->name, nm.len, file->f_pos,
+ if (!dir_emit(ctx, dent->name, nm.len,
le64_to_cpu(dent->inum),
- vfs_dent_type(dent->type));
- if (over)
+ vfs_dent_type(dent->type)))
return 0;
/* Switch to the next entry */
@@ -441,7 +440,7 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir)
}
kfree(file->private_data);
- file->f_pos = key_hash_flash(c, &dent->key);
+ ctx->pos = key_hash_flash(c, &dent->key);
file->private_data = dent;
cond_resched();
}
@@ -454,18 +453,11 @@ out:
kfree(file->private_data);
file->private_data = NULL;
- file->f_pos = 2;
+ /* 2 is a special value indicating that there are no more direntries */
+ ctx->pos = 2;
return 0;
}
-/* If a directory is seeked, we have to free saved readdir() state */
-static loff_t ubifs_dir_llseek(struct file *file, loff_t offset, int origin)
-{
- kfree(file->private_data);
- file->private_data = NULL;
- return generic_file_llseek(file, offset, origin);
-}
-
/* Free saved readdir() state when the directory is closed */
static int ubifs_dir_release(struct inode *dir, struct file *file)
{
@@ -516,31 +508,13 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
* changing the parent inode.
*/
- dbg_gen("dent '%.*s' to ino %lu (nlink %d) in dir ino %lu",
- dentry->d_name.len, dentry->d_name.name, inode->i_ino,
+ dbg_gen("dent '%pd' to ino %lu (nlink %d) in dir ino %lu",
+ dentry, inode->i_ino,
inode->i_nlink, dir->i_ino);
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);
+ err = dbg_check_synced_i_size(c, inode);
if (err)
return err;
@@ -582,6 +556,7 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry)
int sz_change = CALC_DENT_SIZE(dentry->d_name.len);
int err, budgeted = 1;
struct ubifs_budget_req req = { .mod_dent = 1, .dirtied_ino = 2 };
+ unsigned int saved_nlink = inode->i_nlink;
/*
* Budget request settings: deletion direntry, deletion inode (+1 for
@@ -590,12 +565,12 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry)
* deletions.
*/
- dbg_gen("dent '%.*s' from ino %lu (nlink %d) in dir ino %lu",
- dentry->d_name.len, dentry->d_name.name, inode->i_ino,
+ dbg_gen("dent '%pd' from ino %lu (nlink %d) in dir ino %lu",
+ dentry, inode->i_ino,
inode->i_nlink, dir->i_ino);
ubifs_assert(mutex_is_locked(&dir->i_mutex));
ubifs_assert(mutex_is_locked(&inode->i_mutex));
- err = dbg_check_synced_i_size(inode);
+ err = dbg_check_synced_i_size(c, inode);
if (err)
return err;
@@ -621,7 +596,7 @@ 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;
@@ -629,7 +604,7 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry)
out_cancel:
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- inc_nlink(inode);
+ set_nlink(inode, saved_nlink);
unlock_2_inodes(dir, inode);
if (budgeted)
ubifs_release_budget(c, &req);
@@ -680,8 +655,8 @@ static int ubifs_rmdir(struct inode *dir, struct dentry *dentry)
* because we have extra space reserved for deletions.
*/
- dbg_gen("directory '%.*s', ino %lu in dir ino %lu", dentry->d_name.len,
- dentry->d_name.name, inode->i_ino, dir->i_ino);
+ dbg_gen("directory '%pd', ino %lu in dir ino %lu", dentry,
+ inode->i_ino, dir->i_ino);
ubifs_assert(mutex_is_locked(&dir->i_mutex));
ubifs_assert(mutex_is_locked(&inode->i_mutex));
err = check_dir_empty(c, dentry->d_inode);
@@ -711,7 +686,7 @@ 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;
@@ -720,15 +695,14 @@ out_cancel:
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
inc_nlink(dir);
- inc_nlink(inode);
- inc_nlink(inode);
+ set_nlink(inode, 2);
unlock_2_inodes(dir, inode);
if (budgeted)
ubifs_release_budget(c, &req);
return err;
}
-static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
{
struct inode *inode;
struct ubifs_inode *dir_ui = ubifs_inode(dir);
@@ -741,8 +715,8 @@ static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
* directory inode.
*/
- dbg_gen("dent '%.*s', mode %#x in dir ino %lu",
- dentry->d_name.len, dentry->d_name.name, mode, dir->i_ino);
+ dbg_gen("dent '%pd', mode %#hx in dir ino %lu",
+ dentry, mode, dir->i_ino);
err = ubifs_budget_space(c, &req);
if (err)
@@ -785,7 +759,7 @@ out_budg:
}
static int ubifs_mknod(struct inode *dir, struct dentry *dentry,
- int mode, dev_t rdev)
+ umode_t mode, dev_t rdev)
{
struct inode *inode;
struct ubifs_inode *ui;
@@ -803,8 +777,7 @@ static int ubifs_mknod(struct inode *dir, struct dentry *dentry,
* directory inode.
*/
- dbg_gen("dent '%.*s' in dir ino %lu",
- dentry->d_name.len, dentry->d_name.name, dir->i_ino);
+ dbg_gen("dent '%pd' in dir ino %lu", dentry, dir->i_ino);
if (!new_valid_dev(rdev))
return -EINVAL;
@@ -878,8 +851,8 @@ static int ubifs_symlink(struct inode *dir, struct dentry *dentry,
* directory inode.
*/
- dbg_gen("dent '%.*s', target '%s' in dir ino %lu", dentry->d_name.len,
- dentry->d_name.name, symname, dir->i_ino);
+ dbg_gen("dent '%pd', target '%s' in dir ino %lu", dentry,
+ symname, dir->i_ino);
if (len > UBIFS_MAX_INO_DATA)
return -ENAMETOOLONG;
@@ -993,6 +966,7 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry,
struct ubifs_budget_req ino_req = { .dirtied_ino = 1,
.dirtied_ino_d = ALIGN(old_inode_ui->data_len, 8) };
struct timespec time;
+ unsigned int uninitialized_var(saved_nlink);
/*
* Budget request settings: deletion direntry, new direntry, removing
@@ -1003,10 +977,9 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry,
* separately.
*/
- dbg_gen("dent '%.*s' ino %lu in dir ino %lu to dent '%.*s' in "
- "dir ino %lu", old_dentry->d_name.len, old_dentry->d_name.name,
- old_inode->i_ino, old_dir->i_ino, new_dentry->d_name.len,
- new_dentry->d_name.name, new_dir->i_ino);
+ dbg_gen("dent '%pd' ino %lu in dir ino %lu to dent '%pd' in dir ino %lu",
+ old_dentry, old_inode->i_ino, old_dir->i_ino,
+ new_dentry, new_dir->i_ino);
ubifs_assert(mutex_is_locked(&old_dir->i_mutex));
ubifs_assert(mutex_is_locked(&new_dir->i_mutex));
if (unlink)
@@ -1075,13 +1048,14 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry,
if (unlink) {
/*
* Directories cannot have hard-links, so if this is a
- * directory, decrement its @i_nlink twice because an empty
- * directory has @i_nlink 2.
+ * directory, just clear @i_nlink.
*/
+ saved_nlink = new_inode->i_nlink;
if (is_dir)
+ clear_nlink(new_inode);
+ else
drop_nlink(new_inode);
new_inode->i_ctime = time;
- drop_nlink(new_inode);
} else {
new_dir->i_size += new_sz;
ubifs_inode(new_dir)->ui_size = new_dir->i_size;
@@ -1118,9 +1092,7 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry,
out_cancel:
if (unlink) {
- if (is_dir)
- inc_nlink(new_inode);
- inc_nlink(new_inode);
+ set_nlink(new_inode, saved_nlink);
} else {
new_dir->i_size -= new_sz;
ubifs_inode(new_dir)->ui_size = new_dir->i_size;
@@ -1151,16 +1123,7 @@ int ubifs_getattr(struct vfsmount *mnt, struct dentry *dentry,
struct ubifs_inode *ui = ubifs_inode(inode);
mutex_lock(&ui->ui_mutex);
- stat->dev = inode->i_sb->s_dev;
- stat->ino = inode->i_ino;
- stat->mode = inode->i_mode;
- stat->nlink = inode->i_nlink;
- stat->uid = inode->i_uid;
- stat->gid = inode->i_gid;
- stat->rdev = inode->i_rdev;
- stat->atime = inode->i_atime;
- stat->mtime = inode->i_mtime;
- stat->ctime = inode->i_ctime;
+ generic_fillattr(inode, stat);
stat->blksize = UBIFS_BLOCK_SIZE;
stat->size = ui->ui_size;
@@ -1203,19 +1166,17 @@ const struct inode_operations ubifs_dir_inode_operations = {
.rename = ubifs_rename,
.setattr = ubifs_setattr,
.getattr = ubifs_getattr,
-#ifdef CONFIG_UBIFS_FS_XATTR
.setxattr = ubifs_setxattr,
.getxattr = ubifs_getxattr,
.listxattr = ubifs_listxattr,
.removexattr = ubifs_removexattr,
-#endif
};
const struct file_operations ubifs_dir_operations = {
- .llseek = ubifs_dir_llseek,
+ .llseek = generic_file_llseek,
.release = ubifs_dir_release,
.read = generic_read_dir,
- .readdir = ubifs_readdir,
+ .iterate = ubifs_readdir,
.fsync = ubifs_fsync,
.unlocked_ioctl = ubifs_ioctl,
#ifdef CONFIG_COMPAT
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index d77db7e3648..b5b593c4527 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -37,11 +37,11 @@
*
* A thing to keep in mind: inode @i_mutex is locked in most VFS operations we
* implement. However, this is not true for 'ubifs_writepage()', which may be
- * called with @i_mutex unlocked. For example, when pdflush is doing background
- * write-back, it calls 'ubifs_writepage()' with unlocked @i_mutex. At "normal"
- * work-paths the @i_mutex is locked in 'ubifs_writepage()', e.g. in the
- * "sys_write -> alloc_pages -> direct reclaim path". So, in 'ubifs_writepage()'
- * we are only guaranteed that the page is locked.
+ * called with @i_mutex unlocked. For example, when flusher thread is doing
+ * background write-back, it calls 'ubifs_writepage()' with unlocked @i_mutex.
+ * At "normal" work-paths the @i_mutex is locked in 'ubifs_writepage()', e.g.
+ * in the "sys_write -> alloc_pages -> direct reclaim path". So, in
+ * 'ubifs_writepage()' we are only guaranteed that the page is locked.
*
* Similarly, @i_mutex is not always locked in 'ubifs_readpage()', e.g., the
* read-ahead path does not lock it ("sys_read -> generic_file_aio_read ->
@@ -50,6 +50,7 @@
*/
#include "ubifs.h"
+#include <linux/aio.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/slab.h>
@@ -97,7 +98,7 @@ static int read_block(struct inode *inode, void *addr, unsigned int block,
dump:
ubifs_err("bad data node (block %u, inode %lu)",
block, inode->i_ino);
- dbg_dump_node(c, dn);
+ ubifs_dump_node(c, dn);
return -EINVAL;
}
@@ -212,7 +213,7 @@ 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);
}
@@ -448,10 +449,12 @@ 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;
@@ -559,6 +562,7 @@ 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
@@ -899,8 +903,9 @@ static int do_writepage(struct page *page, int len)
struct ubifs_info *c = inode->i_sb->s_fs_info;
#ifdef UBIFS_DEBUG
+ struct ubifs_inode *ui = ubifs_inode(inode);
spin_lock(&ui->ui_lock);
- ubifs_assert(page->index <= ui->synced_i_size << PAGE_CACHE_SIZE);
+ ubifs_assert(page->index <= ui->synced_i_size >> PAGE_CACHE_SHIFT);
spin_unlock(&ui->ui_lock);
#endif
@@ -968,11 +973,11 @@ 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
@@ -1039,10 +1044,10 @@ static int ubifs_writepage(struct page *page, struct writeback_control *wbc)
* the page size, the remaining memory is zeroed when mapped, and
* writes to that region are not written out to the file."
*/
- kaddr = kmap_atomic(page, KM_USER0);
+ kaddr = kmap_atomic(page);
memset(kaddr + len, 0, PAGE_CACHE_SIZE - len);
flush_dcache_page(page);
- kunmap_atomic(kaddr, KM_USER0);
+ kunmap_atomic(kaddr);
if (i_size > synced_i_size) {
err = inode->i_sb->s_op->write_inode(inode, NULL);
@@ -1186,7 +1191,7 @@ 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;
@@ -1260,7 +1265,7 @@ int ubifs_setattr(struct dentry *dentry, struct iattr *attr)
if (err)
return err;
- err = dbg_check_synced_i_size(inode);
+ err = dbg_check_synced_i_size(c, inode);
if (err)
return err;
@@ -1273,13 +1278,14 @@ int ubifs_setattr(struct dentry *dentry, struct iattr *attr)
return err;
}
-static void ubifs_invalidatepage(struct page *page, unsigned long offset)
+static void ubifs_invalidatepage(struct page *page, unsigned int offset,
+ unsigned int length)
{
struct inode *inode = page->mapping->host;
struct ubifs_info *c = inode->i_sb->s_fs_info;
ubifs_assert(PagePrivate(page));
- if (offset)
+ if (offset || length < PAGE_CACHE_SIZE)
/* Partial page remains dirty */
return;
@@ -1301,7 +1307,7 @@ static void *ubifs_follow_link(struct dentry *dentry, struct nameidata *nd)
return NULL;
}
-int ubifs_fsync(struct file *file, int datasync)
+int ubifs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
struct inode *inode = file->f_mapping->host;
struct ubifs_info *c = inode->i_sb->s_fs_info;
@@ -1309,14 +1315,23 @@ int ubifs_fsync(struct file *file, int datasync)
dbg_gen("syncing inode %lu", inode->i_ino);
- /*
- * VFS has already synchronized dirty pages for this inode. Synchronize
- * the inode unless this is a 'datasync()' call.
- */
+ 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;
+
+ err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+ if (err)
+ return err;
+ mutex_lock(&inode->i_mutex);
+
+ /* Synchronize the inode unless this is a 'datasync()' call. */
if (!datasync || (inode->i_state & I_DIRTY_DATASYNC)) {
err = inode->i_sb->s_op->write_inode(inode, NULL);
if (err)
- return err;
+ goto out;
}
/*
@@ -1324,10 +1339,9 @@ int ubifs_fsync(struct file *file, int datasync)
* them.
*/
err = ubifs_sync_wbufs_by_inode(c, inode);
- if (err)
- return err;
-
- return 0;
+out:
+ mutex_unlock(&inode->i_mutex);
+ return err;
}
/**
@@ -1350,17 +1364,17 @@ static inline int mctime_update_needed(const struct inode *inode,
/**
* update_ctime - update mtime and ctime of an inode.
- * @c: UBIFS file-system description object
* @inode: inode to update
*
* This function updates mtime and ctime of the inode if it is not equivalent to
* current time. Returns zero in case of success and a negative error code in
* case of failure.
*/
-static int update_mctime(struct ubifs_info *c, struct inode *inode)
+static int update_mctime(struct inode *inode)
{
struct timespec now = ubifs_current_time(inode);
struct ubifs_inode *ui = ubifs_inode(inode);
+ struct ubifs_info *c = inode->i_sb->s_fs_info;
if (mctime_update_needed(inode, &now)) {
int err, release;
@@ -1383,18 +1397,13 @@ static int update_mctime(struct ubifs_info *c, struct inode *inode)
return 0;
}
-static ssize_t ubifs_aio_write(struct kiocb *iocb, const struct iovec *iov,
- unsigned long nr_segs, loff_t pos)
+static ssize_t ubifs_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
- int err;
- struct inode *inode = iocb->ki_filp->f_mapping->host;
- struct ubifs_info *c = inode->i_sb->s_fs_info;
-
- err = update_mctime(c, inode);
+ int err = update_mctime(file_inode(iocb->ki_filp));
if (err)
return err;
- return generic_file_aio_write(iocb, iov, nr_segs, pos);
+ return generic_file_write_iter(iocb, from);
}
static int ubifs_set_page_dirty(struct page *page)
@@ -1426,13 +1435,14 @@ 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;
+ struct inode *inode = file_inode(vma->vm_file);
struct ubifs_info *c = inode->i_sb->s_fs_info;
struct timespec now = ubifs_current_time(inode);
struct ubifs_budget_req req = { .new_page = 1 };
@@ -1474,8 +1484,8 @@ static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vm
err = ubifs_budget_space(c, &req);
if (unlikely(err)) {
if (err == -ENOSPC)
- ubifs_warn("out of space for mmapped file "
- "(inode number %lu)", inode->i_ino);
+ ubifs_warn("out of space for mmapped file (inode number %lu)",
+ inode->i_ino);
return VM_FAULT_SIGBUS;
}
@@ -1510,8 +1520,8 @@ static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vm
ubifs_release_dirty_inode_budget(c, ui);
}
- unlock_page(page);
- return 0;
+ wait_for_stable_page(page);
+ return VM_FAULT_LOCKED;
out_unlock:
unlock_page(page);
@@ -1523,14 +1533,15 @@ out_unlock:
static const struct vm_operations_struct ubifs_file_vm_ops = {
.fault = filemap_fault,
+ .map_pages = filemap_map_pages,
.page_mkwrite = ubifs_vm_page_mkwrite,
+ .remap_pages = generic_file_remap_pages,
};
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;
@@ -1551,12 +1562,10 @@ const struct address_space_operations ubifs_file_address_operations = {
const struct inode_operations ubifs_file_inode_operations = {
.setattr = ubifs_setattr,
.getattr = ubifs_getattr,
-#ifdef CONFIG_UBIFS_FS_XATTR
.setxattr = ubifs_setxattr,
.getxattr = ubifs_getxattr,
.listxattr = ubifs_listxattr,
.removexattr = ubifs_removexattr,
-#endif
};
const struct inode_operations ubifs_symlink_inode_operations = {
@@ -1568,15 +1577,15 @@ const struct inode_operations ubifs_symlink_inode_operations = {
const struct file_operations ubifs_file_operations = {
.llseek = generic_file_llseek,
- .read = do_sync_read,
- .write = do_sync_write,
- .aio_read = generic_file_aio_read,
- .aio_write = ubifs_aio_write,
+ .read = new_sync_read,
+ .write = new_sync_write,
+ .read_iter = generic_file_read_iter,
+ .write_iter = ubifs_write_iter,
.mmap = ubifs_file_mmap,
.fsync = ubifs_fsync,
.unlocked_ioctl = ubifs_ioctl,
.splice_read = generic_file_splice_read,
- .splice_write = generic_file_splice_write,
+ .splice_write = iter_file_splice_write,
#ifdef CONFIG_COMPAT
.compat_ioctl = ubifs_compat_ioctl,
#endif
diff --git a/fs/ubifs/find.c b/fs/ubifs/find.c
index 1d54383d126..2dcf3d473fe 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 -
@@ -681,8 +681,16 @@ int ubifs_find_free_leb_for_idx(struct ubifs_info *c)
if (!lprops) {
lprops = ubifs_fast_find_freeable(c);
if (!lprops) {
- ubifs_assert(c->freeable_cnt == 0);
- if (c->lst.empty_lebs - c->lst.taken_empty_lebs > 0) {
+ /*
+ * The first condition means the following: go scan the
+ * LPT if there are uncategorized lprops, which means
+ * there may be freeable LEBs there (UBIFS does not
+ * store the information about freeable LEBs in the
+ * master node).
+ */
+ if (c->in_a_category_cnt != c->main_lebs ||
+ c->lst.empty_lebs - c->lst.taken_empty_lebs > 0) {
+ ubifs_assert(c->freeable_cnt == 0);
lprops = scan_for_leb_for_idx(c);
if (IS_ERR(lprops)) {
err = PTR_ERR(lprops);
@@ -939,8 +947,8 @@ static int find_dirtiest_idx_leb(struct ubifs_info *c)
}
dbg_find("LEB %d, dirty %d and free %d flags %#x", lp->lnum, lp->dirty,
lp->free, lp->flags);
- ubifs_assert(lp->flags | LPROPS_TAKEN);
- ubifs_assert(lp->flags | LPROPS_INDEX);
+ ubifs_assert(lp->flags & LPROPS_TAKEN);
+ ubifs_assert(lp->flags & LPROPS_INDEX);
return lnum;
}
diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c
index 151f1088282..9718da86ad0 100644
--- a/fs/ubifs/gc.c
+++ b/fs/ubifs/gc.c
@@ -100,12 +100,16 @@ 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;
c->gc_lnum = -1;
- err = ubifs_wbuf_seek_nolock(wbuf, gc_lnum, 0, UBI_LONGTERM);
+ err = ubifs_wbuf_seek_nolock(wbuf, gc_lnum, 0);
return err;
}
@@ -118,7 +122,7 @@ 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;
@@ -161,7 +165,8 @@ 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)
{
ino_t inuma, inumb;
struct ubifs_info *c = priv;
@@ -473,6 +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).
@@ -632,8 +668,7 @@ int ubifs_garbage_collect(struct ubifs_info *c, int anyway)
ubifs_assert(!wbuf->used);
for (i = 0; ; i++) {
- int space_before = c->leb_size - wbuf->offs - wbuf->used;
- int space_after;
+ int space_before, space_after;
cond_resched();
@@ -678,40 +713,9 @@ int ubifs_garbage_collect(struct ubifs_info *c, int anyway)
break;
}
- dbg_gc("found LEB %d: free %d, dirty %d, sum %d "
- "(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;
- }
+ dbg_gc("found LEB %d: free %d, dirty %d, sum %d (min. space %d)",
+ lp.lnum, lp.free, lp.dirty, lp.free + lp.dirty,
+ min_space);
space_before = c->leb_size - wbuf->offs - wbuf->used;
if (wbuf->lnum == -1)
diff --git a/fs/ubifs/io.c b/fs/ubifs/io.c
index d82173182ee..2290d586672 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>
@@ -66,8 +86,124 @@ void ubifs_ro_mode(struct ubifs_info *c, int err)
c->no_chk_data_crc = 0;
c->vfs_sb->s_flags |= MS_RDONLY;
ubifs_warn("switched to read-only mode, error %d", err);
- dbg_dump_stack();
+ dump_stack();
+ }
+}
+
+/*
+ * Below are simple wrappers over UBI I/O functions which include some
+ * additional checks and UBIFS debugging stuff. See corresponding UBI function
+ * for more information.
+ */
+
+int ubifs_leb_read(const struct ubifs_info *c, int lnum, void *buf, int offs,
+ int len, int even_ebadmsg)
+{
+ int err;
+
+ err = ubi_read(c->ubi, lnum, buf, offs, len);
+ /*
+ * In case of %-EBADMSG print the error message only if the
+ * @even_ebadmsg is true.
+ */
+ if (err && (err != -EBADMSG || even_ebadmsg)) {
+ ubifs_err("reading %d bytes from LEB %d:%d failed, error %d",
+ len, lnum, offs, err);
+ dump_stack();
+ }
+ return err;
+}
+
+int ubifs_leb_write(struct ubifs_info *c, int lnum, const void *buf, int offs,
+ int len)
+{
+ int err;
+
+ ubifs_assert(!c->ro_media && !c->ro_mount);
+ if (c->ro_error)
+ return -EROFS;
+ if (!dbg_is_tst_rcvry(c))
+ err = ubi_leb_write(c->ubi, lnum, buf, offs, len);
+ else
+ err = dbg_leb_write(c, lnum, buf, offs, len);
+ if (err) {
+ ubifs_err("writing %d bytes to LEB %d:%d failed, error %d",
+ len, lnum, offs, err);
+ ubifs_ro_mode(c, err);
+ dump_stack();
+ }
+ return err;
+}
+
+int ubifs_leb_change(struct ubifs_info *c, int lnum, const void *buf, int len)
+{
+ int err;
+
+ ubifs_assert(!c->ro_media && !c->ro_mount);
+ if (c->ro_error)
+ return -EROFS;
+ if (!dbg_is_tst_rcvry(c))
+ err = ubi_leb_change(c->ubi, lnum, buf, len);
+ else
+ err = dbg_leb_change(c, lnum, buf, len);
+ if (err) {
+ ubifs_err("changing %d bytes in LEB %d failed, error %d",
+ len, lnum, err);
+ ubifs_ro_mode(c, err);
+ dump_stack();
+ }
+ return err;
+}
+
+int ubifs_leb_unmap(struct ubifs_info *c, int lnum)
+{
+ int err;
+
+ ubifs_assert(!c->ro_media && !c->ro_mount);
+ if (c->ro_error)
+ return -EROFS;
+ if (!dbg_is_tst_rcvry(c))
+ err = ubi_leb_unmap(c->ubi, lnum);
+ else
+ err = dbg_leb_unmap(c, lnum);
+ if (err) {
+ ubifs_err("unmap LEB %d failed, error %d", lnum, err);
+ ubifs_ro_mode(c, err);
+ dump_stack();
+ }
+ return err;
+}
+
+int ubifs_leb_map(struct ubifs_info *c, int lnum)
+{
+ int err;
+
+ ubifs_assert(!c->ro_media && !c->ro_mount);
+ if (c->ro_error)
+ return -EROFS;
+ if (!dbg_is_tst_rcvry(c))
+ err = ubi_leb_map(c->ubi, lnum);
+ else
+ err = dbg_leb_map(c, lnum);
+ if (err) {
+ ubifs_err("mapping LEB %d failed, error %d", lnum, err);
+ ubifs_ro_mode(c, err);
+ dump_stack();
+ }
+ return err;
+}
+
+int ubifs_is_mapped(const struct ubifs_info *c, int lnum)
+{
+ int err;
+
+ err = ubi_is_mapped(c->ubi, lnum);
+ if (err < 0) {
+ ubifs_err("ubi_is_mapped failed for LEB %d, error %d",
+ lnum, err);
+ dump_stack();
}
+ return err;
}
/**
@@ -88,8 +224,12 @@ 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 +271,8 @@ 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);
@@ -153,8 +293,8 @@ out_len:
out:
if (!quiet) {
ubifs_err("bad node at LEB %d:%d", lnum, offs);
- dbg_dump_node(c, buf);
- dbg_dump_stack();
+ ubifs_dump_node(c, buf);
+ dump_stack();
}
return err;
}
@@ -343,11 +483,17 @@ 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)
@@ -357,27 +503,48 @@ 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(!(wbuf->avail & 7));
- ubifs_assert(wbuf->offs + c->min_io_size <= c->leb_size);
+ 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);
- err = ubi_leb_write(c->ubi, wbuf->lnum, wbuf->buf, wbuf->offs,
- c->min_io_size, wbuf->dtype);
- if (err) {
- ubifs_err("cannot write %d bytes to LEB %d:%d",
- c->min_io_size, wbuf->lnum, wbuf->offs);
- dbg_dump_stack();
+ /*
+ * 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 = ubifs_leb_write(c, wbuf->lnum, wbuf->buf, wbuf->offs, sync_len);
+ if (err)
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);
@@ -393,14 +560,12 @@ int ubifs_wbuf_sync_nolock(struct ubifs_wbuf *wbuf)
* @wbuf: write-buffer
* @lnum: logical eraseblock number to seek to
* @offs: logical eraseblock offset to seek to
- * @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)
+int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs)
{
const struct ubifs_info *c = wbuf->c;
@@ -409,21 +574,20 @@ 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;
return 0;
}
@@ -500,8 +664,9 @@ 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
@@ -510,7 +675,7 @@ 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),
@@ -518,9 +683,15 @@ 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;
@@ -542,15 +713,18 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len)
if (aligned_len == wbuf->avail) {
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->dtype);
+ err = ubifs_leb_write(c, wbuf->lnum, wbuf->buf,
+ wbuf->offs, wbuf->size);
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);
@@ -564,39 +738,63 @@ 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 = ubifs_leb_write(c, wbuf->lnum, wbuf->buf, wbuf->offs,
+ wbuf->size);
+ 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->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 = ubifs_leb_write(c, wbuf->lnum, buf, wbuf->offs,
+ wbuf->size);
+ if (err)
+ goto out;
+
+ 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 = ubifs_leb_write(c, wbuf->lnum, buf + written,
+ wbuf->offs, n);
if (err)
goto out;
- offs += n;
+ wbuf->offs += n;
aligned_len -= n;
len -= n;
written += n;
@@ -606,14 +804,17 @@ 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);
@@ -634,9 +835,9 @@ exit:
out:
ubifs_err("cannot write %d bytes to LEB %d:%d, error %d",
len, wbuf->lnum, wbuf->offs, err);
- dbg_dump_node(c, buf);
- dbg_dump_stack();
- dbg_dump_leb(c, wbuf->lnum);
+ ubifs_dump_node(c, buf);
+ dump_stack();
+ ubifs_dump_leb(c, wbuf->lnum);
return err;
}
@@ -647,7 +848,6 @@ out:
* @len: node length
* @lnum: logical eraseblock number
* @offs: offset within the logical eraseblock
- * @dtype: node life-time hint (%UBI_LONGTERM, %UBI_SHORTTERM, %UBI_UNKNOWN)
*
* This function automatically fills node magic number, assigns sequence
* number, and calculates node CRC checksum. The length of the @buf buffer has
@@ -656,7 +856,7 @@ out:
* success and a negative error code in case of failure.
*/
int ubifs_write_node(struct ubifs_info *c, void *buf, int len, int lnum,
- int offs, int dtype)
+ int offs)
{
int err, buf_len = ALIGN(len, c->min_io_size);
@@ -666,18 +866,15 @@ int ubifs_write_node(struct ubifs_info *c, void *buf, int len, int lnum,
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_error)
return -EROFS;
ubifs_prepare_node(c, buf, len, 1);
- err = ubi_leb_write(c->ubi, lnum, buf, offs, buf_len, dtype);
- if (err) {
- ubifs_err("cannot write %d bytes to LEB %d:%d, error %d",
- buf_len, lnum, offs, err);
- dbg_dump_node(c, buf);
- dbg_dump_stack();
- }
+ err = ubifs_leb_write(c, lnum, buf, offs, buf_len);
+ if (err)
+ ubifs_dump_node(c, buf);
return err;
}
@@ -729,13 +926,9 @@ int ubifs_read_node_wbuf(struct ubifs_wbuf *wbuf, void *buf, int type, int len,
if (rlen > 0) {
/* Read everything that goes before write-buffer */
- err = ubi_read(c->ubi, lnum, buf, offs, rlen);
- if (err && err != -EBADMSG) {
- ubifs_err("failed to read node %d from LEB %d:%d, "
- "error %d", type, lnum, offs, err);
- dbg_dump_stack();
+ err = ubifs_leb_read(c, lnum, buf, offs, rlen, 0);
+ if (err && err != -EBADMSG)
return err;
- }
}
if (type != ch->node_type) {
@@ -760,8 +953,8 @@ int ubifs_read_node_wbuf(struct ubifs_wbuf *wbuf, void *buf, int type, int len,
out:
ubifs_err("bad node at LEB %d:%d", lnum, offs);
- dbg_dump_node(c, buf);
- dbg_dump_stack();
+ ubifs_dump_node(c, buf);
+ dump_stack();
return -EINVAL;
}
@@ -790,38 +983,37 @@ int ubifs_read_node(const struct ubifs_info *c, void *buf, int type, int len,
ubifs_assert(!(offs & 7) && offs < c->leb_size);
ubifs_assert(type >= 0 && type < UBIFS_NODE_TYPES_CNT);
- err = ubi_read(c->ubi, lnum, buf, offs, len);
- if (err && err != -EBADMSG) {
- ubifs_err("cannot read node %d from LEB %d:%d, error %d",
- type, lnum, offs, err);
+ err = ubifs_leb_read(c, lnum, buf, offs, len, 0);
+ if (err && err != -EBADMSG)
return err;
- }
if (type != ch->node_type) {
- ubifs_err("bad node type (%d but expected %d)",
- ch->node_type, type);
+ ubifs_errc(c, "bad node type (%d but expected %d)",
+ ch->node_type, type);
goto out;
}
err = ubifs_check_node(c, buf, lnum, offs, 0, 0);
if (err) {
- ubifs_err("expected node type %d", type);
+ ubifs_errc(c, "expected node type %d", type);
return err;
}
l = le32_to_cpu(ch->len);
if (l != len) {
- ubifs_err("bad node length %d, expected %d", l, len);
+ ubifs_errc(c, "bad node length %d, expected %d", l, len);
goto out;
}
return 0;
out:
- 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();
+ ubifs_errc(c, "bad node at LEB %d:%d, LEB mapping status %d", lnum,
+ offs, ubi_is_mapped(c->ubi, lnum));
+ if (!c->probing) {
+ ubifs_dump_node(c, buf);
+ dump_stack();
+ }
return -EINVAL;
}
@@ -837,11 +1029,11 @@ 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);
@@ -851,8 +1043,14 @@ 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;
- wbuf->dtype = UBI_UNKNOWN;
+ /*
+ * 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->sync_callback = NULL;
mutex_init(&wbuf->io_mutex);
spin_lock_init(&wbuf->lock);
diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c
index 8aacd64957a..648b143606c 100644
--- a/fs/ubifs/ioctl.c
+++ b/fs/ubifs/ioctl.c
@@ -147,7 +147,7 @@ out_unlock:
long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
int flags, err;
- struct inode *inode = file->f_path.dentry->d_inode;
+ struct inode *inode = file_inode(file);
switch (cmd) {
case FS_IOC_GETFLAGS:
@@ -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))
@@ -173,12 +173,12 @@ long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
* Make sure the file-system is read-write and make sure it
* will not become read-only while we are changing the flags.
*/
- err = mnt_want_write(file->f_path.mnt);
+ err = mnt_want_write_file(file);
if (err)
return err;
dbg_gen("set flags: %#x, i_flags %#x", flags, inode->i_flags);
err = setflags(inode, flags);
- mnt_drop_write(file->f_path.mnt);
+ mnt_drop_write_file(file);
return err;
}
diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c
index 914f1bd89e5..0e045e75abd 100644
--- a/fs/ubifs/journal.c
+++ b/fs/ubifs/journal.c
@@ -141,14 +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)
@@ -203,13 +197,24 @@ again:
return 0;
}
- err = ubifs_add_bud_to_log(c, jhead, lnum, 0);
- if (err)
- goto out_return;
offs = 0;
out:
- err = ubifs_wbuf_seek_nolock(wbuf, lnum, offs, wbuf->dtype);
+ /*
+ * 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);
if (err)
goto out_unlock;
@@ -380,11 +385,9 @@ 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_lprops(c);
+ dump_stack();
+ ubifs_dump_budg(c, &c->bi);
+ ubifs_dump_lprops(c);
cmt_retries = dbg_check_lprops(c);
up_write(&c->commit_sem);
}
@@ -466,8 +469,8 @@ static void pack_inode(struct ubifs_info *c, struct ubifs_ino_node *ino,
ino->ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
ino->mtime_sec = cpu_to_le64(inode->i_mtime.tv_sec);
ino->mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
- ino->uid = cpu_to_le32(inode->i_uid);
- ino->gid = cpu_to_le32(inode->i_gid);
+ ino->uid = cpu_to_le32(i_uid_read(inode));
+ ino->gid = cpu_to_le32(i_gid_read(inode));
ino->mode = cpu_to_le32(inode->i_mode);
ino->flags = cpu_to_le32(ui->flags);
ino->size = cpu_to_le64(ui->ui_size);
@@ -666,6 +669,7 @@ out_free:
out_release:
release_head(c, BASEHD);
+ kfree(dent);
out_ro:
ubifs_ro_mode(c, err);
if (last_reference)
@@ -690,17 +694,26 @@ 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",
- (unsigned long)key_inum(c, key), key_block(c, key), len,
- DBGKEY(key));
+ dbg_jnlk(key, "ino %lu, blk %u, len %d, key ",
+ (unsigned long)key_inum(c, key), key_block(c, key), len);
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);
@@ -736,7 +749,10 @@ 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:
@@ -745,7 +761,10 @@ 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;
}
@@ -914,10 +933,8 @@ int ubifs_jnl_rename(struct ubifs_info *c, const struct inode *old_dir,
int move = (old_dir != new_dir);
struct ubifs_inode *uninitialized_var(new_ui);
- dbg_jnl("dent '%.*s' in dir ino %lu to dent '%.*s' in dir ino %lu",
- old_dentry->d_name.len, old_dentry->d_name.name,
- old_dir->i_ino, new_dentry->d_name.len,
- new_dentry->d_name.name, new_dir->i_ino);
+ dbg_jnl("dent '%pd' in dir ino %lu to dent '%pd' in dir ino %lu",
+ old_dentry, old_dir->i_ino, new_dentry, new_dir->i_ino);
ubifs_assert(ubifs_inode(old_dir)->data_len == 0);
ubifs_assert(ubifs_inode(new_dir)->data_len == 0);
ubifs_assert(mutex_is_locked(&ubifs_inode(old_dir)->ui_mutex));
@@ -1157,7 +1174,7 @@ int ubifs_jnl_truncate(struct ubifs_info *c, const struct inode *inode,
dn = (void *)trun + UBIFS_TRUN_NODE_SZ;
blk = new_size >> UBIFS_BLOCK_SHIFT;
data_key_init(c, &key, inum, blk);
- dbg_jnl("last block key %s", DBGKEY(&key));
+ dbg_jnlk(&key, "last block key ");
err = ubifs_tnc_lookup(c, &key, dn);
if (err == -ENOENT)
dlen = 0; /* Not found (so it is a hole) */
@@ -1248,7 +1265,6 @@ out_free:
return err;
}
-#ifdef CONFIG_UBIFS_FS_XATTR
/**
* ubifs_jnl_delete_xattr - delete an extended attribute.
@@ -1443,4 +1459,3 @@ out_free:
return err;
}
-#endif /* CONFIG_UBIFS_FS_XATTR */
diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c
index 4d0cb124146..a902c5919e4 100644
--- a/fs/ubifs/log.c
+++ b/fs/ubifs/log.c
@@ -29,11 +29,7 @@
#include "ubifs.h"
-#ifdef CONFIG_UBIFS_FS_DEBUG
static int dbg_check_bud_bytes(struct ubifs_info *c);
-#else
-#define dbg_check_bud_bytes(c) 0
-#endif
/**
* ubifs_search_bud - search bud LEB.
@@ -100,20 +96,6 @@ 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
*/
@@ -175,26 +157,6 @@ 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
@@ -277,7 +239,7 @@ 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;
}
@@ -296,7 +258,7 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs)
* an unclean reboot, because the target LEB might have been
* unmapped, but not yet physically erased.
*/
- err = ubi_leb_map(c->ubi, bud->lnum, UBI_SHORTTERM);
+ err = ubifs_leb_map(c, bud->lnum);
if (err)
goto out_unlock;
}
@@ -304,7 +266,7 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs)
dbg_log("write ref LEB %d:%d",
c->lhead_lnum, c->lhead_offs);
err = ubifs_write_node(c, ref, UBIFS_REF_NODE_SZ, c->lhead_lnum,
- c->lhead_offs, UBI_SHORTTERM);
+ c->lhead_offs);
if (err)
goto out_unlock;
@@ -317,8 +279,6 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs)
return 0;
out_unlock:
- if (err != -EAGAIN)
- ubifs_ro_mode(c, err);
mutex_unlock(&c->log_mutex);
kfree(ref);
kfree(bud);
@@ -355,17 +315,15 @@ static void remove_buds(struct ubifs_info *c)
* heads (non-closed buds).
*/
c->cmt_bud_bytes += wbuf->offs - bud->start;
- dbg_log("preserve %d:%d, jhead %s, bud bytes %d, "
- "cmt_bud_bytes %lld", bud->lnum, bud->start,
- dbg_jhead(bud->jhead), wbuf->offs - bud->start,
- c->cmt_bud_bytes);
+ dbg_log("preserve %d:%d, jhead %s, bud bytes %d, cmt_bud_bytes %lld",
+ bud->lnum, bud->start, dbg_jhead(bud->jhead),
+ wbuf->offs - bud->start, c->cmt_bud_bytes);
bud->start = wbuf->offs;
} else {
c->cmt_bud_bytes += c->leb_size - bud->start;
- dbg_log("remove %d:%d, jhead %s, bud bytes %d, "
- "cmt_bud_bytes %lld", bud->lnum, bud->start,
- dbg_jhead(bud->jhead), c->leb_size - bud->start,
- c->cmt_bud_bytes);
+ dbg_log("remove %d:%d, jhead %s, bud bytes %d, cmt_bud_bytes %lld",
+ bud->lnum, bud->start, dbg_jhead(bud->jhead),
+ c->leb_size - bud->start, c->cmt_bud_bytes);
rb_erase(p1, &c->buds);
/*
* If the commit does not finish, the recovery will need
@@ -445,7 +403,7 @@ 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;
}
@@ -458,7 +416,7 @@ int ubifs_log_start_commit(struct ubifs_info *c, int *ltail_lnum)
len = ALIGN(len, c->min_io_size);
dbg_log("writing commit start at LEB %d:0, len %d", c->lhead_lnum, len);
- err = ubifs_leb_write(c, c->lhead_lnum, cs, 0, len, UBI_SHORTTERM);
+ err = ubifs_leb_write(c, c->lhead_lnum, cs, 0, len);
if (err)
goto out;
@@ -466,7 +424,7 @@ 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 +511,7 @@ 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)
@@ -616,27 +574,10 @@ static int done_already(struct rb_root *done_tree, int lnum)
*/
static void destroy_done_tree(struct rb_root *done_tree)
{
- struct rb_node *this = done_tree->rb_node;
- struct done_ref *dr;
+ struct done_ref *dr, *n;
- while (this) {
- if (this->rb_left) {
- this = this->rb_left;
- continue;
- } else if (this->rb_right) {
- this = this->rb_right;
- continue;
- }
- dr = rb_entry(this, struct done_ref, rb);
- this = rb_parent(this);
- if (this) {
- if (this->rb_left == &dr->rb)
- this->rb_left = NULL;
- else
- this->rb_right = NULL;
- }
+ rbtree_postorder_for_each_entry_safe(dr, n, done_tree, rb)
kfree(dr);
- }
}
/**
@@ -659,10 +600,10 @@ static int add_node(struct ubifs_info *c, void *buf, int *lnum, int *offs,
int sz = ALIGN(*offs, c->min_io_size), err;
ubifs_pad(c, buf + *offs, sz - *offs);
- err = ubifs_leb_change(c, *lnum, buf, sz, UBI_SHORTTERM);
+ err = ubifs_leb_change(c, *lnum, buf, sz);
if (err)
return err;
- *lnum = next_log_lnum(c, *lnum);
+ *lnum = ubifs_next_log_lnum(c, *lnum);
*offs = 0;
}
memcpy(buf + *offs, node, len);
@@ -732,13 +673,13 @@ 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);
ubifs_pad(c, buf + offs, sz - offs);
- err = ubifs_leb_change(c, write_lnum, buf, sz, UBI_SHORTTERM);
+ err = ubifs_leb_change(c, write_lnum, buf, sz);
if (err)
goto out_free;
offs = ALIGN(offs, c->min_io_size);
@@ -752,7 +693,7 @@ 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;
@@ -770,8 +711,6 @@ out_free:
return err;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
-
/**
* dbg_check_bud_bytes - make sure bud bytes calculation are all right.
* @c: UBIFS file-system description object
@@ -786,7 +725,7 @@ static int dbg_check_bud_bytes(struct ubifs_info *c)
struct ubifs_bud *bud;
long long bud_bytes = 0;
- if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+ if (!dbg_is_chk_gen(c))
return 0;
spin_lock(&c->buds_lock);
@@ -803,5 +742,3 @@ static int dbg_check_bud_bytes(struct ubifs_info *c)
return err;
}
-
-#endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/lprops.c b/fs/ubifs/lprops.c
index 4d4ca388889..46190a7c42a 100644
--- a/fs/ubifs/lprops.c
+++ b/fs/ubifs/lprops.c
@@ -300,8 +300,11 @@ void ubifs_add_to_cat(struct ubifs_info *c, struct ubifs_lprops *lprops,
default:
ubifs_assert(0);
}
+
lprops->flags &= ~LPROPS_CAT_MASK;
lprops->flags |= cat;
+ c->in_a_category_cnt += 1;
+ ubifs_assert(c->in_a_category_cnt <= c->main_lebs);
}
/**
@@ -334,6 +337,9 @@ static void ubifs_remove_from_cat(struct ubifs_info *c,
default:
ubifs_assert(0);
}
+
+ c->in_a_category_cnt -= 1;
+ ubifs_assert(c->in_a_category_cnt >= 0);
}
/**
@@ -447,7 +453,7 @@ static void change_category(struct ubifs_info *c, struct ubifs_lprops *lprops)
int new_cat = ubifs_categorize_lprops(c, lprops);
if (old_cat == new_cat) {
- struct ubifs_lpt_heap *heap = &c->lpt_heap[new_cat - 1];
+ struct ubifs_lpt_heap *heap;
/* lprops on a heap now must be moved up or down */
if (new_cat < 1 || new_cat > LPROPS_HEAP_CNT)
@@ -504,7 +510,7 @@ static int is_lprops_dirty(struct ubifs_info *c, struct ubifs_lprops *lprops)
pnode = (struct ubifs_pnode *)container_of(lprops - pos,
struct ubifs_pnode,
lprops[0]);
- return !test_bit(COW_ZNODE, &pnode->flags) &&
+ return !test_bit(COW_CNODE, &pnode->flags) &&
test_bit(DIRTY_CNODE, &pnode->flags);
}
@@ -846,7 +852,9 @@ const struct ubifs_lprops *ubifs_fast_find_frdi_idx(struct ubifs_info *c)
return lprops;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
+/*
+ * Everything below is related to debugging.
+ */
/**
* dbg_check_cats - check category heaps and lists.
@@ -860,20 +868,20 @@ int dbg_check_cats(struct ubifs_info *c)
struct list_head *pos;
int i, cat;
- if (!(ubifs_chk_flags & (UBIFS_CHK_GEN | UBIFS_CHK_LPROPS)))
+ if (!dbg_is_chk_gen(c) && !dbg_is_chk_lprops(c))
return 0;
list_for_each_entry(lprops, &c->empty_list, list) {
if (lprops->free != c->leb_size) {
- ubifs_err("non-empty LEB %d on empty list "
- "(free %d dirty %d flags %d)", lprops->lnum,
- lprops->free, lprops->dirty, lprops->flags);
+ ubifs_err("non-empty LEB %d on empty list (free %d dirty %d flags %d)",
+ lprops->lnum, lprops->free, lprops->dirty,
+ lprops->flags);
return -EINVAL;
}
if (lprops->flags & LPROPS_TAKEN) {
- ubifs_err("taken LEB %d on empty list "
- "(free %d dirty %d flags %d)", lprops->lnum,
- lprops->free, lprops->dirty, lprops->flags);
+ ubifs_err("taken LEB %d on empty list (free %d dirty %d flags %d)",
+ lprops->lnum, lprops->free, lprops->dirty,
+ lprops->flags);
return -EINVAL;
}
}
@@ -881,15 +889,15 @@ int dbg_check_cats(struct ubifs_info *c)
i = 0;
list_for_each_entry(lprops, &c->freeable_list, list) {
if (lprops->free + lprops->dirty != c->leb_size) {
- ubifs_err("non-freeable LEB %d on freeable list "
- "(free %d dirty %d flags %d)", lprops->lnum,
- lprops->free, lprops->dirty, lprops->flags);
+ ubifs_err("non-freeable LEB %d on freeable list (free %d dirty %d flags %d)",
+ lprops->lnum, lprops->free, lprops->dirty,
+ lprops->flags);
return -EINVAL;
}
if (lprops->flags & LPROPS_TAKEN) {
- ubifs_err("taken LEB %d on freeable list "
- "(free %d dirty %d flags %d)", lprops->lnum,
- lprops->free, lprops->dirty, lprops->flags);
+ ubifs_err("taken LEB %d on freeable list (free %d dirty %d flags %d)",
+ lprops->lnum, lprops->free, lprops->dirty,
+ lprops->flags);
return -EINVAL;
}
i += 1;
@@ -911,21 +919,21 @@ int dbg_check_cats(struct ubifs_info *c)
list_for_each_entry(lprops, &c->frdi_idx_list, list) {
if (lprops->free + lprops->dirty != c->leb_size) {
- ubifs_err("non-freeable LEB %d on frdi_idx list "
- "(free %d dirty %d flags %d)", lprops->lnum,
- lprops->free, lprops->dirty, lprops->flags);
+ ubifs_err("non-freeable LEB %d on frdi_idx list (free %d dirty %d flags %d)",
+ lprops->lnum, lprops->free, lprops->dirty,
+ lprops->flags);
return -EINVAL;
}
if (lprops->flags & LPROPS_TAKEN) {
- ubifs_err("taken LEB %d on frdi_idx list "
- "(free %d dirty %d flags %d)", lprops->lnum,
- lprops->free, lprops->dirty, lprops->flags);
+ ubifs_err("taken LEB %d on frdi_idx list (free %d dirty %d flags %d)",
+ lprops->lnum, lprops->free, lprops->dirty,
+ lprops->flags);
return -EINVAL;
}
if (!(lprops->flags & LPROPS_INDEX)) {
- ubifs_err("non-index LEB %d on frdi_idx list "
- "(free %d dirty %d flags %d)", lprops->lnum,
- lprops->free, lprops->dirty, lprops->flags);
+ ubifs_err("non-index LEB %d on frdi_idx list (free %d dirty %d flags %d)",
+ lprops->lnum, lprops->free, lprops->dirty,
+ lprops->flags);
return -EINVAL;
}
}
@@ -958,7 +966,7 @@ void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat,
{
int i = 0, j, err = 0;
- if (!(ubifs_chk_flags & (UBIFS_CHK_GEN | UBIFS_CHK_LPROPS)))
+ if (!dbg_is_chk_gen(c) && !dbg_is_chk_lprops(c))
return;
for (i = 0; i < heap->cnt; i++) {
@@ -980,9 +988,9 @@ void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat,
goto out;
}
if (lprops != lp) {
- dbg_msg("lprops %zx lp %zx lprops->lnum %d lp->lnum %d",
- (size_t)lprops, (size_t)lp, lprops->lnum,
- lp->lnum);
+ ubifs_err("lprops %zx lp %zx lprops->lnum %d lp->lnum %d",
+ (size_t)lprops, (size_t)lp, lprops->lnum,
+ lp->lnum);
err = 4;
goto out;
}
@@ -1000,28 +1008,18 @@ void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat,
}
out:
if (err) {
- dbg_msg("failed cat %d hpos %d err %d", cat, i, err);
- dbg_dump_stack();
- dbg_dump_heap(c, heap, cat);
+ ubifs_err("failed cat %d hpos %d err %d", cat, i, err);
+ dump_stack();
+ ubifs_dump_heap(c, heap, cat);
}
}
/**
- * 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 +1028,12 @@ 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 +1041,7 @@ 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 +1075,7 @@ 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 +1087,40 @@ 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;
+
+ /*
+ * 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;
+ }
- 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;
+ sleb = ubifs_scan(c, lnum, 0, buf, 0);
+ if (IS_ERR(sleb)) {
+ ret = PTR_ERR(sleb);
+ if (ret == -EUCLEAN) {
+ ubifs_dump_lprops(c);
+ ubifs_dump_budg(c, &c->bi);
}
- data->err = PTR_ERR(sleb);
- return LPT_SCAN_STOP;
+ goto out;
}
is_idx = -1;
@@ -1157,8 +1159,8 @@ static int scan_check_cb(struct ubifs_info *c,
if (free > c->leb_size || free < 0 || dirty > c->leb_size ||
dirty < 0) {
- ubifs_err("bad calculated accounting for LEB %d: "
- "free %d, dirty %d", lnum, free, dirty);
+ ubifs_err("bad calculated accounting for LEB %d: free %d, dirty %d",
+ lnum, free, dirty);
goto out_destroy;
}
@@ -1204,8 +1206,7 @@ static int scan_check_cb(struct ubifs_info *c,
/* Free but not unmapped LEB, it's fine */
is_idx = 0;
else {
- ubifs_err("indexing node without indexing "
- "flag");
+ ubifs_err("indexing node without indexing flag");
goto out_print;
}
}
@@ -1236,18 +1237,19 @@ static int scan_check_cb(struct ubifs_info *c,
}
ubifs_scan_destroy(sleb);
+ vfree(buf);
return LPT_SCAN_CONTINUE;
out_print:
- ubifs_err("bad accounting of LEB %d: free %d, dirty %d flags %#x, "
- "should be free %d, dirty %d",
+ ubifs_err("bad accounting of LEB %d: free %d, dirty %d flags %#x, should be free %d, dirty %d",
lnum, lp->free, lp->dirty, lp->flags, free, dirty);
- dbg_dump_leb(c, lnum);
+ ubifs_dump_leb(c, lnum);
out_destroy:
ubifs_scan_destroy(sleb);
+ ret = -EINVAL;
out:
- data->err = -EINVAL;
- return LPT_SCAN_STOP;
+ vfree(buf);
+ return ret;
}
/**
@@ -1264,10 +1266,9 @@ 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))
+ if (!dbg_is_chk_lprops(c))
return 0;
/*
@@ -1280,42 +1281,34 @@ 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);
- ubifs_err("read from lprops: empty_lebs %d, idx_lebs %d, "
- "total_free %lld, total_dirty %lld, total_used %lld",
+ 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);
+ 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,
c->lst.total_dirty, c->lst.total_used);
err = -EINVAL;
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;
@@ -1326,5 +1319,3 @@ int dbg_check_lprops(struct ubifs_info *c)
out:
return err;
}
-
-#endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c
index 72775d35b99..d46b19ec181 100644
--- a/fs/ubifs/lpt.c
+++ b/fs/ubifs/lpt.c
@@ -701,8 +701,7 @@ int ubifs_create_dflt_lpt(struct ubifs_info *c, int *main_lebs, int lpt_first,
alen = ALIGN(len, c->min_io_size);
set_ltab(c, lnum, c->leb_size - alen, alen - len);
memset(p, 0xff, alen - len);
- err = ubi_leb_change(c->ubi, lnum++, buf, alen,
- UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum++, buf, alen);
if (err)
goto out;
p = buf;
@@ -732,8 +731,7 @@ int ubifs_create_dflt_lpt(struct ubifs_info *c, int *main_lebs, int lpt_first,
set_ltab(c, lnum, c->leb_size - alen,
alen - len);
memset(p, 0xff, alen - len);
- err = ubi_leb_change(c->ubi, lnum++, buf, alen,
- UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum++, buf, alen);
if (err)
goto out;
p = buf;
@@ -780,8 +778,7 @@ int ubifs_create_dflt_lpt(struct ubifs_info *c, int *main_lebs, int lpt_first,
alen = ALIGN(len, c->min_io_size);
set_ltab(c, lnum, c->leb_size - alen, alen - len);
memset(p, 0xff, alen - len);
- err = ubi_leb_change(c->ubi, lnum++, buf, alen,
- UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum++, buf, alen);
if (err)
goto out;
p = buf;
@@ -806,7 +803,7 @@ int ubifs_create_dflt_lpt(struct ubifs_info *c, int *main_lebs, int lpt_first,
alen = ALIGN(len, c->min_io_size);
set_ltab(c, lnum, c->leb_size - alen, alen - len);
memset(p, 0xff, alen - len);
- err = ubi_leb_change(c->ubi, lnum++, buf, alen, UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum++, buf, alen);
if (err)
goto out;
p = buf;
@@ -826,7 +823,7 @@ int ubifs_create_dflt_lpt(struct ubifs_info *c, int *main_lebs, int lpt_first,
/* Write remaining buffer */
memset(p, 0xff, alen - len);
- err = ubi_leb_change(c->ubi, lnum, buf, alen, UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum, buf, alen);
if (err)
goto out;
@@ -926,7 +923,7 @@ static int check_lpt_crc(void *buf, int len)
if (crc != calc_crc) {
ubifs_err("invalid crc in LPT node: crc %hx calc %hx", crc,
calc_crc);
- dbg_dump_stack();
+ dump_stack();
return -EINVAL;
}
return 0;
@@ -949,7 +946,7 @@ static int check_lpt_type(uint8_t **addr, int *pos, int type)
if (node_type != type) {
ubifs_err("invalid type (%d) in LPT node type %d", node_type,
type);
- dbg_dump_stack();
+ dump_stack();
return -EINVAL;
}
return 0;
@@ -1222,7 +1219,7 @@ int ubifs_read_nnode(struct ubifs_info *c, struct ubifs_nnode *parent, int iip)
if (c->big_lpt)
nnode->num = calc_nnode_num_from_parent(c, parent, iip);
} else {
- err = ubi_read(c->ubi, lnum, buf, offs, c->nnode_sz);
+ err = ubifs_leb_read(c, lnum, buf, offs, c->nnode_sz, 1);
if (err)
goto out;
err = ubifs_unpack_nnode(c, buf, nnode);
@@ -1247,6 +1244,7 @@ int ubifs_read_nnode(struct ubifs_info *c, struct ubifs_nnode *parent, int iip)
out:
ubifs_err("error %d reading nnode at %d:%d", err, lnum, offs);
+ dump_stack();
kfree(nnode);
return err;
}
@@ -1270,10 +1268,9 @@ 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
@@ -1291,7 +1288,7 @@ static int read_pnode(struct ubifs_info *c, struct ubifs_nnode *parent, int iip)
lprops->flags = ubifs_categorize_lprops(c, lprops);
}
} else {
- err = ubi_read(c->ubi, lnum, buf, offs, c->pnode_sz);
+ err = ubifs_leb_read(c, lnum, buf, offs, c->pnode_sz, 1);
if (err)
goto out;
err = unpack_pnode(c, buf, pnode);
@@ -1312,8 +1309,9 @@ static int read_pnode(struct ubifs_info *c, struct ubifs_nnode *parent, int iip)
out:
ubifs_err("error %d reading pnode at %d:%d", err, lnum, offs);
- dbg_dump_pnode(c, pnode, parent, iip);
- dbg_msg("calc num: %d", calc_pnode_num_from_parent(c, parent, iip));
+ ubifs_dump_pnode(c, pnode, parent, iip);
+ dump_stack();
+ ubifs_err("calc num: %d", calc_pnode_num_from_parent(c, parent, iip));
kfree(pnode);
return err;
}
@@ -1332,7 +1330,7 @@ static int read_ltab(struct ubifs_info *c)
buf = vmalloc(c->ltab_sz);
if (!buf)
return -ENOMEM;
- err = ubi_read(c->ubi, c->ltab_lnum, buf, c->ltab_offs, c->ltab_sz);
+ err = ubifs_leb_read(c, c->ltab_lnum, buf, c->ltab_offs, c->ltab_sz, 1);
if (err)
goto out;
err = unpack_ltab(c, buf);
@@ -1355,7 +1353,8 @@ static int read_lsave(struct ubifs_info *c)
buf = vmalloc(c->lsave_sz);
if (!buf)
return -ENOMEM;
- err = ubi_read(c->ubi, c->lsave_lnum, buf, c->lsave_offs, c->lsave_sz);
+ err = ubifs_leb_read(c, c->lsave_lnum, buf, c->lsave_offs,
+ c->lsave_sz, 1);
if (err)
goto out;
err = unpack_lsave(c, buf);
@@ -1738,16 +1737,23 @@ int ubifs_lpt_init(struct ubifs_info *c, int rd, int wr)
if (rd) {
err = lpt_init_rd(c);
if (err)
- return err;
+ goto out_err;
}
if (wr) {
err = lpt_init_wr(c);
if (err)
- return err;
+ goto out_err;
}
return 0;
+
+out_err:
+ if (wr)
+ ubifs_lpt_free(c, 1);
+ if (rd)
+ ubifs_lpt_free(c, 0);
+ return err;
}
/**
@@ -1815,8 +1821,8 @@ static struct ubifs_nnode *scan_get_nnode(struct ubifs_info *c,
if (c->big_lpt)
nnode->num = calc_nnode_num_from_parent(c, parent, iip);
} else {
- err = ubi_read(c->ubi, branch->lnum, buf, branch->offs,
- c->nnode_sz);
+ err = ubifs_leb_read(c, branch->lnum, buf, branch->offs,
+ c->nnode_sz, 1);
if (err)
return ERR_PTR(err);
err = ubifs_unpack_nnode(c, buf, nnode);
@@ -1884,8 +1890,8 @@ static struct ubifs_pnode *scan_get_pnode(struct ubifs_info *c,
ubifs_assert(branch->lnum >= c->lpt_first &&
branch->lnum <= c->lpt_last);
ubifs_assert(branch->offs >= 0 && branch->offs < c->leb_size);
- err = ubi_read(c->ubi, branch->lnum, buf, branch->offs,
- c->pnode_sz);
+ err = ubifs_leb_read(c, branch->lnum, buf, branch->offs,
+ c->pnode_sz, 1);
if (err)
return ERR_PTR(err);
err = unpack_pnode(c, buf, pnode);
@@ -1984,12 +1990,11 @@ again:
if (path[h].in_tree)
continue;
- nnode = kmalloc(sz, GFP_NOFS);
+ nnode = kmemdup(&path[h].nnode, sz, GFP_NOFS);
if (!nnode) {
err = -ENOMEM;
goto out;
}
- memcpy(nnode, &path[h].nnode, sz);
parent = nnode->parent;
parent->nbranch[nnode->iip].nnode = nnode;
path[h].ptr.nnode = nnode;
@@ -2002,12 +2007,11 @@ again:
const size_t sz = sizeof(struct ubifs_pnode);
struct ubifs_nnode *parent;
- pnode = kmalloc(sz, GFP_NOFS);
+ pnode = kmemdup(&path[h].pnode, sz, GFP_NOFS);
if (!pnode) {
err = -ENOMEM;
goto out;
}
- memcpy(pnode, &path[h].pnode, sz);
parent = pnode->parent;
parent->nbranch[pnode->iip].pnode = pnode;
path[h].ptr.pnode = pnode;
@@ -2080,8 +2084,6 @@ out:
return err;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
-
/**
* dbg_chk_pnode - check a pnode.
* @c: the UBIFS file-system description object
@@ -2096,8 +2098,8 @@ static int dbg_chk_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
int i;
if (pnode->num != col) {
- dbg_err("pnode num %d expected %d parent num %d iip %d",
- pnode->num, col, pnode->parent->num, pnode->iip);
+ ubifs_err("pnode num %d expected %d parent num %d iip %d",
+ pnode->num, col, pnode->parent->num, pnode->iip);
return -EINVAL;
}
for (i = 0; i < UBIFS_LPT_FANOUT; i++) {
@@ -2111,14 +2113,14 @@ static int dbg_chk_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
if (lnum >= c->leb_cnt)
continue;
if (lprops->lnum != lnum) {
- dbg_err("bad LEB number %d expected %d",
- lprops->lnum, lnum);
+ ubifs_err("bad LEB number %d expected %d",
+ lprops->lnum, lnum);
return -EINVAL;
}
if (lprops->flags & LPROPS_TAKEN) {
if (cat != LPROPS_UNCAT) {
- dbg_err("LEB %d taken but not uncat %d",
- lprops->lnum, cat);
+ ubifs_err("LEB %d taken but not uncat %d",
+ lprops->lnum, cat);
return -EINVAL;
}
continue;
@@ -2130,8 +2132,8 @@ static int dbg_chk_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
case LPROPS_FRDI_IDX:
break;
default:
- dbg_err("LEB %d index but cat %d",
- lprops->lnum, cat);
+ ubifs_err("LEB %d index but cat %d",
+ lprops->lnum, cat);
return -EINVAL;
}
} else {
@@ -2143,8 +2145,8 @@ static int dbg_chk_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
case LPROPS_FREEABLE:
break;
default:
- dbg_err("LEB %d not index but cat %d",
- lprops->lnum, cat);
+ ubifs_err("LEB %d not index but cat %d",
+ lprops->lnum, cat);
return -EINVAL;
}
}
@@ -2184,24 +2186,24 @@ static int dbg_chk_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode,
break;
}
if (!found) {
- dbg_err("LEB %d cat %d not found in cat heap/list",
- lprops->lnum, cat);
+ ubifs_err("LEB %d cat %d not found in cat heap/list",
+ lprops->lnum, cat);
return -EINVAL;
}
switch (cat) {
case LPROPS_EMPTY:
if (lprops->free != c->leb_size) {
- dbg_err("LEB %d cat %d free %d dirty %d",
- lprops->lnum, cat, lprops->free,
- lprops->dirty);
+ ubifs_err("LEB %d cat %d free %d dirty %d",
+ lprops->lnum, cat, lprops->free,
+ lprops->dirty);
return -EINVAL;
}
case LPROPS_FREEABLE:
case LPROPS_FRDI_IDX:
if (lprops->free + lprops->dirty != c->leb_size) {
- dbg_err("LEB %d cat %d free %d dirty %d",
- lprops->lnum, cat, lprops->free,
- lprops->dirty);
+ ubifs_err("LEB %d cat %d free %d dirty %d",
+ lprops->lnum, cat, lprops->free,
+ lprops->dirty);
return -EINVAL;
}
}
@@ -2225,7 +2227,7 @@ int dbg_check_lpt_nodes(struct ubifs_info *c, struct ubifs_cnode *cnode,
struct ubifs_cnode *cn;
int num, iip = 0, err;
- if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
+ if (!dbg_is_chk_lprops(c))
return 0;
while (cnode) {
@@ -2235,9 +2237,9 @@ int dbg_check_lpt_nodes(struct ubifs_info *c, struct ubifs_cnode *cnode,
/* cnode is a nnode */
num = calc_nnode_num(row, col);
if (cnode->num != num) {
- dbg_err("nnode num %d expected %d "
- "parent num %d iip %d", cnode->num, num,
- (nnode ? nnode->num : 0), cnode->iip);
+ ubifs_err("nnode num %d expected %d parent num %d iip %d",
+ cnode->num, num,
+ (nnode ? nnode->num : 0), cnode->iip);
return -EINVAL;
}
nn = (struct ubifs_nnode *)cnode;
@@ -2274,5 +2276,3 @@ int dbg_check_lpt_nodes(struct ubifs_info *c, struct ubifs_cnode *cnode,
}
return 0;
}
-
-#endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c
index 5c90dec5db0..45d4e96a6ba 100644
--- a/fs/ubifs/lpt_commit.c
+++ b/fs/ubifs/lpt_commit.c
@@ -27,8 +27,11 @@
#include <linux/crc16.h>
#include <linux/slab.h>
+#include <linux/random.h>
#include "ubifs.h"
+static int dbg_populate_lsave(struct ubifs_info *c);
+
/**
* first_dirty_cnode - find first dirty cnode.
* @c: UBIFS file-system description object
@@ -110,8 +113,8 @@ static int get_cnodes_to_commit(struct ubifs_info *c)
return 0;
cnt += 1;
while (1) {
- ubifs_assert(!test_bit(COW_ZNODE, &cnode->flags));
- __set_bit(COW_ZNODE, &cnode->flags);
+ ubifs_assert(!test_bit(COW_CNODE, &cnode->flags));
+ __set_bit(COW_CNODE, &cnode->flags);
cnext = next_dirty_cnode(cnode);
if (!cnext) {
cnode->cnext = c->lpt_cnext;
@@ -317,11 +320,10 @@ static int layout_cnodes(struct ubifs_info *c)
return 0;
no_space:
- ubifs_err("LPT out of space");
- dbg_err("LPT out of space at LEB %d:%d needing %d, done_ltab %d, "
- "done_lsave %d", lnum, offs, len, done_ltab, done_lsave);
- dbg_dump_lpt_info(c);
- dbg_dump_lpt_lebs(c);
+ ubifs_err("LPT out of space at LEB %d:%d needing %d, done_ltab %d, done_lsave %d",
+ lnum, offs, len, done_ltab, done_lsave);
+ ubifs_dump_lpt_info(c);
+ ubifs_dump_lpt_lebs(c);
dump_stack();
return err;
}
@@ -414,7 +416,7 @@ static int write_cnodes(struct ubifs_info *c)
alen = ALIGN(wlen, c->min_io_size);
memset(buf + offs, 0xff, alen - wlen);
err = ubifs_leb_write(c, lnum, buf + from, from,
- alen, UBI_SHORTTERM);
+ alen);
if (err)
return err;
}
@@ -458,9 +460,9 @@ static int write_cnodes(struct ubifs_info *c)
* important.
*/
clear_bit(DIRTY_CNODE, &cnode->flags);
- smp_mb__before_clear_bit();
- clear_bit(COW_ZNODE, &cnode->flags);
- smp_mb__after_clear_bit();
+ smp_mb__before_atomic();
+ clear_bit(COW_CNODE, &cnode->flags);
+ smp_mb__after_atomic();
offs += len;
dbg_chk_lpt_sz(c, 1, len);
cnode = cnode->cnext;
@@ -472,8 +474,7 @@ static int write_cnodes(struct ubifs_info *c)
wlen = offs - from;
alen = ALIGN(wlen, c->min_io_size);
memset(buf + offs, 0xff, alen - wlen);
- err = ubifs_leb_write(c, lnum, buf + from, from, alen,
- UBI_SHORTTERM);
+ err = ubifs_leb_write(c, lnum, buf + from, from, alen);
if (err)
return err;
dbg_chk_lpt_sz(c, 2, c->leb_size - offs);
@@ -499,8 +500,7 @@ static int write_cnodes(struct ubifs_info *c)
wlen = offs - from;
alen = ALIGN(wlen, c->min_io_size);
memset(buf + offs, 0xff, alen - wlen);
- err = ubifs_leb_write(c, lnum, buf + from, from, alen,
- UBI_SHORTTERM);
+ err = ubifs_leb_write(c, lnum, buf + from, from, alen);
if (err)
return err;
dbg_chk_lpt_sz(c, 2, c->leb_size - offs);
@@ -524,7 +524,7 @@ static int write_cnodes(struct ubifs_info *c)
wlen = offs - from;
alen = ALIGN(wlen, c->min_io_size);
memset(buf + offs, 0xff, alen - wlen);
- err = ubifs_leb_write(c, lnum, buf + from, from, alen, UBI_SHORTTERM);
+ err = ubifs_leb_write(c, lnum, buf + from, from, alen);
if (err)
return err;
@@ -545,11 +545,10 @@ static int write_cnodes(struct ubifs_info *c)
return 0;
no_space:
- ubifs_err("LPT out of space mismatch");
- dbg_err("LPT out of space mismatch at LEB %d:%d needing %d, done_ltab "
- "%d, done_lsave %d", lnum, offs, len, done_ltab, done_lsave);
- dbg_dump_lpt_info(c);
- dbg_dump_lpt_lebs(c);
+ ubifs_err("LPT out of space mismatch at LEB %d:%d needing %d, done_ltab %d, done_lsave %d",
+ lnum, offs, len, done_ltab, done_lsave);
+ ubifs_dump_lpt_info(c);
+ ubifs_dump_lpt_lebs(c);
dump_stack();
return err;
}
@@ -586,7 +585,7 @@ 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);
@@ -815,6 +814,10 @@ 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)
@@ -1150,11 +1153,11 @@ static int lpt_gc_lnum(struct ubifs_info *c, int lnum)
void *buf = c->lpt_buf;
dbg_lp("LEB %d", lnum);
- err = ubi_read(c->ubi, lnum, buf, 0, c->leb_size);
- if (err) {
- ubifs_err("cannot read LEB %d, error %d", lnum, err);
+
+ err = ubifs_leb_read(c, lnum, buf, 0, c->leb_size, 1);
+ if (err)
return err;
- }
+
while (1) {
if (!is_a_node(c, buf, len)) {
int pad_len;
@@ -1486,7 +1489,9 @@ void ubifs_lpt_free(struct ubifs_info *c, int wr_only)
kfree(c->lpt_nod_buf);
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
+/*
+ * Everything below is related to debugging.
+ */
/**
* dbg_is_all_ff - determine if a buffer contains only 0xFF bytes.
@@ -1628,56 +1633,65 @@ 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))
+ if (!dbg_is_chk_lprops(c))
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;
+ 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 = ubifs_leb_read(c, lnum, buf, 0, c->leb_size, 1);
+ if (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)) {
- dbg_msg("invalid empty space in LEB %d at %d",
- lnum, c->leb_size - len);
+ if (!dbg_is_all_ff(p, len)) {
+ ubifs_err("invalid empty space in LEB %d at %d",
+ lnum, c->leb_size - len);
err = -EINVAL;
}
i = lnum - c->lpt_first;
if (len != c->ltab[i].free) {
- dbg_msg("invalid free space in LEB %d "
- "(free %d, expected %d)",
- lnum, len, c->ltab[i].free);
+ ubifs_err("invalid free space in LEB %d (free %d, expected %d)",
+ lnum, len, c->ltab[i].free);
err = -EINVAL;
}
if (dirty != c->ltab[i].dirty) {
- dbg_msg("invalid dirty space in LEB %d "
- "(dirty %d, expected %d)",
- lnum, dirty, c->ltab[i].dirty);
+ ubifs_err("invalid dirty space in LEB %d (dirty %d, expected %d)",
+ 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;
}
/**
@@ -1690,7 +1704,7 @@ int dbg_check_ltab(struct ubifs_info *c)
{
int lnum, err, i, cnt;
- if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
+ if (!dbg_is_chk_lprops(c))
return 0;
/* Bring the entire tree into memory */
@@ -1713,7 +1727,7 @@ int dbg_check_ltab(struct ubifs_info *c)
for (lnum = c->lpt_first; lnum <= c->lpt_last; lnum++) {
err = dbg_check_ltab_lnum(c, lnum);
if (err) {
- dbg_err("failed at LEB %d", lnum);
+ ubifs_err("failed at LEB %d", lnum);
return err;
}
}
@@ -1733,7 +1747,7 @@ int dbg_chk_lpt_free_spc(struct ubifs_info *c)
long long free = 0;
int i;
- if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
+ if (!dbg_is_chk_lprops(c))
return 0;
for (i = 0; i < c->lpt_lebs; i++) {
@@ -1745,10 +1759,10 @@ int dbg_chk_lpt_free_spc(struct ubifs_info *c)
free += c->leb_size;
}
if (free < c->lpt_sz) {
- dbg_err("LPT space error: free %lld lpt_sz %lld",
- free, c->lpt_sz);
- dbg_dump_lpt_info(c);
- dbg_dump_lpt_lebs(c);
+ ubifs_err("LPT space error: free %lld lpt_sz %lld",
+ free, c->lpt_sz);
+ ubifs_dump_lpt_info(c);
+ ubifs_dump_lpt_lebs(c);
dump_stack();
return -EINVAL;
}
@@ -1775,7 +1789,7 @@ int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len)
long long chk_lpt_sz, lpt_sz;
int err = 0;
- if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS))
+ if (!dbg_is_chk_lprops(c))
return 0;
switch (action) {
@@ -1785,13 +1799,13 @@ int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len)
d->chk_lpt_lebs = 0;
d->chk_lpt_wastage = 0;
if (c->dirty_pn_cnt > c->pnode_cnt) {
- dbg_err("dirty pnodes %d exceed max %d",
- c->dirty_pn_cnt, c->pnode_cnt);
+ ubifs_err("dirty pnodes %d exceed max %d",
+ c->dirty_pn_cnt, c->pnode_cnt);
err = -EINVAL;
}
if (c->dirty_nn_cnt > c->nnode_cnt) {
- dbg_err("dirty nnodes %d exceed max %d",
- c->dirty_nn_cnt, c->nnode_cnt);
+ ubifs_err("dirty nnodes %d exceed max %d",
+ c->dirty_nn_cnt, c->nnode_cnt);
err = -EINVAL;
}
return err;
@@ -1808,23 +1822,23 @@ int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len)
chk_lpt_sz *= d->chk_lpt_lebs;
chk_lpt_sz += len - c->nhead_offs;
if (d->chk_lpt_sz != chk_lpt_sz) {
- dbg_err("LPT wrote %lld but space used was %lld",
- d->chk_lpt_sz, chk_lpt_sz);
+ ubifs_err("LPT wrote %lld but space used was %lld",
+ d->chk_lpt_sz, chk_lpt_sz);
err = -EINVAL;
}
if (d->chk_lpt_sz > c->lpt_sz) {
- dbg_err("LPT wrote %lld but lpt_sz is %lld",
- d->chk_lpt_sz, c->lpt_sz);
+ ubifs_err("LPT wrote %lld but lpt_sz is %lld",
+ d->chk_lpt_sz, c->lpt_sz);
err = -EINVAL;
}
if (d->chk_lpt_sz2 && d->chk_lpt_sz != d->chk_lpt_sz2) {
- dbg_err("LPT layout size %lld but wrote %lld",
- d->chk_lpt_sz, d->chk_lpt_sz2);
+ ubifs_err("LPT layout size %lld but wrote %lld",
+ d->chk_lpt_sz, d->chk_lpt_sz2);
err = -EINVAL;
}
if (d->chk_lpt_sz2 && d->new_nhead_offs != len) {
- dbg_err("LPT new nhead offs: expected %d was %d",
- d->new_nhead_offs, len);
+ ubifs_err("LPT new nhead offs: expected %d was %d",
+ d->new_nhead_offs, len);
err = -EINVAL;
}
lpt_sz = (long long)c->pnode_cnt * c->pnode_sz;
@@ -1833,13 +1847,13 @@ int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len)
if (c->big_lpt)
lpt_sz += c->lsave_sz;
if (d->chk_lpt_sz - d->chk_lpt_wastage > lpt_sz) {
- dbg_err("LPT chk_lpt_sz %lld + waste %lld exceeds %lld",
- d->chk_lpt_sz, d->chk_lpt_wastage, lpt_sz);
+ ubifs_err("LPT chk_lpt_sz %lld + waste %lld exceeds %lld",
+ d->chk_lpt_sz, d->chk_lpt_wastage, lpt_sz);
err = -EINVAL;
}
if (err) {
- dbg_dump_lpt_info(c);
- dbg_dump_lpt_lebs(c);
+ ubifs_dump_lpt_info(c);
+ ubifs_dump_lpt_lebs(c);
dump_stack();
}
d->chk_lpt_sz2 = d->chk_lpt_sz;
@@ -1858,7 +1872,7 @@ int dbg_chk_lpt_sz(struct ubifs_info *c, int action, int len)
}
/**
- * dbg_dump_lpt_leb - dump an LPT LEB.
+ * ubifs_dump_lpt_leb - dump an LPT LEB.
* @c: UBIFS file-system description object
* @lnum: LEB number to dump
*
@@ -1870,45 +1884,48 @@ 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);
- err = ubi_read(c->ubi, lnum, buf, 0, c->leb_size);
- if (err) {
- ubifs_err("cannot read LEB %d, error %d", lnum, err);
+ pr_err("(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 = ubifs_leb_read(c, lnum, buf, 0, c->leb_size, 1);
+ if (err)
+ 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",
+ pr_err("LEB %d:%d, pad %d bytes\n",
lnum, offs, pad_len);
- buf += pad_len;
+ p += pad_len;
len -= pad_len;
continue;
}
if (len)
- printk(KERN_DEBUG "LEB %d:%d, free %d bytes\n",
+ pr_err("LEB %d:%d, free %d bytes\n",
lnum, offs, len);
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:
{
node_len = c->pnode_sz;
if (c->big_lpt)
- printk(KERN_DEBUG "LEB %d:%d, pnode num %d\n",
+ pr_err("LEB %d:%d, pnode num %d\n",
lnum, offs, node_num);
else
- printk(KERN_DEBUG "LEB %d:%d, pnode\n",
- lnum, offs);
+ pr_err("LEB %d:%d, pnode\n", lnum, offs);
break;
}
case UBIFS_LPT_NNODE:
@@ -1918,60 +1935,100 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
node_len = c->nnode_sz;
if (c->big_lpt)
- printk(KERN_DEBUG "LEB %d:%d, nnode num %d, ",
+ pr_err("LEB %d:%d, nnode num %d, ",
lnum, offs, node_num);
else
- printk(KERN_DEBUG "LEB %d:%d, nnode, ",
+ pr_err("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,
+ pr_cont("%d:%d", nnode.nbranch[i].lnum,
nnode.nbranch[i].offs);
if (i != UBIFS_LPT_FANOUT - 1)
- printk(KERN_CONT ", ");
+ pr_cont(", ");
}
- printk(KERN_CONT "\n");
+ pr_cont("\n");
break;
}
case UBIFS_LPT_LTAB:
node_len = c->ltab_sz;
- printk(KERN_DEBUG "LEB %d:%d, ltab\n",
- lnum, offs);
+ pr_err("LEB %d:%d, ltab\n", lnum, offs);
break;
case UBIFS_LPT_LSAVE:
node_len = c->lsave_sz;
- printk(KERN_DEBUG "LEB %d:%d, lsave len\n", lnum, offs);
+ pr_err("LEB %d:%d, lsave len\n", lnum, offs);
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);
+ pr_err("(pid %d) finish dumping LEB %d\n", current->pid, lnum);
+out:
+ vfree(buf);
+ return;
}
/**
- * dbg_dump_lpt_lebs - dump LPT lebs.
+ * ubifs_dump_lpt_lebs - dump LPT lebs.
* @c: UBIFS file-system description object
*
* This function dumps all LPT LEBs. The caller has to make sure the LPT is
* locked.
*/
-void dbg_dump_lpt_lebs(const struct ubifs_info *c)
+void ubifs_dump_lpt_lebs(const struct ubifs_info *c)
{
int i;
- printk(KERN_DEBUG "(pid %d) start dumping all LPT LEBs\n",
- current->pid);
+ pr_err("(pid %d) start dumping all LPT LEBs\n", current->pid);
for (i = 0; i < c->lpt_lebs; i++)
dump_lpt_leb(c, i + c->lpt_first);
- printk(KERN_DEBUG "(pid %d) finish dumping all LPT LEBs\n",
- current->pid);
+ pr_err("(pid %d) finish dumping all LPT LEBs\n", current->pid);
}
-#endif /* CONFIG_UBIFS_FS_DEBUG */
+/**
+ * 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 (!dbg_is_chk_gen(c))
+ return 0;
+ if (prandom_u32() & 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[prandom_u32() % c->lsave_cnt] = lprops->lnum;
+ list_for_each_entry(lprops, &c->freeable_list, list)
+ c->lsave[prandom_u32() % c->lsave_cnt] = lprops->lnum;
+ list_for_each_entry(lprops, &c->frdi_idx_list, list)
+ c->lsave[prandom_u32() % c->lsave_cnt] = lprops->lnum;
+
+ heap = &c->lpt_heap[LPROPS_DIRTY_IDX - 1];
+ for (i = 0; i < heap->cnt; i++)
+ c->lsave[prandom_u32() % c->lsave_cnt] = heap->arr[i]->lnum;
+ heap = &c->lpt_heap[LPROPS_DIRTY - 1];
+ for (i = 0; i < heap->cnt; i++)
+ c->lsave[prandom_u32() % c->lsave_cnt] = heap->arr[i]->lnum;
+ heap = &c->lpt_heap[LPROPS_FREE - 1];
+ for (i = 0; i < heap->cnt; i++)
+ c->lsave[prandom_u32() % c->lsave_cnt] = heap->arr[i]->lnum;
+
+ return 1;
+}
diff --git a/fs/ubifs/master.c b/fs/ubifs/master.c
index 21f47afdacf..ab83ace9910 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;
}
@@ -241,7 +241,7 @@ static int validate_master(const struct ubifs_info *c)
out:
ubifs_err("bad master node at offset %d error %d", c->mst_offs, err);
- dbg_dump_node(c, c->mst_node);
+ ubifs_dump_node(c, c->mst_node);
return -EINVAL;
}
@@ -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;
@@ -317,7 +317,7 @@ int ubifs_read_master(struct ubifs_info *c)
if (c->leb_cnt < old_leb_cnt ||
c->leb_cnt < UBIFS_MIN_LEB_CNT) {
ubifs_err("bad leb_cnt on master node");
- dbg_dump_node(c, c->mst_node);
+ ubifs_dump_node(c, c->mst_node);
return -EINVAL;
}
@@ -379,7 +379,7 @@ int ubifs_write_master(struct ubifs_info *c)
c->mst_offs = offs;
c->mst_node->highest_inum = cpu_to_le64(c->highest_inum);
- err = ubifs_write_node(c, c->mst_node, len, lnum, offs, UBI_SHORTTERM);
+ err = ubifs_write_node(c, c->mst_node, len, lnum, offs);
if (err)
return err;
@@ -390,7 +390,7 @@ int ubifs_write_master(struct ubifs_info *c)
if (err)
return err;
}
- err = ubifs_write_node(c, c->mst_node, len, lnum, offs, UBI_SHORTTERM);
+ err = ubifs_write_node(c, c->mst_node, len, lnum, offs);
return err;
}
diff --git a/fs/ubifs/misc.h b/fs/ubifs/misc.h
index c3de04dc952..ee7cb5ebb6e 100644
--- a/fs/ubifs/misc.h
+++ b/fs/ubifs/misc.h
@@ -39,6 +39,29 @@ static inline int ubifs_zn_dirty(const struct ubifs_znode *znode)
}
/**
+ * ubifs_zn_obsolete - check if znode is obsolete.
+ * @znode: znode to check
+ *
+ * This helper function returns %1 if @znode is obsolete and %0 otherwise.
+ */
+static inline int ubifs_zn_obsolete(const struct ubifs_znode *znode)
+{
+ return !!test_bit(OBSOLETE_ZNODE, &znode->flags);
+}
+
+/**
+ * ubifs_zn_cow - check if znode has to be copied on write.
+ * @znode: znode to check
+ *
+ * This helper function returns %1 if @znode is has COW flag set and %0
+ * otherwise.
+ */
+static inline int ubifs_zn_cow(const struct ubifs_znode *znode)
+{
+ return !!test_bit(COW_ZNODE, &znode->flags);
+}
+
+/**
* ubifs_wake_up_bgt - wake up background thread.
* @c: UBIFS file-system description object
*/
@@ -122,86 +145,6 @@ static inline int ubifs_wbuf_sync(struct ubifs_wbuf *wbuf)
}
/**
- * ubifs_leb_unmap - unmap an LEB.
- * @c: UBIFS file-system description object
- * @lnum: LEB number to unmap
- *
- * This function returns %0 on success and a negative error code on failure.
- */
-static inline int ubifs_leb_unmap(const struct ubifs_info *c, int lnum)
-{
- int err;
-
- ubifs_assert(!c->ro_media && !c->ro_mount);
- if (c->ro_error)
- return -EROFS;
- err = ubi_leb_unmap(c->ubi, lnum);
- if (err) {
- ubifs_err("unmap LEB %d failed, error %d", lnum, err);
- return err;
- }
-
- return 0;
-}
-
-/**
- * ubifs_leb_write - write to a LEB.
- * @c: UBIFS file-system description object
- * @lnum: LEB number to write
- * @buf: buffer to write from
- * @offs: offset within LEB to write to
- * @len: length to write
- * @dtype: data type
- *
- * This function returns %0 on success and a negative error code on failure.
- */
-static inline int ubifs_leb_write(const struct ubifs_info *c, int lnum,
- const void *buf, int offs, int len, int dtype)
-{
- int err;
-
- 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) {
- ubifs_err("writing %d bytes at %d:%d, error %d",
- len, lnum, offs, err);
- return err;
- }
-
- return 0;
-}
-
-/**
- * ubifs_leb_change - atomic LEB change.
- * @c: UBIFS file-system description object
- * @lnum: LEB number to write
- * @buf: buffer to write from
- * @len: length to write
- * @dtype: data type
- *
- * This function returns %0 on success and a negative error code on failure.
- */
-static inline int ubifs_leb_change(const struct ubifs_info *c, int lnum,
- const void *buf, int len, int dtype)
-{
- int err;
-
- 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) {
- ubifs_err("changing %d bytes in LEB %d, error %d",
- len, lnum, err);
- return err;
- }
-
- return 0;
-}
-
-/**
* ubifs_encode_dev - encode device node IDs.
* @dev: UBIFS device node information
* @rdev: device IDs to encode
@@ -340,4 +283,21 @@ 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 82009c74b6a..f1c3e5a1b31 100644
--- a/fs/ubifs/orphan.c
+++ b/fs/ubifs/orphan.c
@@ -52,11 +52,7 @@
* than the maximum number of orphans allowed.
*/
-#ifdef CONFIG_UBIFS_FS_DEBUG
static int dbg_check_orphans(struct ubifs_info *c);
-#else
-#define dbg_check_orphans(c) 0
-#endif
/**
* ubifs_add_orphan - add an orphan.
@@ -92,7 +88,7 @@ int ubifs_add_orphan(struct ubifs_info *c, ino_t inum)
else if (inum > o->inum)
p = &(*p)->rb_right;
else {
- dbg_err("orphaned twice");
+ ubifs_err("orphaned twice");
spin_unlock(&c->orphan_lock);
kfree(orphan);
return 0;
@@ -130,13 +126,14 @@ void ubifs_delete_orphan(struct ubifs_info *c, ino_t inum)
else if (inum > o->inum)
p = p->rb_right;
else {
- if (o->dnext) {
+ if (o->del) {
spin_unlock(&c->orphan_lock);
dbg_gen("deleted twice ino %lu",
(unsigned long)inum);
return;
}
- if (o->cnext) {
+ if (o->cmt) {
+ o->del = 1;
o->dnext = c->orph_dnext;
c->orph_dnext = o;
spin_unlock(&c->orphan_lock);
@@ -158,8 +155,8 @@ void ubifs_delete_orphan(struct ubifs_info *c, ino_t inum)
}
}
spin_unlock(&c->orphan_lock);
- dbg_err("missing orphan ino %lu", (unsigned long)inum);
- dbg_dump_stack();
+ ubifs_err("missing orphan ino %lu", (unsigned long)inum);
+ dump_stack();
}
/**
@@ -176,11 +173,13 @@ int ubifs_orphan_start_commit(struct ubifs_info *c)
last = &c->orph_cnext;
list_for_each_entry(orphan, &c->orph_new, new_list) {
ubifs_assert(orphan->new);
+ ubifs_assert(!orphan->cmt);
orphan->new = 0;
+ orphan->cmt = 1;
*last = orphan;
last = &orphan->cnext;
}
- *last = orphan->cnext;
+ *last = NULL;
c->cmt_orphans = c->new_orphans;
c->new_orphans = 0;
dbg_cmt("%d orphans to commit", c->cmt_orphans);
@@ -248,8 +247,7 @@ static int do_write_orph_node(struct ubifs_info *c, int len, int atomic)
ubifs_assert(c->ohead_offs == 0);
ubifs_prepare_node(c, c->orph_buf, len, 1);
len = ALIGN(len, c->min_io_size);
- err = ubifs_leb_change(c, c->ohead_lnum, c->orph_buf, len,
- UBI_SHORTTERM);
+ err = ubifs_leb_change(c, c->ohead_lnum, c->orph_buf, len);
} else {
if (c->ohead_offs == 0) {
/* Ensure LEB has been unmapped */
@@ -258,7 +256,7 @@ static int do_write_orph_node(struct ubifs_info *c, int len, int atomic)
return err;
}
err = ubifs_write_node(c, c->orph_buf, len, c->ohead_lnum,
- c->ohead_offs, UBI_SHORTTERM);
+ c->ohead_offs);
}
return err;
}
@@ -304,7 +302,9 @@ static int write_orph_node(struct ubifs_info *c, int atomic)
cnext = c->orph_cnext;
for (i = 0; i < cnt; i++) {
orphan = cnext;
+ ubifs_assert(orphan->cmt);
orph->inos[i] = cpu_to_le64(orphan->inum);
+ orphan->cmt = 0;
cnext = orphan->cnext;
orphan->cnext = NULL;
}
@@ -383,11 +383,12 @@ static int consolidate(struct ubifs_info *c)
list_for_each_entry(orphan, &c->orph_list, list) {
if (orphan->new)
continue;
+ orphan->cmt = 1;
*last = orphan;
last = &orphan->cnext;
cnt += 1;
}
- *last = orphan->cnext;
+ *last = NULL;
ubifs_assert(cnt == c->tot_orphans - c->new_orphans);
c->cmt_orphans = cnt;
c->ohead_lnum = c->orph_first;
@@ -447,6 +448,7 @@ static void erase_deleted(struct ubifs_info *c)
orphan = dnext;
dnext = orphan->dnext;
ubifs_assert(!orphan->new);
+ ubifs_assert(orphan->del);
rb_erase(&orphan->rb, &c->orph_tree);
list_del(&orphan->list);
c->tot_orphans -= 1;
@@ -536,6 +538,7 @@ static int insert_dead_orphan(struct ubifs_info *c, ino_t inum)
rb_link_node(&orphan->rb, parent, p);
rb_insert_color(&orphan->rb, &c->orph_tree);
list_add_tail(&orphan->list, &c->orph_list);
+ orphan->del = 1;
orphan->dnext = c->orph_dnext;
c->orph_dnext = orphan;
dbg_mnt("ino %lu, new %d, tot %d", (unsigned long)inum,
@@ -567,9 +570,9 @@ static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
list_for_each_entry(snod, &sleb->nodes, list) {
if (snod->type != UBIFS_ORPH_NODE) {
- ubifs_err("invalid node type %d in orphan area at "
- "%d:%d", snod->type, sleb->lnum, snod->offs);
- dbg_dump_node(c, snod->node);
+ ubifs_err("invalid node type %d in orphan area at %d:%d",
+ snod->type, sleb->lnum, snod->offs);
+ ubifs_dump_node(c, snod->node);
return -EINVAL;
}
@@ -594,10 +597,9 @@ static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
* number. That makes this orphan node, out of date.
*/
if (!first) {
- ubifs_err("out of order commit number %llu in "
- "orphan node at %d:%d",
+ ubifs_err("out of order commit number %llu in orphan node at %d:%d",
cmt_no, sleb->lnum, snod->offs);
- dbg_dump_node(c, snod->node);
+ ubifs_dump_node(c, snod->node);
return -EINVAL;
}
dbg_rcvry("out of date LEB %d", sleb->lnum);
@@ -673,7 +675,8 @@ 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;
@@ -724,7 +727,9 @@ int ubifs_mount_orphans(struct ubifs_info *c, int unclean, int read_only)
return err;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
+/*
+ * Everything below is related to debugging.
+ */
struct check_orphan {
struct rb_node rb;
@@ -810,27 +815,10 @@ static int dbg_find_check_orphan(struct rb_root *root, ino_t inum)
static void dbg_free_check_tree(struct rb_root *root)
{
- struct rb_node *this = root->rb_node;
- struct check_orphan *o;
+ struct check_orphan *o, *n;
- while (this) {
- if (this->rb_left) {
- this = this->rb_left;
- continue;
- } else if (this->rb_right) {
- this = this->rb_right;
- continue;
- }
- o = rb_entry(this, struct check_orphan, rb);
- this = rb_parent(this);
- if (this) {
- if (this->rb_left == &o->rb)
- this->rb_left = NULL;
- else
- this->rb_right = NULL;
- }
+ rbtree_postorder_for_each_entry_safe(o, n, root, rb)
kfree(o);
- }
}
static int dbg_orphan_check(struct ubifs_info *c, struct ubifs_zbranch *zbr,
@@ -892,15 +880,22 @@ 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 +907,7 @@ static int dbg_scan_orphans(struct ubifs_info *c, struct check_info *ci)
break;
}
+ vfree(buf);
return err;
}
@@ -920,7 +916,7 @@ static int dbg_check_orphans(struct ubifs_info *c)
struct check_info ci;
int err;
- if (!(ubifs_chk_flags & UBIFS_CHK_ORPH))
+ if (!dbg_is_chk_orph(c))
return 0;
ci.last_ino = 0;
@@ -959,5 +955,3 @@ out:
kfree(ci.node);
return err;
}
-
-#endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c
index 77e9b874b6c..c14adb2f420 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>
@@ -100,7 +117,7 @@ static int get_master_node(const struct ubifs_info *c, int lnum, void **pbuf,
if (!sbuf)
return -ENOMEM;
- err = ubi_read(c->ubi, lnum, sbuf, 0, c->leb_size);
+ err = ubifs_leb_read(c, lnum, sbuf, 0, c->leb_size, 0);
if (err && err != -EBADMSG)
goto out_free;
@@ -196,10 +213,10 @@ static int write_rcvrd_mst_node(struct ubifs_info *c,
mst->flags |= cpu_to_le32(UBIFS_MST_RCVRY);
ubifs_prepare_node(c, mst, UBIFS_MST_NODE_SZ, 1);
- err = ubi_leb_change(c->ubi, lnum, mst, sz, UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum, mst, sz);
if (err)
goto out;
- err = ubi_leb_change(c->ubi, lnum + 1, mst, sz, UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum + 1, mst, sz);
if (err)
goto out;
out:
@@ -257,7 +274,8 @@ int ubifs_recover_master_node(struct ubifs_info *c)
if (cor1)
goto out_err;
mst = mst1;
- } else if (offs1 == 0 && offs2 + sz >= c->leb_size) {
+ } else if (offs1 == 0 &&
+ c->leb_size - offs2 - sz < sz) {
/* 1st LEB was unmapped and written, 2nd not */
if (cor1)
goto out_err;
@@ -300,6 +318,32 @@ 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;
@@ -318,12 +362,12 @@ out_err:
out_free:
ubifs_err("failed to recover master node");
if (mst1) {
- dbg_err("dumping first master node");
- dbg_dump_node(c, mst1);
+ ubifs_err("dumping first master node");
+ ubifs_dump_node(c, mst1);
}
if (mst2) {
- dbg_err("dumping second master node");
- dbg_dump_node(c, mst2);
+ ubifs_err("dumping second master node");
+ ubifs_dump_node(c, mst2);
}
vfree(buf2);
vfree(buf1);
@@ -362,8 +406,9 @@ 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 +416,10 @@ 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 +474,7 @@ 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 +486,7 @@ 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;
@@ -495,8 +540,8 @@ static int fix_unclean_leb(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
int len = ALIGN(endpt, c->min_io_size);
if (start) {
- err = ubi_read(c->ubi, lnum, sleb->buf, 0,
- start);
+ err = ubifs_leb_read(c, lnum, sleb->buf, 0,
+ start, 1);
if (err)
return err;
}
@@ -510,8 +555,7 @@ static int fix_unclean_leb(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
ubifs_pad(c, buf, pad_len);
}
}
- err = ubi_leb_change(c->ubi, lnum, sleb->buf, len,
- UBI_UNKNOWN);
+ err = ubifs_leb_change(c, lnum, sleb->buf, len);
if (err)
return err;
}
@@ -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 +581,41 @@ 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 +624,8 @@ 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 +633,21 @@ 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 +657,7 @@ 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 +670,127 @@ 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) at %d:%d",
+ ret, lnum, offs);
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 {
+ ubifs_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);
- ubifs_err("corrupt empty space LEB %d:%d, corruption "
- "starts at %d", lnum, offs, corruption);
+ /*
+ * 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 */
+ ubifs_err("corruption %d", ret);
+ ubifs_scan_a_node(c, buf, len, lnum, offs, 1);
corrupted:
ubifs_scanned_corruption(c, lnum, offs, buf);
err = -EUCLEAN;
@@ -733,22 +821,23 @@ static int get_cs_sqnum(struct ubifs_info *c, int lnum, int offs,
return -ENOMEM;
if (c->leb_size - offs < UBIFS_CS_NODE_SZ)
goto out_err;
- err = ubi_read(c->ubi, lnum, (void *)cs_node, offs, UBIFS_CS_NODE_SZ);
+ err = ubifs_leb_read(c, lnum, (void *)cs_node, offs,
+ UBIFS_CS_NODE_SZ, 0);
if (err && err != -EBADMSG)
goto out_free;
ret = ubifs_scan_a_node(c, cs_node, UBIFS_CS_NODE_SZ, lnum, offs, 0);
if (ret != SCANNED_A_NODE) {
- dbg_err("Not a valid node");
+ ubifs_err("Not a valid node");
goto out_err;
}
if (cs_node->ch.node_type != UBIFS_CS_NODE) {
- dbg_err("Node a CS node, type is %d", cs_node->ch.node_type);
+ ubifs_err("Node a CS node, type is %d", cs_node->ch.node_type);
goto out_err;
}
if (le64_to_cpu(cs_node->cmt_no) != c->cmt_no) {
- dbg_err("CS node cmt_no %llu != current cmt_no %llu",
- (unsigned long long)le64_to_cpu(cs_node->cmt_no),
- c->cmt_no);
+ ubifs_err("CS node cmt_no %llu != current cmt_no %llu",
+ (unsigned long long)le64_to_cpu(cs_node->cmt_no),
+ c->cmt_no);
goto out_err;
}
*cs_sqnum = le64_to_cpu(cs_node->ch.sqnum);
@@ -811,15 +900,15 @@ struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum,
}
}
if (snod->sqnum > cs_sqnum) {
- ubifs_err("unrecoverable log corruption "
- "in LEB %d", lnum);
+ ubifs_err("unrecoverable log corruption in LEB %d",
+ lnum);
ubifs_scan_destroy(sleb);
return ERR_PTR(-EUCLEAN);
}
}
ubifs_scan_destroy(sleb);
}
- return ubifs_recover_leb(c, lnum, offs, sbuf, 0);
+ return ubifs_recover_leb(c, lnum, offs, sbuf, -1);
}
/**
@@ -833,15 +922,10 @@ struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum,
*
* This function returns %0 on success and a negative error code on failure.
*/
-static int recover_head(const struct ubifs_info *c, int lnum, int offs,
- void *sbuf)
+static int recover_head(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;
@@ -849,15 +933,15 @@ static int recover_head(const struct ubifs_info *c, int lnum, int offs,
return 0;
/* Read at the head location and check it is empty flash */
- err = ubi_read(c->ubi, lnum, sbuf, offs, len);
+ err = ubifs_leb_read(c, lnum, sbuf, offs, len, 1);
if (err || !is_empty(sbuf, len)) {
dbg_rcvry("cleaning head at %d:%d", lnum, offs);
if (offs == 0)
return ubifs_leb_unmap(c, lnum);
- err = ubi_read(c->ubi, lnum, sbuf, 0, offs);
+ err = ubifs_leb_read(c, lnum, sbuf, 0, offs, 1);
if (err)
return err;
- return ubi_leb_change(c->ubi, lnum, sbuf, offs, UBI_UNKNOWN);
+ return ubifs_leb_change(c, lnum, sbuf, offs);
}
return 0;
@@ -880,7 +964,7 @@ static int recover_head(const struct ubifs_info *c, int lnum, int offs,
*
* This function returns %0 on success and a negative error code on failure.
*/
-int ubifs_recover_inl_heads(const struct ubifs_info *c, void *sbuf)
+int ubifs_recover_inl_heads(struct ubifs_info *c, void *sbuf)
{
int err;
@@ -900,7 +984,7 @@ int ubifs_recover_inl_heads(const struct ubifs_info *c, void *sbuf)
}
/**
- * clean_an_unclean_leb - read and write a LEB to remove corruption.
+ * clean_an_unclean_leb - read and write a LEB to remove corruption.
* @c: UBIFS file-system description object
* @ucleb: unclean LEB information
* @sbuf: LEB-sized buffer to use
@@ -911,7 +995,7 @@ int ubifs_recover_inl_heads(const struct ubifs_info *c, void *sbuf)
*
* This function returns %0 on success and a negative error code on failure.
*/
-static int clean_an_unclean_leb(const struct ubifs_info *c,
+static int clean_an_unclean_leb(struct ubifs_info *c,
struct ubifs_unclean_leb *ucleb, void *sbuf)
{
int err, lnum = ucleb->lnum, offs = 0, len = ucleb->endpt, quiet = 1;
@@ -927,7 +1011,7 @@ static int clean_an_unclean_leb(const struct ubifs_info *c,
return 0;
}
- err = ubi_read(c->ubi, lnum, buf, offs, len);
+ err = ubifs_leb_read(c, lnum, buf, offs, len, 0);
if (err && err != -EBADMSG)
return err;
@@ -987,7 +1071,7 @@ static int clean_an_unclean_leb(const struct ubifs_info *c,
}
/* Write back the LEB atomically */
- err = ubi_leb_change(c->ubi, lnum, sbuf, len, UBI_UNKNOWN);
+ err = ubifs_leb_change(c, lnum, sbuf, len);
if (err)
return err;
@@ -1007,7 +1091,7 @@ static int clean_an_unclean_leb(const struct ubifs_info *c,
*
* This function returns %0 on success and a negative error code on failure.
*/
-int ubifs_clean_lebs(const struct ubifs_info *c, void *sbuf)
+int ubifs_clean_lebs(struct ubifs_info *c, void *sbuf)
{
dbg_rcvry("recovery");
while (!list_empty(&c->unclean_leb_list)) {
@@ -1026,6 +1110,53 @@ 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) {
+ ubifs_err("could not find an empty LEB");
+ ubifs_dump_lprops(c);
+ ubifs_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
*
@@ -1047,71 +1178,26 @@ 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.
@@ -1120,11 +1206,8 @@ 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) {
@@ -1135,42 +1218,22 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c)
}
mutex_unlock(&wbuf->io_mutex);
if (err < 0) {
- dbg_err("GC failed, error %d", err);
+ ubifs_err("GC failed, error %d", err);
if (err == -EAGAIN)
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;
}
/**
@@ -1272,29 +1335,14 @@ static void remove_ino(struct ubifs_info *c, ino_t inum)
*/
void ubifs_destroy_size_tree(struct ubifs_info *c)
{
- struct rb_node *this = c->size_tree.rb_node;
- struct size_entry *e;
+ struct size_entry *e, *n;
- while (this) {
- if (this->rb_left) {
- this = this->rb_left;
- continue;
- } else if (this->rb_right) {
- this = this->rb_right;
- continue;
- }
- e = rb_entry(this, struct size_entry, rb);
+ rbtree_postorder_for_each_entry_safe(e, n, &c->size_tree, rb) {
if (e->inode)
iput(e->inode);
- this = rb_parent(this);
- if (this) {
- if (this->rb_left == &e->rb)
- this->rb_left = NULL;
- else
- this->rb_right = NULL;
- }
kfree(e);
}
+
c->size_tree = RB_ROOT;
}
@@ -1393,7 +1441,7 @@ static int fix_size_in_place(struct ubifs_info *c, struct size_entry *e)
if (i_size >= e->d_size)
return 0;
/* Read the LEB */
- err = ubi_read(c->ubi, lnum, c->sbuf, 0, c->leb_size);
+ err = ubifs_leb_read(c, lnum, c->sbuf, 0, c->leb_size, 1);
if (err)
goto out;
/* Change the size field and recalculate the CRC */
@@ -1409,10 +1457,10 @@ static int fix_size_in_place(struct ubifs_info *c, struct size_entry *e)
len -= 1;
len = ALIGN(len + 1, c->min_io_size);
/* Atomically write the fixed LEB back again */
- err = ubi_leb_change(c->ubi, lnum, c->sbuf, len, UBI_UNKNOWN);
+ err = ubifs_leb_change(c, lnum, c->sbuf, len);
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;
@@ -1461,20 +1509,27 @@ 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->ro_mount) {
+ 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;
@@ -1489,9 +1544,11 @@ 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 eed0fcff8d7..3187925e987 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;
/*
@@ -149,22 +141,48 @@ static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r)
* during the replay.
*/
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);
+ dbg_mnt("LEB %d lp: %d free %d 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);
+
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 +218,22 @@ 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_mntk(&r->key, "LEB %d:%d len %d deletion %d sqnum %llu key ",
+ r->lnum, r->offs, r->len, r->deletion, r->sqnum);
/* 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 +256,7 @@ 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 +264,77 @@ 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 +346,25 @@ 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_mntk(key, "add LEB %d:%d, key ", lnum, offs);
+
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 +374,18 @@ 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 +397,25 @@ 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_mntk(key, "add LEB %d:%d, key ", lnum, offs);
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 +427,15 @@ 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 +472,90 @@ 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 = ubifs_leb_read(c, next->lnum, (char *)&data, next->start, 4, 1);
+ 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))
@@ -620,19 +671,14 @@ 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->ro_mount)
- 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);
@@ -640,61 +686,12 @@ out:
out_dump:
ubifs_err("bad node is at LEB %d:%d", lnum, snod->offs);
- dbg_dump_node(c, snod->node);
+ ubifs_dump_node(c, snod->node);
ubifs_scan_destroy(sleb);
return -EINVAL;
}
/**
- * 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
*
@@ -704,17 +701,16 @@ 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;
@@ -865,16 +861,15 @@ static int replay_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf)
* numbers.
*/
if (snod->type != UBIFS_CS_NODE) {
- dbg_err("first log node at LEB %d:%d is not CS node",
- lnum, offs);
+ ubifs_err("first log node at LEB %d:%d is not CS node",
+ lnum, offs);
goto out_dump;
}
if (le64_to_cpu(node->cmt_no) != c->cmt_no) {
- dbg_err("first CS node at LEB %d:%d has wrong "
- "commit number %llu expected %llu",
- lnum, offs,
- (unsigned long long)le64_to_cpu(node->cmt_no),
- c->cmt_no);
+ ubifs_err("first CS node at LEB %d:%d has wrong commit number %llu expected %llu",
+ lnum, offs,
+ (unsigned long long)le64_to_cpu(node->cmt_no),
+ c->cmt_no);
goto out_dump;
}
@@ -896,7 +891,7 @@ static int replay_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf)
/* Make sure the first node sits at offset zero of the LEB */
if (snod->offs != 0) {
- dbg_err("first node is not at zero offset");
+ ubifs_err("first node is not at zero offset");
goto out_dump;
}
@@ -909,8 +904,8 @@ static int replay_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf)
}
if (snod->sqnum < c->cs_sqnum) {
- dbg_err("bad sqnum %llu, commit sqnum %llu",
- snod->sqnum, c->cs_sqnum);
+ ubifs_err("bad sqnum %llu, commit sqnum %llu",
+ snod->sqnum, c->cs_sqnum);
goto out_dump;
}
@@ -962,7 +957,7 @@ out:
out_dump:
ubifs_err("log error detected while replaying the log at LEB %d:%d",
lnum, offs + snod->offs);
- dbg_dump_node(c, snod->node);
+ ubifs_dump_node(c, snod->node);
ubifs_scan_destroy(sleb);
return -EINVAL;
}
@@ -1012,7 +1007,7 @@ out:
*/
int ubifs_replay_journal(struct ubifs_info *c)
{
- int err, i, lnum, offs, free;
+ int err, lnum, free;
BUILD_BUG_ON(UBIFS_TRUN_KEY > 5);
@@ -1030,49 +1025,44 @@ int ubifs_replay_journal(struct ubifs_info *c)
dbg_mnt("start replaying the journal");
c->replaying = 1;
lnum = c->ltail_lnum = c->lhead_lnum;
- offs = c->lhead_offs;
- for (i = 0; i < c->log_lebs; i++, lnum++) {
- if (lnum >= UBIFS_LOG_LNUM + c->log_lebs) {
- /*
- * The log is logically circular, we reached the last
- * LEB, switch to the first one.
- */
- lnum = UBIFS_LOG_LNUM;
- offs = 0;
- }
- err = replay_log_leb(c, lnum, offs, c->sbuf);
+ do {
+ err = replay_log_leb(c, lnum, 0, c->sbuf);
if (err == 1)
/* We hit the end of the log */
break;
if (err)
goto out;
- offs = 0;
- }
+ lnum = ubifs_next_log_lnum(c, lnum);
+ } while (lnum != c->ltail_lnum);
err = replay_buds(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,
+ 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);
c->replaying = 0;
return err;
diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c
index bf31b4729e5..4c37607a958 100644
--- a/fs/ubifs/sb.c
+++ b/fs/ubifs/sb.c
@@ -130,7 +130,6 @@ static int create_default_filesystem(struct ubifs_info *c)
* orphan node.
*/
orph_lebs = UBIFS_MIN_ORPH_LEBS;
-#ifdef CONFIG_UBIFS_FS_DEBUG
if (c->leb_cnt - min_leb_cnt > 1)
/*
* For debugging purposes it is better to have at least 2
@@ -138,7 +137,6 @@ static int create_default_filesystem(struct ubifs_info *c)
* consolidations and would be stressed more.
*/
orph_lebs += 1;
-#endif
main_lebs = c->leb_cnt - UBIFS_SB_LEBS - UBIFS_MST_LEBS - log_lebs;
main_lebs -= orph_lebs;
@@ -196,7 +194,7 @@ static int create_default_filesystem(struct ubifs_info *c)
sup->rp_size = cpu_to_le64(tmp64);
sup->ro_compat_version = cpu_to_le32(UBIFS_RO_COMPAT_VERSION);
- err = ubifs_write_node(c, sup, UBIFS_SB_NODE_SZ, 0, 0, UBI_LONGTERM);
+ err = ubifs_write_node(c, sup, UBIFS_SB_NODE_SZ, 0, 0);
kfree(sup);
if (err)
return err;
@@ -247,19 +245,18 @@ static int create_default_filesystem(struct ubifs_info *c)
mst->total_dirty = cpu_to_le64(tmp64);
/* The indexing LEB does not contribute to dark space */
- tmp64 = (c->main_lebs - 1) * c->dark_wm;
+ tmp64 = ((long long)(c->main_lebs - 1) * c->dark_wm);
mst->total_dark = cpu_to_le64(tmp64);
mst->total_used = cpu_to_le64(UBIFS_INO_NODE_SZ);
- err = ubifs_write_node(c, mst, UBIFS_MST_NODE_SZ, UBIFS_MST_LNUM, 0,
- UBI_UNKNOWN);
+ err = ubifs_write_node(c, mst, UBIFS_MST_NODE_SZ, UBIFS_MST_LNUM, 0);
if (err) {
kfree(mst);
return err;
}
- err = ubifs_write_node(c, mst, UBIFS_MST_NODE_SZ, UBIFS_MST_LNUM + 1, 0,
- UBI_UNKNOWN);
+ err = ubifs_write_node(c, mst, UBIFS_MST_NODE_SZ, UBIFS_MST_LNUM + 1,
+ 0);
kfree(mst);
if (err)
return err;
@@ -282,8 +279,7 @@ static int create_default_filesystem(struct ubifs_info *c)
key_write_idx(c, &key, &br->key);
br->lnum = cpu_to_le32(main_first + DEFAULT_DATA_LEB);
br->len = cpu_to_le32(UBIFS_INO_NODE_SZ);
- err = ubifs_write_node(c, idx, tmp, main_first + DEFAULT_IDX_LEB, 0,
- UBI_UNKNOWN);
+ err = ubifs_write_node(c, idx, tmp, main_first + DEFAULT_IDX_LEB, 0);
kfree(idx);
if (err)
return err;
@@ -315,8 +311,7 @@ static int create_default_filesystem(struct ubifs_info *c)
ino->flags = cpu_to_le32(UBIFS_COMPR_FL);
err = ubifs_write_node(c, ino, UBIFS_INO_NODE_SZ,
- main_first + DEFAULT_DATA_LEB, 0,
- UBI_UNKNOWN);
+ main_first + DEFAULT_DATA_LEB, 0);
kfree(ino);
if (err)
return err;
@@ -335,8 +330,7 @@ static int create_default_filesystem(struct ubifs_info *c)
return -ENOMEM;
cs->ch.node_type = UBIFS_CS_NODE;
- err = ubifs_write_node(c, cs, UBIFS_CS_NODE_SZ, UBIFS_LOG_LNUM,
- 0, UBI_UNKNOWN);
+ err = ubifs_write_node(c, cs, UBIFS_CS_NODE_SZ, UBIFS_LOG_LNUM, 0);
kfree(cs);
ubifs_msg("default file-system created");
@@ -397,9 +391,8 @@ static int validate_sb(struct ubifs_info *c, struct ubifs_sb_node *sup)
min_leb_cnt += c->lpt_lebs + c->orph_lebs + c->jhead_cnt + 6;
if (c->leb_cnt < min_leb_cnt || c->leb_cnt > c->vi.size) {
- ubifs_err("bad LEB count: %d in superblock, %d on UBI volume, "
- "%d minimum required", c->leb_cnt, c->vi.size,
- min_leb_cnt);
+ ubifs_err("bad LEB count: %d in superblock, %d on UBI volume, %d minimum required",
+ c->leb_cnt, c->vi.size, min_leb_cnt);
goto failed;
}
@@ -410,13 +403,22 @@ static int validate_sb(struct ubifs_info *c, struct ubifs_sb_node *sup)
}
if (c->main_lebs < UBIFS_MIN_MAIN_LEBS) {
- err = 7;
+ ubifs_err("too few main LEBs count %d, must be at least %d",
+ c->main_lebs, UBIFS_MIN_MAIN_LEBS);
goto failed;
}
- if (c->max_bud_bytes < (long long)c->leb_size * UBIFS_MIN_BUD_LEBS ||
- c->max_bud_bytes > (long long)c->leb_size * c->main_lebs) {
- err = 8;
+ max_bytes = (long long)c->leb_size * UBIFS_MIN_BUD_LEBS;
+ if (c->max_bud_bytes < max_bytes) {
+ ubifs_err("too small journal (%lld bytes), must be at least %lld bytes",
+ c->max_bud_bytes, max_bytes);
+ goto failed;
+ }
+
+ max_bytes = (long long)c->leb_size * c->main_lebs;
+ if (c->max_bud_bytes > max_bytes) {
+ ubifs_err("too large journal size (%lld bytes), only %lld bytes available in the main area",
+ c->max_bud_bytes, max_bytes);
goto failed;
}
@@ -450,7 +452,6 @@ static int validate_sb(struct ubifs_info *c, struct ubifs_sb_node *sup)
goto failed;
}
- max_bytes = c->main_lebs * (long long)c->leb_size;
if (c->rp_size < 0 || max_bytes < c->rp_size) {
err = 14;
goto failed;
@@ -466,7 +467,7 @@ static int validate_sb(struct ubifs_info *c, struct ubifs_sb_node *sup)
failed:
ubifs_err("bad superblock, error %d", err);
- dbg_dump_node(c, sup);
+ ubifs_dump_node(c, sup);
return -EINVAL;
}
@@ -475,7 +476,8 @@ 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)
{
@@ -508,7 +510,7 @@ int ubifs_write_sb_node(struct ubifs_info *c, struct ubifs_sb_node *sup)
int len = ALIGN(UBIFS_SB_NODE_SZ, c->min_io_size);
ubifs_prepare_node(c, sup, UBIFS_SB_NODE_SZ, 1);
- return ubifs_leb_change(c, UBIFS_SB_LNUM, sup, len, UBI_LONGTERM);
+ return ubifs_leb_change(c, UBIFS_SB_LNUM, sup, len);
}
/**
@@ -545,10 +547,9 @@ int ubifs_read_superblock(struct ubifs_info *c)
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 "
- "w%d/r%d", c->fmt_version,
- c->ro_compat_version, UBIFS_FORMAT_VERSION,
+ ubifs_err("on-flash format version is w%d/r%d, but software only supports up to version w%d/r%d",
+ c->fmt_version, c->ro_compat_version,
+ UBIFS_FORMAT_VERSION,
UBIFS_RO_COMPAT_VERSION);
if (c->ro_compat_version <= UBIFS_RO_COMPAT_VERSION) {
ubifs_msg("only R/O mounting is possible");
@@ -607,8 +608,8 @@ int ubifs_read_superblock(struct ubifs_info *c)
c->fanout = le32_to_cpu(sup->fanout);
c->lsave_cnt = le32_to_cpu(sup->lsave_cnt);
c->rp_size = le64_to_cpu(sup->rp_size);
- c->rp_uid = le32_to_cpu(sup->rp_uid);
- c->rp_gid = le32_to_cpu(sup->rp_gid);
+ c->rp_uid = make_kuid(&init_user_ns, le32_to_cpu(sup->rp_uid));
+ c->rp_gid = make_kgid(&init_user_ns, le32_to_cpu(sup->rp_gid));
sup_flags = le32_to_cpu(sup->flags);
if (!c->mount_opts.override_compr)
c->default_compr = le16_to_cpu(sup->default_compr);
@@ -616,6 +617,7 @@ 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;
@@ -650,3 +652,156 @@ 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 ubifs_leb_unmap(c, lnum);
+ }
+
+ dbg_mnt("fixup LEB %d, data len %d", lnum, len);
+ err = ubifs_leb_read(c, lnum, c->sbuf, 0, len, 1);
+ if (err)
+ return err;
+
+ return ubifs_leb_change(c, lnum, c->sbuf, len);
+}
+
+/**
+ * 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 log head which contains the only a CS node at the
+ * beginning.
+ */
+ err = fixup_leb(c, c->lhead_lnum,
+ ALIGN(UBIFS_CS_NODE_SZ, c->min_io_size));
+ 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 3e1ee57dbea..58aa05df2bb 100644
--- a/fs/ubifs/scan.c
+++ b/fs/ubifs/scan.c
@@ -75,7 +75,7 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum,
magic = le32_to_cpu(ch->magic);
if (magic == 0xFFFFFFFF) {
- dbg_scan("hit empty space");
+ dbg_scan("hit empty space at LEB %d:%d", lnum, offs);
return SCANNED_EMPTY_SPACE;
}
@@ -85,7 +85,8 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum,
if (len < UBIFS_CH_SZ)
return SCANNED_GARBAGE;
- dbg_scan("scanning %s", dbg_ntype(ch->node_type));
+ dbg_scan("scanning %s at LEB %d:%d",
+ dbg_ntype(ch->node_type), lnum, offs);
if (ubifs_check_node(c, buf, lnum, offs, quiet, 1))
return SCANNED_A_CORRUPT_NODE;
@@ -101,7 +102,7 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum,
if (!quiet) {
ubifs_err("bad pad node at LEB %d:%d",
lnum, offs);
- dbg_dump_node(c, pad);
+ ubifs_dump_node(c, pad);
}
return SCANNED_A_BAD_PAD_NODE;
}
@@ -109,13 +110,13 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum,
/* Make the node pads to 8-byte boundary */
if ((node_len + pad_len) & 7) {
if (!quiet)
- dbg_err("bad padding length %d - %d",
- offs, offs + node_len + pad_len);
+ ubifs_err("bad padding length %d - %d",
+ offs, offs + node_len + pad_len);
return SCANNED_A_BAD_PAD_NODE;
}
- dbg_scan("%d bytes padded, offset now %d",
- pad_len, ALIGN(offs + node_len + pad_len, 8));
+ dbg_scan("%d bytes padded at LEB %d:%d, offset now %d", pad_len,
+ lnum, offs, ALIGN(offs + node_len + pad_len, 8));
return node_len + pad_len;
}
@@ -148,10 +149,10 @@ struct ubifs_scan_leb *ubifs_start_scan(const struct ubifs_info *c, int lnum,
INIT_LIST_HEAD(&sleb->nodes);
sleb->buf = sbuf;
- err = ubi_read(c->ubi, lnum, sbuf + offs, offs, c->leb_size - offs);
+ err = ubifs_leb_read(c, lnum, sbuf + offs, offs, c->leb_size - offs, 0);
if (err && err != -EBADMSG) {
- ubifs_err("cannot read %d bytes from LEB %d:%d,"
- " error %d", c->leb_size - offs, lnum, offs, err);
+ ubifs_err("cannot read %d bytes from LEB %d:%d, error %d",
+ c->leb_size - offs, lnum, offs, err);
kfree(sleb);
return ERR_PTR(err);
}
@@ -240,12 +241,10 @@ void ubifs_scanned_corruption(const struct ubifs_info *c, int lnum, int offs,
int len;
ubifs_err("corruption at LEB %d:%d", lnum, offs);
- if (dbg_failure_mode)
- return;
len = c->leb_size - offs;
if (len > 8192)
len = 8192;
- dbg_err("first %d bytes from LEB %d:%d", len, lnum, offs);
+ ubifs_err("first %d bytes from LEB %d:%d", len, lnum, offs);
print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 4, buf, len, 1);
}
@@ -300,16 +299,16 @@ struct ubifs_scan_leb *ubifs_scan(const struct ubifs_info *c, int lnum,
switch (ret) {
case SCANNED_GARBAGE:
- dbg_err("garbage");
+ ubifs_err("garbage");
goto corrupted;
case SCANNED_A_NODE:
break;
case SCANNED_A_CORRUPT_NODE:
case SCANNED_A_BAD_PAD_NODE:
- dbg_err("bad node");
+ ubifs_err("bad node");
goto corrupted;
default:
- dbg_err("unknown");
+ ubifs_err("unknown");
err = -EINVAL;
goto error;
}
@@ -328,7 +327,7 @@ 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 46961c00323..9a9fb94a41c 100644
--- a/fs/ubifs/shrinker.c
+++ b/fs/ubifs/shrinker.c
@@ -128,7 +128,6 @@ static int shrink_tnc(struct ubifs_info *c, int nr, int age, int *contention)
freed = ubifs_destroy_tnc_subtree(znode);
atomic_long_sub(freed, &ubifs_clean_zn_cnt);
atomic_long_sub(freed, &c->clean_zn_cnt);
- ubifs_assert(atomic_long_read(&c->clean_zn_cnt) >= 0);
total_freed += freed;
znode = zprev;
}
@@ -277,13 +276,25 @@ static int kick_a_thread(void)
return 0;
}
-int ubifs_shrinker(struct shrinker *shrink, int nr, gfp_t gfp_mask)
+unsigned long ubifs_shrink_count(struct shrinker *shrink,
+ struct shrink_control *sc)
{
- 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;
+}
+
+unsigned long ubifs_shrink_scan(struct shrinker *shrink,
+ struct shrink_control *sc)
+{
+ unsigned long nr = sc->nr_to_scan;
+ int contention = 0;
+ unsigned long freed;
+ long clean_zn_cnt = atomic_long_read(&ubifs_clean_zn_cnt);
if (!clean_zn_cnt) {
/*
@@ -311,10 +322,10 @@ int ubifs_shrinker(struct shrinker *shrink, int nr, gfp_t gfp_mask)
if (!freed && contention) {
dbg_tnc("freed nothing, but contention");
- return -1;
+ return SHRINK_STOP;
}
out:
- dbg_tnc("%d znodes were freed, requested %d", freed, nr);
+ dbg_tnc("%lu znodes were freed, requested %lu", freed, nr);
return freed;
}
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 6e11c2975dc..3904c8574ef 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -49,7 +49,8 @@ struct kmem_cache *ubifs_inode_slab;
/* UBIFS TNC shrinker description */
static struct shrinker ubifs_shrinker_info = {
- .shrink = ubifs_shrinker,
+ .scan_objects = ubifs_shrink_scan,
+ .count_objects = ubifs_shrink_count,
.seeks = DEFAULT_SEEKS,
};
@@ -85,16 +86,15 @@ static int validate_inode(struct ubifs_info *c, const struct inode *inode)
if (ui->data_len < 0 || ui->data_len > UBIFS_MAX_INO_DATA)
return 4;
- if (ui->xattr && (inode->i_mode & S_IFMT) != S_IFREG)
+ if (ui->xattr && !S_ISREG(inode->i_mode))
return 5;
if (!ubifs_compr_present(ui->compr_type)) {
- ubifs_warn("inode %lu uses '%s' compression, but it was not "
- "compiled in", inode->i_ino,
- ubifs_compr_name(ui->compr_type));
+ ubifs_warn("inode %lu uses '%s' compression, but it was not compiled in",
+ inode->i_ino, ubifs_compr_name(ui->compr_type));
}
- err = dbg_check_dir_size(c, inode);
+ err = dbg_check_dir(c, inode);
return err;
}
@@ -129,9 +129,9 @@ struct inode *ubifs_iget(struct super_block *sb, unsigned long inum)
goto out_ino;
inode->i_flags |= (S_NOCMTIME | S_NOATIME);
- inode->i_nlink = le32_to_cpu(ino->nlink);
- inode->i_uid = le32_to_cpu(ino->uid);
- inode->i_gid = le32_to_cpu(ino->gid);
+ set_nlink(inode, le32_to_cpu(ino->nlink));
+ i_uid_write(inode, le32_to_cpu(ino->uid));
+ i_gid_write(inode, le32_to_cpu(ino->gid));
inode->i_atime.tv_sec = (int64_t)le64_to_cpu(ino->atime_sec);
inode->i_atime.tv_nsec = le32_to_cpu(ino->atime_nsec);
inode->i_mtime.tv_sec = (int64_t)le64_to_cpu(ino->mtime_sec);
@@ -246,8 +246,8 @@ struct inode *ubifs_iget(struct super_block *sb, unsigned long inum)
out_invalid:
ubifs_err("inode %lu validation failed, error %d", inode->i_ino, err);
- dbg_dump_node(c, ino);
- dbg_dump_inode(c, inode);
+ ubifs_dump_node(c, ino);
+ ubifs_dump_inode(c, inode);
err = -EINVAL;
out_ino:
kfree(ino);
@@ -276,7 +276,6 @@ 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);
}
@@ -304,7 +303,7 @@ static int ubifs_write_inode(struct inode *inode, struct writeback_control *wbc)
mutex_lock(&ui->ui_mutex);
/*
* Due to races between write-back forced by budgeting
- * (see 'sync_some_inodes()') and pdflush write-back, the inode may
+ * (see 'sync_some_inodes()') and background write-back, the inode may
* have already been synchronized, do not do this again. This might
* also happen if it was synchronized in an VFS operation, e.g.
* 'ubifs_link()'.
@@ -352,7 +351,7 @@ static void ubifs_evict_inode(struct inode *inode)
dbg_gen("inode %lu, mode %#x", inode->i_ino, (int)inode->i_mode);
ubifs_assert(!atomic_read(&inode->i_count));
- truncate_inode_pages(&inode->i_data, 0);
+ truncate_inode_pages_final(&inode->i_data);
if (inode->i_nlink)
goto done;
@@ -375,14 +374,14 @@ 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);
+ clear_inode(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);
@@ -420,9 +419,9 @@ static int ubifs_statfs(struct dentry *dentry, struct kstatfs *buf)
return 0;
}
-static int ubifs_show_options(struct seq_file *s, struct vfsmount *mnt)
+static int ubifs_show_options(struct seq_file *s, struct dentry *root)
{
- struct ubifs_info *c = mnt->mnt_sb->s_fs_info;
+ struct ubifs_info *c = root->d_sb->s_fs_info;
if (c->mount_opts.unmount_mode == 2)
seq_printf(s, ",fast_unmount");
@@ -512,9 +511,12 @@ 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",
@@ -534,6 +536,18 @@ 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.
@@ -541,6 +555,10 @@ 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);
@@ -650,8 +668,8 @@ static int init_constants_sb(struct ubifs_info *c)
tmp = UBIFS_CS_NODE_SZ + UBIFS_REF_NODE_SZ * c->jhead_cnt;
tmp = ALIGN(tmp, c->min_io_size);
if (tmp > c->leb_size) {
- dbg_err("too small LEB size %d, at least %d needed",
- c->leb_size, tmp);
+ ubifs_err("too small LEB size %d, at least %d needed",
+ c->leb_size, tmp);
return -EINVAL;
}
@@ -665,8 +683,8 @@ static int init_constants_sb(struct ubifs_info *c)
tmp /= c->leb_size;
tmp += 1;
if (c->log_lebs < tmp) {
- dbg_err("too small log %d LEBs, required min. %d LEBs",
- c->log_lebs, tmp);
+ ubifs_err("too small log %d LEBs, required min. %d LEBs",
+ c->log_lebs, tmp);
return -EINVAL;
}
@@ -675,11 +693,11 @@ 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
@@ -723,7 +741,7 @@ 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);
/*
@@ -792,15 +810,15 @@ 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.
+ * Garbage Collector head 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;
}
@@ -842,7 +860,7 @@ static void free_orphans(struct ubifs_info *c)
orph = list_entry(c->orph_list.next, struct ubifs_orphan, list);
list_del(&orph->list);
kfree(orph);
- dbg_err("orphan list not empty at unmount");
+ ubifs_err("orphan list not empty at unmount");
}
vfree(c->orph_buf);
@@ -855,26 +873,10 @@ static void free_orphans(struct ubifs_info *c)
*/
static void free_buds(struct ubifs_info *c)
{
- struct rb_node *this = c->buds.rb_node;
- struct ubifs_bud *bud;
-
- while (this) {
- if (this->rb_left)
- this = this->rb_left;
- else if (this->rb_right)
- this = this->rb_right;
- else {
- bud = rb_entry(this, struct ubifs_bud, rb);
- this = rb_parent(this);
- if (this) {
- if (this->rb_left == &bud->rb)
- this->rb_left = NULL;
- else
- this->rb_right = NULL;
- }
- kfree(bud);
- }
- }
+ struct ubifs_bud *bud, *n;
+
+ rbtree_postorder_for_each_entry_safe(bud, n, &c->buds, rb)
+ kfree(bud);
}
/**
@@ -892,7 +894,7 @@ static int check_volume_empty(struct ubifs_info *c)
c->empty = 1;
for (lnum = 0; lnum < c->leb_cnt; lnum++) {
- err = ubi_is_mapped(c->ubi, lnum);
+ err = ubifs_is_mapped(c, lnum);
if (unlikely(err < 0))
return err;
if (err == 1) {
@@ -1043,8 +1045,8 @@ static int ubifs_parse_options(struct ubifs_info *c, char *options,
flag = parse_standard_option(p);
if (!flag) {
- ubifs_err("unrecognized mount option \"%s\" "
- "or missing value", p);
+ ubifs_err("unrecognized mount option \"%s\" or missing value",
+ p);
return -EINVAL;
}
sb->s_flags |= flag;
@@ -1106,8 +1108,8 @@ again:
}
/* Just disable bulk-read */
- ubifs_warn("Cannot allocate %d bytes of memory for bulk-read, "
- "disabling it", c->max_bu_buf_len);
+ ubifs_warn("cannot allocate %d bytes of memory for bulk-read, disabling it",
+ c->max_bu_buf_len);
c->mount_opts.bulk_read = 1;
c->bulk_read = 0;
return;
@@ -1125,9 +1127,9 @@ 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);
- dbg_dump_lprops(c);
+ ubifs_err("insufficient free space to mount in R/W mode");
+ ubifs_dump_budg(c, &c->bi);
+ ubifs_dump_lprops(c);
return -ENOSPC;
}
return 0;
@@ -1139,17 +1141,17 @@ static int check_free_space(struct ubifs_info *c)
*
* This function mounts UBIFS file system. Returns zero in case of success and
* a negative error code in case of failure.
- *
- * Note, the function does not de-allocate resources it it fails half way
- * through, and the caller has to do this instead.
*/
static int mount_ubifs(struct ubifs_info *c)
{
int err;
- long long x;
+ long long x, y;
size_t sz;
c->ro_mount = !!(c->vfs_sb->s_flags & MS_RDONLY);
+ /* Suppress error messages while probing if MS_SILENT is set */
+ c->probing = !!(c->vfs_sb->s_flags & MS_SILENT);
+
err = init_constants_early(c);
if (err)
return err;
@@ -1202,16 +1204,21 @@ 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)
goto out_free;
+ c->probing = 0;
+
/*
* Make sure the compressor which is set as default in the superblock
* or overridden by mount options is actually compiled in.
@@ -1235,12 +1242,12 @@ static int mount_ubifs(struct ubifs_info *c)
goto out_free;
}
+ 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) {
- err = alloc_wbufs(c);
- if (err)
- goto out_cbuf;
-
/* Create background thread */
c->bgt = kthread_create(ubifs_bg_thread, c, "%s", c->bgt_name);
if (IS_ERR(c->bgt)) {
@@ -1262,12 +1269,25 @@ 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 (!c->ro_mount) {
- err = ubifs_recover_inl_heads(c, c->sbuf);
- if (err)
- goto out_master;
- }
- } else if (!c->ro_mount) {
+ }
+
+ 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_lpt;
+ }
+
+ if (!c->ro_mount) {
/*
* Set the "dirty" flag so that if we reboot uncleanly we
* will notice this immediately on the next mount.
@@ -1275,14 +1295,10 @@ 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, !c->ro_mount);
- 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;
@@ -1291,7 +1307,7 @@ 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, c->ro_mount);
if (err)
@@ -1382,75 +1398,71 @@ 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 (c->ro_mount)
- ubifs_msg("mounted read-only");
+ ubifs_msg("mounted UBI device %d, volume %d, name \"%s\"%s",
+ c->vi.ubi_num, c->vi.vol_id, c->vi.name,
+ c->ro_mount ? ", R/O mode" : "");
x = (long long)c->main_lebs * c->leb_size;
- ubifs_msg("file system size: %lld bytes (%lld KiB, %lld MiB, %d "
- "LEBs)", x, x >> 10, x >> 20, c->main_lebs);
- x = (long long)c->log_lebs * c->leb_size + c->max_bud_bytes;
- ubifs_msg("journal size: %lld bytes (%lld KiB, %lld MiB, %d "
- "LEBs)", x, x >> 10, x >> 20, c->log_lebs + c->max_bud_cnt);
- ubifs_msg("media format: w%d/r%d (latest is w%d/r%d)",
+ y = (long long)c->log_lebs * c->leb_size + c->max_bud_bytes;
+ ubifs_msg("LEB size: %d bytes (%d KiB), min./max. I/O unit sizes: %d bytes/%d bytes",
+ c->leb_size, c->leb_size >> 10, c->min_io_size,
+ c->max_write_size);
+ ubifs_msg("FS size: %lld bytes (%lld MiB, %d LEBs), journal size %lld bytes (%lld MiB, %d LEBs)",
+ x, x >> 20, c->main_lebs,
+ y, y >> 20, c->log_lebs + c->max_bud_cnt);
+ ubifs_msg("reserved for root: %llu bytes (%llu KiB)",
+ c->report_rp_size, c->report_rp_size >> 10);
+ ubifs_msg("media format: w%d/r%d (latest is w%d/r%d), UUID %pUB%s",
c->fmt_version, c->ro_compat_version,
- UBIFS_FORMAT_VERSION, UBIFS_RO_COMPAT_VERSION);
- ubifs_msg("default compressor: %s", ubifs_compr_name(c->default_compr));
- ubifs_msg("reserved for root: %llu bytes (%llu KiB)",
- c->report_rp_size, c->report_rp_size >> 10);
-
- dbg_msg("compiled on: " __DATE__ " at " __TIME__);
- dbg_msg("min. I/O unit size: %d bytes", c->min_io_size);
- dbg_msg("LEB size: %d bytes (%d KiB)",
- c->leb_size, c->leb_size >> 10);
- dbg_msg("data journal heads: %d",
+ UBIFS_FORMAT_VERSION, UBIFS_RO_COMPAT_VERSION, c->uuid,
+ c->big_lpt ? ", big LPT model" : ", small LPT model");
+
+ dbg_gen("default compressor: %s", ubifs_compr_name(c->default_compr));
+ dbg_gen("data journal heads: %d",
c->jhead_cnt - NONDATA_JHEADS_CNT);
- dbg_msg("UUID: %pUB", c->uuid);
- dbg_msg("big_lpt %d", c->big_lpt);
- dbg_msg("log LEBs: %d (%d - %d)",
+ dbg_gen("log LEBs: %d (%d - %d)",
c->log_lebs, UBIFS_LOG_LNUM, c->log_last);
- dbg_msg("LPT area LEBs: %d (%d - %d)",
+ dbg_gen("LPT area LEBs: %d (%d - %d)",
c->lpt_lebs, c->lpt_first, c->lpt_last);
- dbg_msg("orphan area LEBs: %d (%d - %d)",
+ dbg_gen("orphan area LEBs: %d (%d - %d)",
c->orph_lebs, c->orph_first, c->orph_last);
- dbg_msg("main area LEBs: %d (%d - %d)",
+ dbg_gen("main area LEBs: %d (%d - %d)",
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);
- 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);
- dbg_msg("first main LEB: %d", c->main_first);
- dbg_msg("max. znode size %d", c->max_znode_sz);
- dbg_msg("max. index node size %d", c->max_idx_node_sz);
- dbg_msg("node sizes: data %zu, inode %zu, dentry %zu",
+ dbg_gen("index LEBs: %d", c->lst.idx_lebs);
+ dbg_gen("total index bytes: %lld (%lld KiB, %lld MiB)",
+ c->bi.old_idx_sz, c->bi.old_idx_sz >> 10,
+ c->bi.old_idx_sz >> 20);
+ dbg_gen("key hash type: %d", c->key_hash_type);
+ dbg_gen("tree fanout: %d", c->fanout);
+ dbg_gen("reserved GC LEB: %d", c->gc_lnum);
+ dbg_gen("max. znode size %d", c->max_znode_sz);
+ dbg_gen("max. index node size %d", c->max_idx_node_sz);
+ dbg_gen("node sizes: data %zu, inode %zu, dentry %zu",
UBIFS_DATA_NODE_SZ, UBIFS_INO_NODE_SZ, UBIFS_DENT_NODE_SZ);
- dbg_msg("node sizes: trun %zu, sb %zu, master %zu",
+ dbg_gen("node sizes: trun %zu, sb %zu, master %zu",
UBIFS_TRUN_NODE_SZ, UBIFS_SB_NODE_SZ, UBIFS_MST_NODE_SZ);
- dbg_msg("node sizes: ref %zu, cmt. start %zu, orph %zu",
+ dbg_gen("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("dead watermark: %d", c->dead_wm);
- dbg_msg("dark watermark: %d", c->dark_wm);
- dbg_msg("LEB overhead: %d", c->leb_overhead);
+ dbg_gen("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_gen("dead watermark: %d", c->dead_wm);
+ dbg_gen("dark watermark: %d", c->dark_wm);
+ dbg_gen("LEB overhead: %d", c->leb_overhead);
x = (long long)c->main_lebs * c->dark_wm;
- dbg_msg("max. dark space: %lld (%lld KiB, %lld MiB)",
+ dbg_gen("max. dark space: %lld (%lld KiB, %lld MiB)",
x, x >> 10, x >> 20);
- dbg_msg("maximum bud bytes: %lld (%lld KiB, %lld MiB)",
+ dbg_gen("maximum bud bytes: %lld (%lld KiB, %lld MiB)",
c->max_bud_bytes, c->max_bud_bytes >> 10,
c->max_bud_bytes >> 20);
- dbg_msg("BG commit bud bytes: %lld (%lld KiB, %lld MiB)",
+ dbg_gen("BG commit bud bytes: %lld (%lld KiB, %lld MiB)",
c->bg_bud_bytes, c->bg_bud_bytes >> 10,
c->bg_bud_bytes >> 20);
- dbg_msg("current bud bytes %lld (%lld KiB, %lld MiB)",
+ dbg_gen("current bud bytes %lld (%lld KiB, %lld MiB)",
c->bud_bytes, c->bud_bytes >> 10, c->bud_bytes >> 20);
- dbg_msg("max. seq. number: %llu", c->max_sqnum);
- dbg_msg("commit number: %llu", c->cmt_no);
+ dbg_gen("max. seq. number: %llu", c->max_sqnum);
+ dbg_gen("commit number: %llu", c->cmt_no);
return 0;
@@ -1474,6 +1486,7 @@ 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);
@@ -1512,6 +1525,7 @@ 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);
@@ -1533,17 +1547,22 @@ static int ubifs_remount_rw(struct ubifs_info *c)
if (c->rw_incompat) {
ubifs_err("the file-system is not R/W-compatible");
- ubifs_msg("on-flash format version is w%d/r%d, but software "
- "only supports up to version w%d/r%d", c->fmt_version,
- c->ro_compat_version, UBIFS_FORMAT_VERSION,
- UBIFS_RO_COMPAT_VERSION);
+ ubifs_msg("on-flash format version is w%d/r%d, but software only supports up to version w%d/r%d",
+ c->fmt_version, c->ro_compat_version,
+ UBIFS_FORMAT_VERSION, UBIFS_RO_COMPAT_VERSION);
return -EROFS;
}
mutex_lock(&c->umount_mutex);
dbg_save_space_info(c);
c->remounting_rw = 1;
- c->always_chk_crc = 1;
+ c->ro_mount = 0;
+
+ if (c->space_fixup) {
+ err = ubifs_fixup_free_space(c);
+ if (err)
+ goto out;
+ }
err = check_free_space(c);
if (err)
@@ -1559,6 +1578,7 @@ 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;
}
@@ -1598,16 +1618,16 @@ 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) {
+ err = -ENOMEM;
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)) {
@@ -1642,20 +1662,31 @@ 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);
}
- dbg_gen("re-mounted read-write");
- c->ro_mount = 0;
- 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) {
@@ -1663,11 +1694,12 @@ 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;
}
@@ -1704,9 +1736,10 @@ 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);
@@ -1730,10 +1763,11 @@ 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
@@ -1778,8 +1812,8 @@ static void ubifs_put_super(struct super_block *sb)
* next mount, so we just print a message and
* continue to unmount normally.
*/
- ubifs_err("failed to write master node, "
- "error %d", err);
+ 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 */
@@ -1791,7 +1825,6 @@ 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)
@@ -1799,6 +1832,7 @@ static int ubifs_remount_fs(struct super_block *sb, int *flags, char *data)
int err;
struct ubifs_info *c = sb->s_fs_info;
+ sync_filesystem(sb);
dbg_gen("old flags %#lx, new flags %#x", sb->s_flags, *flags);
err = ubifs_parse_options(c, data, 1);
@@ -1914,59 +1948,65 @@ 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;
}
/*
@@ -1979,7 +2019,6 @@ 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;
@@ -2016,15 +2055,15 @@ static int ubifs_fill_super(struct super_block *sb, void *data, int silent)
goto out_umount;
}
- sb->s_root = d_alloc_root(root);
- if (!sb->s_root)
- goto out_iput;
+ sb->s_root = d_make_root(root);
+ if (!sb->s_root) {
+ err = -ENOMEM;
+ goto out_umount;
+ }
mutex_unlock(&c->umount_mutex);
return 0;
-out_iput:
- iput(root);
out_umount:
ubifs_umount(c);
out_unlock:
@@ -2033,24 +2072,29 @@ 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 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;
@@ -2063,23 +2107,29 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags,
*/
ubi = open_ubi(name, UBI_READONLY);
if (IS_ERR(ubi)) {
- dbg_err("cannot open \"%s\", error %d",
- name, (int)PTR_ERR(ubi));
+ ubifs_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;
+ }
+
+ dbg_gen("opened ubi%d_%d", c->vi.ubi_num, c->vi.vol_id);
- sb = sget(fs_type, &sb_test, &set_anon_super, &vi.cdev);
+ sb = sget(fs_type, sb_test, sb_set, flags, 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 & MS_RDONLY) != c1->ro_mount) {
@@ -2087,12 +2137,6 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags,
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;
@@ -2112,12 +2156,20 @@ out_close:
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,
.mount = ubifs_mount,
- .kill_sb = kill_anon_super,
+ .kill_sb = kill_ubifs_super,
};
+MODULE_ALIAS_FS("ubifs");
/*
* Inode slab cache constructor.
@@ -2184,25 +2236,17 @@ static int __init ubifs_init(void)
* UBIFS_BLOCK_SIZE. It is assumed that both are powers of 2.
*/
if (PAGE_CACHE_SIZE < UBIFS_BLOCK_SIZE) {
- ubifs_err("VFS page cache size is %u bytes, but UBIFS requires"
- " at least 4096 bytes",
+ ubifs_err("VFS page cache size is %u bytes, but UBIFS requires at least 4096 bytes",
(unsigned int)PAGE_CACHE_SIZE);
return -EINVAL;
}
- err = register_filesystem(&ubifs_fs_type);
- if (err) {
- ubifs_err("cannot register file system, error %d", err);
- return err;
- }
-
- err = -ENOMEM;
ubifs_inode_slab = kmem_cache_create("ubifs_inode_slab",
sizeof(struct ubifs_inode), 0,
SLAB_MEM_SPREAD | SLAB_RECLAIM_ACCOUNT,
&inode_slab_ctor);
if (!ubifs_inode_slab)
- goto out_reg;
+ return -ENOMEM;
register_shrinker(&ubifs_shrinker_info);
@@ -2214,15 +2258,20 @@ static int __init ubifs_init(void)
if (err)
goto out_compr;
+ err = register_filesystem(&ubifs_fs_type);
+ if (err) {
+ ubifs_err("cannot register file system, error %d", err);
+ goto out_dbg;
+ }
return 0;
+out_dbg:
+ dbg_debugfs_exit();
out_compr:
ubifs_compressors_exit();
out_shrinker:
unregister_shrinker(&ubifs_shrinker_info);
kmem_cache_destroy(ubifs_inode_slab);
-out_reg:
- unregister_filesystem(&ubifs_fs_type);
return err;
}
/* late_initcall to let compressors initialize first */
@@ -2236,6 +2285,12 @@ static void __exit ubifs_exit(void)
dbg_debugfs_exit();
ubifs_compressors_exit();
unregister_shrinker(&ubifs_shrinker_info);
+
+ /*
+ * Make sure all delayed rcu free inodes are flushed before we
+ * destroy cache.
+ */
+ rcu_barrier();
kmem_cache_destroy(ubifs_inode_slab);
unregister_filesystem(&ubifs_fs_type);
}
diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c
index ad9cf013362..8a40cf9c02d 100644
--- a/fs/ubifs/tnc.c
+++ b/fs/ubifs/tnc.c
@@ -178,27 +178,11 @@ static int ins_clr_old_idx_znode(struct ubifs_info *c,
*/
void destroy_old_idx(struct ubifs_info *c)
{
- struct rb_node *this = c->old_idx.rb_node;
- struct ubifs_old_idx *old_idx;
+ struct ubifs_old_idx *old_idx, *n;
- while (this) {
- if (this->rb_left) {
- this = this->rb_left;
- continue;
- } else if (this->rb_right) {
- this = this->rb_right;
- continue;
- }
- old_idx = rb_entry(this, struct ubifs_old_idx, rb);
- this = rb_parent(this);
- if (this) {
- if (this->rb_left == &old_idx->rb)
- this->rb_left = NULL;
- else
- this->rb_right = NULL;
- }
+ rbtree_postorder_for_each_entry_safe(old_idx, n, &c->old_idx, rb)
kfree(old_idx);
- }
+
c->old_idx = RB_ROOT;
}
@@ -223,7 +207,7 @@ static struct ubifs_znode *copy_znode(struct ubifs_info *c,
__set_bit(DIRTY_ZNODE, &zn->flags);
__clear_bit(COW_ZNODE, &zn->flags);
- ubifs_assert(!test_bit(OBSOLETE_ZNODE, &znode->flags));
+ ubifs_assert(!ubifs_zn_obsolete(znode));
__set_bit(OBSOLETE_ZNODE, &znode->flags);
if (znode->level != 0) {
@@ -271,7 +255,7 @@ static struct ubifs_znode *dirty_cow_znode(struct ubifs_info *c,
struct ubifs_znode *zn;
int err;
- if (!test_bit(COW_ZNODE, &znode->flags)) {
+ if (!ubifs_zn_cow(znode)) {
/* znode is not being committed */
if (!test_and_set_bit(DIRTY_ZNODE, &znode->flags)) {
atomic_long_inc(&c->dirty_zn_cnt);
@@ -339,17 +323,16 @@ static int lnc_add(struct ubifs_info *c, struct ubifs_zbranch *zbr,
err = ubifs_validate_entry(c, dent);
if (err) {
- dbg_dump_stack();
- dbg_dump_node(c, dent);
+ dump_stack();
+ ubifs_dump_node(c, dent);
return err;
}
- lnc_node = kmalloc(zbr->len, GFP_NOFS);
+ lnc_node = kmemdup(node, zbr->len, GFP_NOFS);
if (!lnc_node)
/* We don't have to have the cache, so no error */
return 0;
- memcpy(lnc_node, node, zbr->len);
zbr->leaf = lnc_node;
return 0;
}
@@ -373,8 +356,8 @@ static int lnc_add_directly(struct ubifs_info *c, struct ubifs_zbranch *zbr,
err = ubifs_validate_entry(c, node);
if (err) {
- dbg_dump_stack();
- dbg_dump_node(c, node);
+ dump_stack();
+ ubifs_dump_node(c, node);
return err;
}
@@ -447,8 +430,11 @@ 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)
@@ -459,7 +445,7 @@ static int try_read_node(const struct ubifs_info *c, void *buf, int type,
dbg_io("LEB %d:%d, %s, length %d", lnum, offs, dbg_ntype(type), len);
- err = ubi_read(c->ubi, lnum, buf, offs, len);
+ err = ubifs_leb_read(c, lnum, buf, offs, len, 1);
if (err) {
ubifs_err("cannot read node type %d from LEB %d:%d, error %d",
type, lnum, offs, err);
@@ -476,7 +462,8 @@ 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);
@@ -502,7 +489,7 @@ static int fallible_read_node(struct ubifs_info *c, const union ubifs_key *key,
{
int ret;
- dbg_tnc("LEB %d:%d, key %s", zbr->lnum, zbr->offs, DBGKEY(key));
+ dbg_tnck(key, "LEB %d:%d, key ", zbr->lnum, zbr->offs);
ret = try_read_node(c, node, key_type(c, key), zbr->len, zbr->lnum,
zbr->offs);
@@ -516,8 +503,8 @@ static int fallible_read_node(struct ubifs_info *c, const union ubifs_key *key,
ret = 0;
}
if (ret == 0 && c->replaying)
- dbg_mnt("dangling branch LEB %d:%d len %d, key %s",
- zbr->lnum, zbr->offs, zbr->len, DBGKEY(key));
+ dbg_mntk(key, "dangling branch LEB %d:%d len %d, key ",
+ zbr->lnum, zbr->offs, zbr->len);
return ret;
}
@@ -992,9 +979,9 @@ static int fallible_resolve_collision(struct ubifs_info *c,
if (adding || !o_znode)
return 0;
- dbg_mnt("dangling match LEB %d:%d len %d %s",
+ dbg_mntk(key, "dangling match LEB %d:%d len %d key ",
o_znode->zbranch[o_n].lnum, o_znode->zbranch[o_n].offs,
- o_znode->zbranch[o_n].len, DBGKEY(key));
+ o_znode->zbranch[o_n].len);
*zn = o_znode;
*n = o_n;
return 1;
@@ -1176,7 +1163,7 @@ int ubifs_lookup_level0(struct ubifs_info *c, const union ubifs_key *key,
struct ubifs_znode *znode;
unsigned long time = get_seconds();
- dbg_tnc("search key %s", DBGKEY(key));
+ dbg_tnck(key, "search key ");
ubifs_assert(key_type(c, key) < UBIFS_INVALID_KEY);
znode = c->zroot.znode;
@@ -1312,7 +1299,7 @@ static int lookup_level0_dirty(struct ubifs_info *c, const union ubifs_key *key,
struct ubifs_znode *znode;
unsigned long time = get_seconds();
- dbg_tnc("search and dirty key %s", DBGKEY(key));
+ dbg_tnck(key, "search and dirty key ");
znode = c->zroot.znode;
if (unlikely(!znode)) {
@@ -1662,7 +1649,7 @@ static int read_wbuf(struct ubifs_wbuf *wbuf, void *buf, int len, int lnum,
if (!overlap) {
/* We may safely unlock the write-buffer and read the data */
spin_unlock(&wbuf->lock);
- return ubi_read(c->ubi, lnum, buf, offs, len);
+ return ubifs_leb_read(c, lnum, buf, offs, len, 0);
}
/* Don't read under wbuf */
@@ -1676,7 +1663,7 @@ static int read_wbuf(struct ubifs_wbuf *wbuf, void *buf, int len, int lnum,
if (rlen > 0)
/* Read everything that goes before write-buffer */
- return ubi_read(c->ubi, lnum, buf, offs, rlen);
+ return ubifs_leb_read(c, lnum, buf, offs, rlen, 0);
return 0;
}
@@ -1719,8 +1706,8 @@ static int validate_data_node(struct ubifs_info *c, void *buf,
if (!keys_eq(c, &zbr->key, &key1)) {
ubifs_err("bad key in node at LEB %d:%d",
zbr->lnum, zbr->offs);
- dbg_tnc("looked for key %s found node's key %s",
- DBGKEY(&zbr->key), DBGKEY1(&key1));
+ dbg_tnck(&zbr->key, "looked for key ");
+ dbg_tnck(&key1, "found node's key ");
goto out_err;
}
@@ -1730,8 +1717,8 @@ out_err:
err = -EINVAL;
out:
ubifs_err("bad node at LEB %d:%d", zbr->lnum, zbr->offs);
- dbg_dump_node(c, buf);
- dbg_dump_stack();
+ ubifs_dump_node(c, buf);
+ dump_stack();
return err;
}
@@ -1763,7 +1750,7 @@ int ubifs_tnc_bulk_read(struct ubifs_info *c, struct bu_info *bu)
if (wbuf)
err = read_wbuf(wbuf, bu->buf, len, lnum, offs);
else
- err = ubi_read(c->ubi, lnum, bu->buf, offs, len);
+ err = ubifs_leb_read(c, lnum, bu->buf, offs, len, 0);
/* Check for a race with GC */
if (maybe_leb_gced(c, lnum, bu->gc_seq))
@@ -1772,8 +1759,8 @@ int ubifs_tnc_bulk_read(struct ubifs_info *c, struct bu_info *bu)
if (err && err != -EBADMSG) {
ubifs_err("failed to read from LEB %d:%d, error %d",
lnum, offs, err);
- dbg_dump_stack();
- dbg_tnc("key %s", DBGKEY(&bu->key));
+ dump_stack();
+ dbg_tnck(&bu->key, "key ");
return err;
}
@@ -1808,7 +1795,7 @@ static int do_lookup_nm(struct ubifs_info *c, const union ubifs_key *key,
int found, n, err;
struct ubifs_znode *znode;
- dbg_tnc("name '%.*s' key %s", nm->len, nm->name, DBGKEY(key));
+ dbg_tnck(key, "name '%.*s' key ", nm->len, nm->name);
mutex_lock(&c->tnc_mutex);
found = ubifs_lookup_level0(c, key, &znode, &n);
if (!found) {
@@ -1982,8 +1969,7 @@ again:
zp = znode->parent;
if (znode->child_cnt < c->fanout) {
ubifs_assert(n != c->fanout);
- dbg_tnc("inserted at %d level %d, key %s", n, znode->level,
- DBGKEY(key));
+ dbg_tnck(key, "inserted at %d level %d, key ", n, znode->level);
insert_zbranch(znode, zbr, n);
@@ -1998,7 +1984,7 @@ again:
* Unfortunately, @znode does not have more empty slots and we have to
* split it.
*/
- dbg_tnc("splitting level %d, key %s", znode->level, DBGKEY(key));
+ dbg_tnck(key, "splitting level %d, key ", znode->level);
if (znode->alt)
/*
@@ -2092,7 +2078,7 @@ do_split:
}
/* Insert new key and branch */
- dbg_tnc("inserting at %d level %d, key %s", n, zn->level, DBGKEY(key));
+ dbg_tnck(key, "inserting at %d level %d, key ", n, zn->level);
insert_zbranch(zi, zbr, n);
@@ -2168,7 +2154,7 @@ int ubifs_tnc_add(struct ubifs_info *c, const union ubifs_key *key, int lnum,
struct ubifs_znode *znode;
mutex_lock(&c->tnc_mutex);
- dbg_tnc("%d:%d, len %d, key %s", lnum, offs, len, DBGKEY(key));
+ dbg_tnck(key, "%d:%d, len %d, key ", lnum, offs, len);
found = lookup_level0_dirty(c, key, &znode, &n);
if (!found) {
struct ubifs_zbranch zbr;
@@ -2217,8 +2203,8 @@ int ubifs_tnc_replace(struct ubifs_info *c, const union ubifs_key *key,
struct ubifs_znode *znode;
mutex_lock(&c->tnc_mutex);
- dbg_tnc("old LEB %d:%d, new LEB %d:%d, len %d, key %s", old_lnum,
- old_offs, lnum, offs, len, DBGKEY(key));
+ dbg_tnck(key, "old LEB %d:%d, new LEB %d:%d, len %d, key ", old_lnum,
+ old_offs, lnum, offs, len);
found = lookup_level0_dirty(c, key, &znode, &n);
if (found < 0) {
err = found;
@@ -2300,8 +2286,8 @@ int ubifs_tnc_add_nm(struct ubifs_info *c, const union ubifs_key *key,
struct ubifs_znode *znode;
mutex_lock(&c->tnc_mutex);
- dbg_tnc("LEB %d:%d, name '%.*s', key %s", lnum, offs, nm->len, nm->name,
- DBGKEY(key));
+ dbg_tnck(key, "LEB %d:%d, name '%.*s', key ",
+ lnum, offs, nm->len, nm->name);
found = lookup_level0_dirty(c, key, &znode, &n);
if (found < 0) {
err = found;
@@ -2359,7 +2345,7 @@ int ubifs_tnc_add_nm(struct ubifs_info *c, const union ubifs_key *key,
* by passing 'ubifs_tnc_remove_nm()' the same key but
* an unmatchable name.
*/
- struct qstr noname = { .len = 0, .name = "" };
+ struct qstr noname = { .name = "" };
err = dbg_check_tnc(c, 0);
mutex_unlock(&c->tnc_mutex);
@@ -2394,14 +2380,14 @@ static int tnc_delete(struct ubifs_info *c, struct ubifs_znode *znode, int n)
/* Delete without merge for now */
ubifs_assert(znode->level == 0);
ubifs_assert(n >= 0 && n < c->fanout);
- dbg_tnc("deleting %s", DBGKEY(&znode->zbranch[n].key));
+ dbg_tnck(&znode->zbranch[n].key, "deleting key ");
zbr = &znode->zbranch[n];
lnc_free(zbr);
err = ubifs_add_dirt(c, zbr->lnum, zbr->len);
if (err) {
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
return err;
}
@@ -2419,7 +2405,7 @@ static int tnc_delete(struct ubifs_info *c, struct ubifs_znode *znode, int n)
*/
do {
- ubifs_assert(!test_bit(OBSOLETE_ZNODE, &znode->flags));
+ ubifs_assert(!ubifs_zn_obsolete(znode));
ubifs_assert(ubifs_zn_dirty(znode));
zp = znode->parent;
@@ -2475,9 +2461,8 @@ static int tnc_delete(struct ubifs_info *c, struct ubifs_znode *znode, int n)
c->zroot.offs = zbr->offs;
c->zroot.len = zbr->len;
c->zroot.znode = znode;
- ubifs_assert(!test_bit(OBSOLETE_ZNODE,
- &zp->flags));
- ubifs_assert(test_bit(DIRTY_ZNODE, &zp->flags));
+ ubifs_assert(!ubifs_zn_obsolete(zp));
+ ubifs_assert(ubifs_zn_dirty(zp));
atomic_long_dec(&c->dirty_zn_cnt);
if (zp->cnext) {
@@ -2505,7 +2490,7 @@ int ubifs_tnc_remove(struct ubifs_info *c, const union ubifs_key *key)
struct ubifs_znode *znode;
mutex_lock(&c->tnc_mutex);
- dbg_tnc("key %s", DBGKEY(key));
+ dbg_tnck(key, "key ");
found = lookup_level0_dirty(c, key, &znode, &n);
if (found < 0) {
err = found;
@@ -2536,7 +2521,7 @@ int ubifs_tnc_remove_nm(struct ubifs_info *c, const union ubifs_key *key,
struct ubifs_znode *znode;
mutex_lock(&c->tnc_mutex);
- dbg_tnc("%.*s, key %s", nm->len, nm->name, DBGKEY(key));
+ dbg_tnck(key, "%.*s, key ", nm->len, nm->name);
err = lookup_level0_dirty(c, key, &znode, &n);
if (err < 0)
goto out_unlock;
@@ -2553,11 +2538,11 @@ 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);
}
@@ -2648,10 +2633,10 @@ int ubifs_tnc_remove_range(struct ubifs_info *c, union ubifs_key *from_key,
err = ubifs_add_dirt(c, znode->zbranch[i].lnum,
znode->zbranch[i].len);
if (err) {
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
goto out_unlock;
}
- dbg_tnc("removing %s", DBGKEY(key));
+ dbg_tnck(key, "removing key ");
}
if (k) {
for (i = n + 1 + k; i < znode->child_cnt; i++)
@@ -2771,7 +2756,7 @@ struct ubifs_dent_node *ubifs_tnc_next_ent(struct ubifs_info *c,
struct ubifs_zbranch *zbr;
union ubifs_key *dkey;
- dbg_tnc("%s %s", nm->name ? (char *)nm->name : "(lowest)", DBGKEY(key));
+ dbg_tnck(key, "%s ", nm->name ? (char *)nm->name : "(lowest)");
ubifs_assert(is_hash_key(c, key));
mutex_lock(&c->tnc_mutex);
@@ -2861,7 +2846,7 @@ static void tnc_destroy_cnext(struct ubifs_info *c)
struct ubifs_znode *znode = cnext;
cnext = cnext->cnext;
- if (test_bit(OBSOLETE_ZNODE, &znode->flags))
+ if (ubifs_zn_obsolete(znode))
kfree(znode);
} while (cnext && cnext != c->cnext);
}
@@ -2872,12 +2857,14 @@ 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, freed;
+
+ n = atomic_long_read(&c->clean_zn_cnt);
+ freed = ubifs_destroy_tnc_subtree(c->zroot.znode);
+ ubifs_assert(freed == n);
+ atomic_long_sub(n, &ubifs_clean_zn_cnt);
}
kfree(c->gap_lebs);
kfree(c->ilebs);
@@ -3273,8 +3260,6 @@ out_unlock:
return err;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
-
/**
* dbg_check_inode_size - check if inode size is correct.
* @c: UBIFS file-system description object
@@ -3296,7 +3281,7 @@ int dbg_check_inode_size(struct ubifs_info *c, const struct inode *inode,
if (!S_ISREG(inode->i_mode))
return 0;
- if (!(ubifs_chk_flags & UBIFS_CHK_GEN))
+ if (!dbg_is_chk_gen(c))
return 0;
block = (size + UBIFS_BLOCK_SIZE - 1) >> UBIFS_BLOCK_SHIFT;
@@ -3329,16 +3314,15 @@ int dbg_check_inode_size(struct ubifs_info *c, const struct inode *inode,
out_dump:
block = key_block(c, key);
- ubifs_err("inode %lu has size %lld, but there are data at offset %lld "
- "(data key %s)", (unsigned long)inode->i_ino, size,
- ((loff_t)block) << UBIFS_BLOCK_SHIFT, DBGKEY(key));
- dbg_dump_inode(c, inode);
- dbg_dump_stack();
- err = -EINVAL;
+ ubifs_err("inode %lu has size %lld, but there are data at offset %lld",
+ (unsigned long)inode->i_ino, size,
+ ((loff_t)block) << UBIFS_BLOCK_SHIFT);
+ mutex_unlock(&c->tnc_mutex);
+ ubifs_dump_inode(c, inode);
+ dump_stack();
+ return -EINVAL;
out_unlock:
mutex_unlock(&c->tnc_mutex);
return err;
}
-
-#endif /* CONFIG_UBIFS_FS_DEBUG */
diff --git a/fs/ubifs/tnc_commit.c b/fs/ubifs/tnc_commit.c
index 53288e5d604..3600994f841 100644
--- a/fs/ubifs/tnc_commit.c
+++ b/fs/ubifs/tnc_commit.c
@@ -22,6 +22,7 @@
/* This file implements TNC functions for committing */
+#include <linux/random.h>
#include "ubifs.h"
/**
@@ -53,18 +54,16 @@ static int make_idx_node(struct ubifs_info *c, struct ubifs_idx_node *idx,
br->len = cpu_to_le32(zbr->len);
if (!zbr->lnum || !zbr->len) {
ubifs_err("bad ref in znode");
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
if (zbr->znode)
- dbg_dump_znode(c, zbr->znode);
+ ubifs_dump_znode(c, zbr->znode);
}
}
ubifs_prepare_node(c, idx, len, 0);
-#ifdef CONFIG_UBIFS_FS_DEBUG
znode->lnum = lnum;
znode->offs = offs;
znode->len = len;
-#endif
err = insert_old_idx_znode(c, znode);
@@ -87,8 +86,12 @@ static int make_idx_node(struct ubifs_info *c, struct ubifs_idx_node *idx,
atomic_long_dec(&c->dirty_zn_cnt);
ubifs_assert(ubifs_zn_dirty(znode));
- ubifs_assert(test_bit(COW_ZNODE, &znode->flags));
+ ubifs_assert(ubifs_zn_cow(znode));
+ /*
+ * Note, unlike 'write_index()' we do not add memory barriers here
+ * because this function is called with @c->tnc_mutex locked.
+ */
__clear_bit(DIRTY_ZNODE, &znode->flags);
__clear_bit(COW_ZNODE, &znode->flags);
@@ -317,8 +320,7 @@ static int layout_leb_in_gaps(struct ubifs_info *c, int *p)
0, 0, 0);
if (err)
return err;
- err = ubifs_leb_change(c, lnum, c->ileb_buf, c->ileb_len,
- UBI_SHORTTERM);
+ err = ubifs_leb_change(c, lnum, c->ileb_buf, c->ileb_len);
if (err)
return err;
dbg_gc("LEB %d wrote %d index nodes", lnum, tot_written);
@@ -377,16 +379,14 @@ 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_is_chk_index(c)) {
/*
* 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);
- dbg_dump_lprops(c);
+ ubifs_warn("out of space");
+ ubifs_dump_budg(c, &c->bi);
+ ubifs_dump_lprops(c);
}
/* Try to commit anyway */
err = 0;
@@ -453,11 +453,9 @@ static int layout_in_empty_space(struct ubifs_info *c)
offs = buf_offs + used;
-#ifdef CONFIG_UBIFS_FS_DEBUG
znode->lnum = lnum;
znode->offs = offs;
znode->len = len;
-#endif
/* Update the parent */
zp = znode->parent;
@@ -493,25 +491,6 @@ static int layout_in_empty_space(struct ubifs_info *c)
else
next_len = ubifs_idx_node_sz(c, cnext->child_cnt);
- if (c->min_io_size == 1) {
- buf_offs += ALIGN(len, 8);
- if (next_len) {
- if (buf_offs + next_len <= c->leb_size)
- continue;
- err = ubifs_update_one_lp(c, lnum, 0,
- c->leb_size - buf_offs, 0, 0);
- if (err)
- return err;
- lnum = -1;
- continue;
- }
- err = ubifs_update_one_lp(c, lnum,
- c->leb_size - buf_offs, 0, 0, 0);
- if (err)
- return err;
- break;
- }
-
/* Update buffer positions */
wlen = used + len;
used += ALIGN(len, 8);
@@ -552,10 +531,8 @@ static int layout_in_empty_space(struct ubifs_info *c)
break;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
c->dbg->new_ihead_lnum = lnum;
c->dbg->new_ihead_offs = buf_offs;
-#endif
return 0;
}
@@ -660,7 +637,7 @@ static int get_znodes_to_commit(struct ubifs_info *c)
}
cnt += 1;
while (1) {
- ubifs_assert(!test_bit(COW_ZNODE, &znode->flags));
+ ubifs_assert(!ubifs_zn_cow(znode));
__set_bit(COW_ZNODE, &znode->flags);
znode->alt = 0;
cnext = find_next_dirty(znode);
@@ -706,7 +683,7 @@ static int alloc_idx_lebs(struct ubifs_info *c, int cnt)
c->ilebs[c->ileb_cnt++] = lnum;
dbg_cmt("LEB %d", lnum);
}
- if (dbg_force_in_the_gaps())
+ if (dbg_is_chk_index(c) && !(prandom_u32() & 7))
return -ENOSPC;
return 0;
}
@@ -796,16 +773,16 @@ 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);
@@ -832,7 +809,7 @@ static int write_index(struct ubifs_info *c)
struct ubifs_idx_node *idx;
struct ubifs_znode *znode, *cnext;
int i, lnum, offs, len, next_len, buf_len, buf_offs, used;
- int avail, wlen, err, lnum_pos = 0;
+ int avail, wlen, err, lnum_pos = 0, blen, nxt_offs;
cnext = c->enext;
if (!cnext)
@@ -880,9 +857,9 @@ static int write_index(struct ubifs_info *c)
br->len = cpu_to_le32(zbr->len);
if (!zbr->lnum || !zbr->len) {
ubifs_err("bad ref in znode");
- dbg_dump_znode(c, znode);
+ ubifs_dump_znode(c, znode);
if (zbr->znode)
- dbg_dump_znode(c, zbr->znode);
+ ubifs_dump_znode(c, zbr->znode);
}
}
len = ubifs_idx_node_sz(c, znode->child_cnt);
@@ -897,19 +874,17 @@ static int write_index(struct ubifs_info *c)
}
offs = buf_offs + used;
-#ifdef CONFIG_UBIFS_FS_DEBUG
if (lnum != znode->lnum || offs != znode->offs ||
len != znode->len) {
ubifs_err("inconsistent znode posn");
return -EINVAL;
}
-#endif
/* Grab some stuff from znode while we still can */
cnext = znode->cnext;
ubifs_assert(ubifs_zn_dirty(znode));
- ubifs_assert(test_bit(COW_ZNODE, &znode->flags));
+ ubifs_assert(ubifs_zn_cow(znode));
/*
* It is important that other threads should see %DIRTY_ZNODE
@@ -920,9 +895,31 @@ static int write_index(struct ubifs_info *c)
* the reason for the second barrier.
*/
clear_bit(DIRTY_ZNODE, &znode->flags);
- smp_mb__before_clear_bit();
+ smp_mb__before_atomic();
clear_bit(COW_ZNODE, &znode->flags);
- smp_mb__after_clear_bit();
+ smp_mb__after_atomic();
+
+ /*
+ * We have marked the znode as clean but have not updated the
+ * @c->clean_zn_cnt counter. If this znode becomes dirty again
+ * before 'free_obsolete_znodes()' is called, then
+ * @c->clean_zn_cnt will be decremented before it gets
+ * incremented (resulting in 2 decrements for the same znode).
+ * This means that @c->clean_zn_cnt may become negative for a
+ * while.
+ *
+ * Q: why we cannot increment @c->clean_zn_cnt?
+ * A: because we do not have the @c->tnc_mutex locked, and the
+ * following code would be racy and buggy:
+ *
+ * if (!ubifs_zn_obsolete(znode)) {
+ * atomic_long_inc(&c->clean_zn_cnt);
+ * atomic_long_inc(&ubifs_clean_zn_cnt);
+ * }
+ *
+ * Thus, we just delay the @c->clean_zn_cnt update until we
+ * have the mutex locked.
+ */
/* Do not access znode from this point on */
@@ -940,76 +937,46 @@ static int write_index(struct ubifs_info *c)
else
next_len = ubifs_idx_node_sz(c, cnext->child_cnt);
- if (c->min_io_size == 1) {
- /*
- * Write the prepared index node immediately if there is
- * no minimum IO size
- */
- err = ubifs_leb_write(c, lnum, c->cbuf, buf_offs,
- wlen, UBI_SHORTTERM);
- if (err)
- return err;
- buf_offs += ALIGN(wlen, 8);
- if (next_len) {
- used = 0;
- avail = buf_len;
- if (buf_offs + next_len > c->leb_size) {
- err = ubifs_update_one_lp(c, lnum,
- LPROPS_NC, 0, 0, LPROPS_TAKEN);
- if (err)
- return err;
- lnum = -1;
- }
+ nxt_offs = buf_offs + used + next_len;
+ if (next_len && nxt_offs <= c->leb_size) {
+ if (avail > 0)
continue;
- }
+ else
+ blen = buf_len;
} else {
- int blen, nxt_offs = buf_offs + used + next_len;
-
- if (next_len && nxt_offs <= c->leb_size) {
- if (avail > 0)
- continue;
- else
- blen = buf_len;
- } else {
- wlen = ALIGN(wlen, 8);
- blen = ALIGN(wlen, c->min_io_size);
- ubifs_pad(c, c->cbuf + wlen, blen - wlen);
- }
- /*
- * The buffer is full or there are no more znodes
- * to do
- */
- err = ubifs_leb_write(c, lnum, c->cbuf, buf_offs,
- blen, UBI_SHORTTERM);
- if (err)
- return err;
- buf_offs += blen;
- if (next_len) {
- if (nxt_offs > c->leb_size) {
- err = ubifs_update_one_lp(c, lnum,
- LPROPS_NC, 0, 0, LPROPS_TAKEN);
- if (err)
- return err;
- lnum = -1;
- }
- used -= blen;
- if (used < 0)
- used = 0;
- avail = buf_len - used;
- memmove(c->cbuf, c->cbuf + blen, used);
- continue;
+ wlen = ALIGN(wlen, 8);
+ blen = ALIGN(wlen, c->min_io_size);
+ ubifs_pad(c, c->cbuf + wlen, blen - wlen);
+ }
+
+ /* The buffer is full or there are no more znodes to do */
+ err = ubifs_leb_write(c, lnum, c->cbuf, buf_offs, blen);
+ if (err)
+ return err;
+ buf_offs += blen;
+ if (next_len) {
+ if (nxt_offs > c->leb_size) {
+ err = ubifs_update_one_lp(c, lnum, LPROPS_NC, 0,
+ 0, LPROPS_TAKEN);
+ if (err)
+ return err;
+ lnum = -1;
}
+ used -= blen;
+ if (used < 0)
+ used = 0;
+ avail = buf_len - used;
+ memmove(c->cbuf, c->cbuf + blen, used);
+ continue;
}
break;
}
-#ifdef CONFIG_UBIFS_FS_DEBUG
if (lnum != c->dbg->new_ihead_lnum ||
buf_offs != c->dbg->new_ihead_offs) {
ubifs_err("inconsistent ihead");
return -EINVAL;
}
-#endif
c->ihead_lnum = lnum;
c->ihead_offs = buf_offs;
@@ -1031,7 +998,7 @@ static void free_obsolete_znodes(struct ubifs_info *c)
do {
znode = cnext;
cnext = znode->cnext;
- if (test_bit(OBSOLETE_ZNODE, &znode->flags))
+ if (ubifs_zn_obsolete(znode))
kfree(znode);
else {
znode->cnext = NULL;
diff --git a/fs/ubifs/tnc_misc.c b/fs/ubifs/tnc_misc.c
index b48db999903..f6bf8995c7b 100644
--- a/fs/ubifs/tnc_misc.c
+++ b/fs/ubifs/tnc_misc.c
@@ -293,10 +293,10 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len,
lnum, offs, znode->level, znode->child_cnt);
if (znode->child_cnt > c->fanout || znode->level > UBIFS_MAX_LEVELS) {
- dbg_err("current fanout %d, branch count %d",
- c->fanout, znode->child_cnt);
- dbg_err("max levels %d, znode level %d",
- UBIFS_MAX_LEVELS, znode->level);
+ ubifs_err("current fanout %d, branch count %d",
+ c->fanout, znode->child_cnt);
+ ubifs_err("max levels %d, znode level %d",
+ UBIFS_MAX_LEVELS, znode->level);
err = 1;
goto out_dump;
}
@@ -316,7 +316,7 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len,
if (zbr->lnum < c->main_first ||
zbr->lnum >= c->leb_cnt || zbr->offs < 0 ||
zbr->offs + zbr->len > c->leb_size || zbr->offs & 7) {
- dbg_err("bad branch %d", i);
+ ubifs_err("bad branch %d", i);
err = 2;
goto out_dump;
}
@@ -328,8 +328,8 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len,
case UBIFS_XENT_KEY:
break;
default:
- dbg_msg("bad key type at slot %d: %s", i,
- DBGKEY(&zbr->key));
+ ubifs_err("bad key type at slot %d: %d",
+ i, key_type(c, &zbr->key));
err = 3;
goto out_dump;
}
@@ -340,19 +340,19 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len,
type = key_type(c, &zbr->key);
if (c->ranges[type].max_len == 0) {
if (zbr->len != c->ranges[type].len) {
- dbg_err("bad target node (type %d) length (%d)",
- type, zbr->len);
- dbg_err("have to be %d", c->ranges[type].len);
+ ubifs_err("bad target node (type %d) length (%d)",
+ type, zbr->len);
+ ubifs_err("have to be %d", c->ranges[type].len);
err = 4;
goto out_dump;
}
} else if (zbr->len < c->ranges[type].min_len ||
zbr->len > c->ranges[type].max_len) {
- dbg_err("bad target node (type %d) length (%d)",
- type, zbr->len);
- dbg_err("have to be in range of %d-%d",
- c->ranges[type].min_len,
- c->ranges[type].max_len);
+ ubifs_err("bad target node (type %d) length (%d)",
+ type, zbr->len);
+ ubifs_err("have to be in range of %d-%d",
+ c->ranges[type].min_len,
+ c->ranges[type].max_len);
err = 5;
goto out_dump;
}
@@ -370,13 +370,13 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len,
cmp = keys_cmp(c, key1, key2);
if (cmp > 0) {
- dbg_err("bad key order (keys %d and %d)", i, i + 1);
+ ubifs_err("bad key order (keys %d and %d)", i, i + 1);
err = 6;
goto out_dump;
} else if (cmp == 0 && !is_hash_key(c, key1)) {
/* These can only be keys with colliding hash */
- dbg_err("keys %d and %d are not hashed but equivalent",
- i, i + 1);
+ ubifs_err("keys %d and %d are not hashed but equivalent",
+ i, i + 1);
err = 7;
goto out_dump;
}
@@ -387,7 +387,7 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len,
out_dump:
ubifs_err("bad indexing node at LEB %d:%d, error %d", lnum, offs, err);
- dbg_dump_node(c, idx);
+ ubifs_dump_node(c, idx);
kfree(idx);
return -EINVAL;
}
@@ -475,7 +475,7 @@ int ubifs_tnc_read_node(struct ubifs_info *c, struct ubifs_zbranch *zbr,
zbr->offs);
if (err) {
- dbg_tnc("key %s", DBGKEY(key));
+ dbg_tnck(key, "key ");
return err;
}
@@ -484,9 +484,9 @@ int ubifs_tnc_read_node(struct ubifs_info *c, struct ubifs_zbranch *zbr,
if (!keys_eq(c, key, &key1)) {
ubifs_err("bad key in node at LEB %d:%d",
zbr->lnum, zbr->offs);
- dbg_tnc("looked for key %s found node's key %s",
- DBGKEY(key), DBGKEY1(&key1));
- dbg_dump_node(c, node);
+ dbg_tnck(key, "looked for key ");
+ dbg_tnck(&key1, "but found node's key ");
+ ubifs_dump_node(c, node);
return -EINVAL;
}
diff --git a/fs/ubifs/ubifs-media.h b/fs/ubifs/ubifs-media.h
index 191ca7863fe..e24380cf46e 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 381d6b207a5..c1f71fe17cc 100644
--- a/fs/ubifs/ubifs.h
+++ b/fs/ubifs/ubifs.h
@@ -42,16 +42,24 @@
#define UBIFS_VERSION 1
/* Normal UBIFS messages */
-#define ubifs_msg(fmt, ...) \
- printk(KERN_NOTICE "UBIFS: " fmt "\n", ##__VA_ARGS__)
+#define ubifs_msg(fmt, ...) pr_notice("UBIFS: " fmt "\n", ##__VA_ARGS__)
/* UBIFS error messages */
-#define ubifs_err(fmt, ...) \
- printk(KERN_ERR "UBIFS error (pid %d): %s: " fmt "\n", current->pid, \
+#define ubifs_err(fmt, ...) \
+ pr_err("UBIFS error (pid %d): %s: " fmt "\n", current->pid, \
__func__, ##__VA_ARGS__)
/* UBIFS warning messages */
-#define ubifs_warn(fmt, ...) \
- printk(KERN_WARNING "UBIFS warning (pid %d): %s: " fmt "\n", \
- current->pid, __func__, ##__VA_ARGS__)
+#define ubifs_warn(fmt, ...) \
+ pr_warn("UBIFS warning (pid %d): %s: " fmt "\n", \
+ current->pid, __func__, ##__VA_ARGS__)
+/*
+ * A variant of 'ubifs_err()' which takes the UBIFS file-sytem description
+ * object as an argument.
+ */
+#define ubifs_errc(c, fmt, ...) \
+ do { \
+ if (!(c)->probing) \
+ ubifs_err(fmt, ##__VA_ARGS__); \
+ } while (0)
/* UBIFS file system VFS magic number */
#define UBIFS_SUPER_MAGIC 0x24051905
@@ -84,9 +92,6 @@
#define INUM_WARN_WATERMARK 0xFFF00000
#define INUM_WATERMARK 0xFFFFFF00
-/* Largest key size supported in this implementation */
-#define CUR_MAX_KEY_LEN UBIFS_SK_LEN
-
/* Maximum number of entries in each LPT (LEB category) heap */
#define LPT_HEAP_SZ 256
@@ -151,6 +156,12 @@
*/
#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
@@ -224,14 +235,14 @@ enum {
* LPT cnode flag bits.
*
* DIRTY_CNODE: cnode is dirty
- * COW_CNODE: cnode is being committed and must be copied before writing
* OBSOLETE_CNODE: cnode is being committed and has been copied (or deleted),
- * so it can (and must) be freed when the commit is finished
+ * so it can (and must) be freed when the commit is finished
+ * COW_CNODE: cnode is being committed and must be copied before writing
*/
enum {
DIRTY_CNODE = 0,
- COW_CNODE = 1,
- OBSOLETE_CNODE = 2,
+ OBSOLETE_CNODE = 1,
+ COW_CNODE = 2,
};
/*
@@ -271,10 +282,10 @@ struct ubifs_old_idx {
/* The below union makes it easier to deal with keys */
union ubifs_key {
- uint8_t u8[CUR_MAX_KEY_LEN];
- uint32_t u32[CUR_MAX_KEY_LEN/4];
- uint64_t u64[CUR_MAX_KEY_LEN/8];
- __le32 j32[CUR_MAX_KEY_LEN/4];
+ uint8_t u8[UBIFS_SK_LEN];
+ uint32_t u32[UBIFS_SK_LEN/4];
+ uint64_t u64[UBIFS_SK_LEN/8];
+ __le32 j32[UBIFS_SK_LEN/4];
};
/**
@@ -383,9 +394,9 @@ 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 {
@@ -646,8 +657,7 @@ 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
- * @dtype: type of data stored in this LEB (%UBI_LONGTERM, %UBI_SHORTTERM,
- * %UBI_UNKNOWN)
+ * @size: write-buffer size (in [@c->min_io_size, @c->max_write_size] range)
* @jhead: journal head the mutex belongs to (note, needed only to shut lockdep
* up by 'mutex_lock_nested()).
* @sync_callback: write-buffer synchronization callback
@@ -680,7 +690,7 @@ struct ubifs_wbuf {
int offs;
int avail;
int used;
- int dtype;
+ int size;
int jhead;
int (*sync_callback)(struct ubifs_info *c, int lnum, int free, int pad);
struct mutex io_mutex;
@@ -714,12 +724,14 @@ 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;
};
/**
@@ -755,6 +767,9 @@ struct ubifs_zbranch {
* @offs: offset of the corresponding indexing node
* @len: length of the corresponding indexing node
* @zbranch: array of znode branches (@c->fanout elements)
+ *
+ * Note! The @lnum, @offs, and @len fields are not really needed - we have them
+ * only for internal consistency check. They could be removed to save some RAM.
*/
struct ubifs_znode {
struct ubifs_znode *parent;
@@ -765,9 +780,9 @@ struct ubifs_znode {
int child_cnt;
int iip;
int alt;
-#ifdef CONFIG_UBIFS_FS_DEBUG
- int lnum, offs, len;
-#endif
+ int lnum;
+ int offs;
+ int len;
struct ubifs_zbranch zbranch[];
};
@@ -898,6 +913,8 @@ struct ubifs_budget_req {
* @dnext: next orphan to delete
* @inum: inode number
* @new: %1 => added since the last commit, otherwise %0
+ * @cmt: %1 => commit pending, otherwise %0
+ * @del: %1 => delete pending, otherwise %0
*/
struct ubifs_orphan {
struct rb_node rb;
@@ -906,7 +923,9 @@ struct ubifs_orphan {
struct ubifs_orphan *cnext;
struct ubifs_orphan *dnext;
ino_t inum;
- int new;
+ unsigned new:1;
+ unsigned cmt:1;
+ unsigned del:1;
};
/**
@@ -929,6 +948,40 @@ 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;
/**
@@ -972,6 +1025,7 @@ 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
@@ -1003,6 +1057,11 @@ 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
@@ -1024,7 +1083,12 @@ 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)
@@ -1039,32 +1103,14 @@ struct ubifs_debug_info;
* @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
@@ -1150,6 +1196,8 @@ struct ubifs_debug_info;
* @freeable_list: list of freeable non-index LEBs (free + dirty == @leb_size)
* @frdi_idx_list: list of freeable index LEBs (free + dirty == @leb_size)
* @freeable_cnt: number of freeable LEBs in @freeable_list
+ * @in_a_category_cnt: count of lprops which are in a certain category, which
+ * basically meants that they were loaded from the flash
*
* @ltab_lnum: LEB number of LPT's own lprops table
* @ltab_offs: offset of LPT's own lprops table
@@ -1166,22 +1214,21 @@ 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
+ * @probing: %1 while attempting to mount if MS_SILENT mount flag is set
+ * @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 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 re-mounting from R/O mode to R/W mode
- * @always_chk_crc: always check CRCs (while mounting and remounting to R/W
- * mode)
* @mount_opts: UBIFS-specific mount options
*
* @dbg: debugging-related information
@@ -1221,6 +1268,7 @@ 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;
@@ -1250,6 +1298,9 @@ 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;
@@ -1271,7 +1322,10 @@ 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;
@@ -1285,21 +1339,10 @@ struct ubifs_info {
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;
@@ -1385,6 +1428,7 @@ struct ubifs_info {
struct list_head freeable_list;
struct list_head frdi_idx_list;
int freeable_cnt;
+ int in_a_category_cnt;
int ltab_lnum;
int ltab_offs;
@@ -1398,28 +1442,26 @@ struct ubifs_info {
long long rp_size;
long long report_rp_size;
- uid_t rp_uid;
- gid_t rp_gid;
+ kuid_t rp_uid;
+ kgid_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;
+ unsigned int probing: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
struct ubifs_debug_info *dbg;
-#endif
};
extern struct list_head ubifs_infos;
@@ -1438,16 +1480,23 @@ extern struct ubifs_compressor *ubifs_compressors[UBIFS_COMPR_TYPES_CNT];
/* io.c */
void ubifs_ro_mode(struct ubifs_info *c, int err);
+int ubifs_leb_read(const struct ubifs_info *c, int lnum, void *buf, int offs,
+ int len, int even_ebadmsg);
+int ubifs_leb_write(struct ubifs_info *c, int lnum, const void *buf, int offs,
+ int len);
+int ubifs_leb_change(struct ubifs_info *c, int lnum, const void *buf, int len);
+int ubifs_leb_unmap(struct ubifs_info *c, int lnum);
+int ubifs_leb_map(struct ubifs_info *c, int lnum);
+int ubifs_is_mapped(const struct ubifs_info *c, int lnum);
int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len);
-int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs,
- int dtype);
+int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs);
int ubifs_wbuf_init(struct ubifs_info *c, struct ubifs_wbuf *wbuf);
int ubifs_read_node(const struct ubifs_info *c, void *buf, int type, int len,
int lnum, int offs);
int ubifs_read_node_wbuf(struct ubifs_wbuf *wbuf, void *buf, int type, int len,
int lnum, int offs);
int ubifs_write_node(struct ubifs_info *c, void *node, int len, int lnum,
- int offs, int dtype);
+ int offs);
int ubifs_check_node(const struct ubifs_info *c, const void *buf, int lnum,
int offs, int quiet, int must_chk_crc);
void ubifs_prepare_node(struct ubifs_info *c, void *buf, int len, int pad);
@@ -1586,7 +1635,10 @@ 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);
+unsigned long ubifs_shrink_scan(struct shrinker *shrink,
+ struct shrink_control *sc);
+unsigned long ubifs_shrink_count(struct shrinker *shrink,
+ struct shrink_control *sc);
/* commit.c */
int ubifs_bg_thread(void *info);
@@ -1605,6 +1657,7 @@ 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,
@@ -1689,12 +1742,12 @@ const struct ubifs_lprops *ubifs_fast_find_frdi_idx(struct ubifs_info *c);
int ubifs_calc_dark(const struct ubifs_info *c, int spc);
/* file.c */
-int ubifs_fsync(struct file *file, int datasync);
+int ubifs_fsync(struct file *file, loff_t start, loff_t end, int datasync);
int ubifs_setattr(struct dentry *dentry, struct iattr *attr);
/* dir.c */
struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir,
- int mode);
+ umode_t mode);
int ubifs_getattr(struct vfsmount *mnt, struct dentry *dentry,
struct kstat *stat);
@@ -1713,11 +1766,11 @@ 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);
-int ubifs_clean_lebs(const struct ubifs_info *c, void *sbuf);
+int ubifs_recover_inl_heads(struct ubifs_info *c, void *sbuf);
+int ubifs_clean_lebs(struct ubifs_info *c, void *sbuf);
int ubifs_rcvry_gc_commit(struct ubifs_info *c);
int ubifs_recover_size_accum(struct ubifs_info *c, union ubifs_key *key,
int deletion, loff_t new_size);
diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c
index c74400f88fe..5e0a63b1b0d 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.
@@ -130,20 +130,19 @@ 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);
ui->xattr = 1;
ui->flags |= UBIFS_XATTR_FL;
- ui->data = kmalloc(size, GFP_NOFS);
+ ui->data = kmemdup(value, size, GFP_NOFS);
if (!ui->data) {
err = -ENOMEM;
goto out_free;
}
- memcpy(ui->data, value, size);
inode->i_size = ui->ui_size = size;
ui->data_len = size;
@@ -204,12 +203,11 @@ static int change_xattr(struct ubifs_info *c, struct inode *host,
return err;
kfree(ui->data);
- ui->data = kmalloc(size, GFP_NOFS);
+ ui->data = kmemdup(value, size, GFP_NOFS);
if (!ui->data) {
err = -ENOMEM;
goto out_free;
}
- memcpy(ui->data, value, size);
inode->i_size = ui->ui_size = size;
ui->data_len = size;
@@ -300,13 +298,13 @@ int ubifs_setxattr(struct dentry *dentry, const char *name,
{
struct inode *inode, *host = dentry->d_inode;
struct ubifs_info *c = host->i_sb->s_fs_info;
- struct qstr nm = { .name = name, .len = strlen(name) };
+ struct qstr nm = QSTR_INIT(name, strlen(name));
struct ubifs_dent_node *xent;
union ubifs_key key;
int err, type;
- dbg_gen("xattr '%s', host ino %lu ('%.*s'), size %zd", name,
- host->i_ino, dentry->d_name.len, dentry->d_name.name, size);
+ dbg_gen("xattr '%s', host ino %lu ('%pd'), size %zd", name,
+ host->i_ino, dentry, size);
ubifs_assert(mutex_is_locked(&host->i_mutex));
if (size > UBIFS_MAX_INO_DATA)
@@ -363,14 +361,14 @@ ssize_t ubifs_getxattr(struct dentry *dentry, const char *name, void *buf,
{
struct inode *inode, *host = dentry->d_inode;
struct ubifs_info *c = host->i_sb->s_fs_info;
- struct qstr nm = { .name = name, .len = strlen(name) };
+ struct qstr nm = QSTR_INIT(name, strlen(name));
struct ubifs_inode *ui;
struct ubifs_dent_node *xent;
union ubifs_key key;
int err;
- dbg_gen("xattr '%s', ino %lu ('%.*s'), buf size %zd", name,
- host->i_ino, dentry->d_name.len, dentry->d_name.name, size);
+ dbg_gen("xattr '%s', ino %lu ('%pd'), buf size %zd", name,
+ host->i_ino, dentry, size);
err = check_namespace(&nm);
if (err < 0)
@@ -401,8 +399,8 @@ ssize_t ubifs_getxattr(struct dentry *dentry, const char *name, void *buf,
if (buf) {
/* If @buf is %NULL we are supposed to return the length */
if (ui->data_len > size) {
- dbg_err("buffer size %zd, xattr len %d",
- size, ui->data_len);
+ ubifs_err("buffer size %zd, xattr len %d",
+ size, ui->data_len);
err = -ERANGE;
goto out_iput;
}
@@ -428,8 +426,8 @@ ssize_t ubifs_listxattr(struct dentry *dentry, char *buffer, size_t size)
int err, len, written = 0;
struct qstr nm = { .name = NULL };
- dbg_gen("ino %lu ('%.*s'), buffer size %zd", host->i_ino,
- dentry->d_name.len, dentry->d_name.name, size);
+ dbg_gen("ino %lu ('%pd'), buffer size %zd", host->i_ino,
+ dentry, size);
len = host_ui->xattr_names + host_ui->xattr_cnt;
if (!buffer)
@@ -526,13 +524,13 @@ int ubifs_removexattr(struct dentry *dentry, const char *name)
{
struct inode *inode, *host = dentry->d_inode;
struct ubifs_info *c = host->i_sb->s_fs_info;
- struct qstr nm = { .name = name, .len = strlen(name) };
+ struct qstr nm = QSTR_INIT(name, strlen(name));
struct ubifs_dent_node *xent;
union ubifs_key key;
int err;
- dbg_gen("xattr '%s', ino %lu ('%.*s')", name,
- host->i_ino, dentry->d_name.len, dentry->d_name.name);
+ dbg_gen("xattr '%s', ino %lu ('%pd')", name,
+ host->i_ino, dentry);
ubifs_assert(mutex_is_locked(&host->i_mutex));
err = check_namespace(&nm);
@@ -558,10 +556,10 @@ int ubifs_removexattr(struct dentry *dentry, const char *name)
}
ubifs_assert(inode->i_nlink == 1);
- inode->i_nlink = 0;
+ clear_nlink(inode);
err = remove_xattr(c, host, inode, &nm);
if (err)
- inode->i_nlink = 1;
+ set_nlink(inode, 1);
/* If @i_nlink is 0, 'iput()' will delete the inode */
iput(inode);
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-09 22:14:56 +0000