| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Sat, 16 Aug 2014 16:46:48 -0700
From: "Darrick J. Wong" <darrick.wong@...cle.com>
To: tytso@....edu, darrick.wong@...cle.com
Cc: linux-ext4@...r.kernel.org
Subject: [PATCH 09/27] e2fsck: resync jbd2 revoke code from Linux 3.16
Synchronize e2fsck's copy of revoke.c with the kernel's copy in
fs/jbd2.
Signed-off-by: Darrick J. Wong <darrick.wong@...cle.com>
---
e2fsck/jfs_user.h | 2
e2fsck/revoke.c | 477 ++++++++++++++++++++++++++++++-----------------
lib/ext2fs/jfs_compat.h | 15 +
lib/ext2fs/kernel-jbd.h | 5
4 files changed, 320 insertions(+), 179 deletions(-)
diff --git a/e2fsck/jfs_user.h b/e2fsck/jfs_user.h
index 799c6da..9405e47 100644
--- a/e2fsck/jfs_user.h
+++ b/e2fsck/jfs_user.h
@@ -54,7 +54,7 @@ typedef struct {
#define kmem_cache_alloc(cache,flags) malloc((cache)->object_length)
#define kmem_cache_free(cache,obj) free(obj)
-#define kmem_cache_create(name,len,a,b,c,d) do_cache_create(len)
+#define kmem_cache_create(name,len,a,b,c) do_cache_create(len)
#define kmem_cache_destroy(cache) do_cache_destroy(cache)
#define kmalloc(len,flags) malloc(len)
#define kfree(p) free(p)
diff --git a/e2fsck/revoke.c b/e2fsck/revoke.c
index 38c265e..383164e 100644
--- a/e2fsck/revoke.c
+++ b/e2fsck/revoke.c
@@ -1,5 +1,5 @@
/*
- * linux/fs/revoke.c
+ * linux/fs/jbd2/revoke.c
*
* Written by Stephen C. Tweedie <sct@...hat.com>, 2000
*
@@ -47,6 +47,10 @@
* overwriting the new data. We don't even need to clear the revoke
* bit here.
*
+ * We cache revoke status of a buffer in the current transaction in b_states
+ * bits. As the name says, revokevalid flag indicates that the cached revoke
+ * status of a buffer is valid and we can rely on the cached status.
+ *
* Revoke information on buffers is a tri-state value:
*
* RevokeValid clear: no cached revoke status, need to look it up
@@ -55,40 +59,58 @@
* need do nothing.
* RevokeValid set, Revoked set:
* buffer has been revoked.
+ *
+ * Locking rules:
+ * We keep two hash tables of revoke records. One hashtable belongs to the
+ * running transaction (is pointed to by journal->j_revoke), the other one
+ * belongs to the committing transaction. Accesses to the second hash table
+ * happen only from the kjournald and no other thread touches this table. Also
+ * journal_switch_revoke_table() which switches which hashtable belongs to the
+ * running and which to the committing transaction is called only from
+ * kjournald. Therefore we need no locks when accessing the hashtable belonging
+ * to the committing transaction.
+ *
+ * All users operating on the hash table belonging to the running transaction
+ * have a handle to the transaction. Therefore they are safe from kjournald
+ * switching hash tables under them. For operations on the lists of entries in
+ * the hash table j_revoke_lock is used.
+ *
+ * Finally, also replay code uses the hash tables but at this moment no one else
+ * can touch them (filesystem isn't mounted yet) and hence no locking is
+ * needed.
*/
#ifndef __KERNEL__
-#include "config.h"
#include "jfs_user.h"
#else
-#include <linux/sched.h>
+#include <linux/time.h>
#include <linux/fs.h>
-#include <linux/jbd.h>
+#include <linux/jbd2.h>
#include <linux/errno.h>
#include <linux/slab.h>
-#include <linux/locks.h>
#include <linux/list.h>
-#include <linux/smp_lock.h>
#include <linux/init.h>
+#include <linux/bio.h>
+#include <linux/log2.h>
#endif
-static lkmem_cache_t *revoke_record_cache;
-static lkmem_cache_t *revoke_table_cache;
+static lkmem_cache_t *jbd2_revoke_record_cache;
+static lkmem_cache_t *jbd2_revoke_table_cache;
/* Each revoke record represents one single revoked block. During
journal replay, this involves recording the transaction ID of the
last transaction to revoke this block. */
-struct jbd_revoke_record_s
+struct jbd2_revoke_record_s
{
struct list_head hash;
tid_t sequence; /* Used for recovery only */
- unsigned long blocknr;
+ unsigned long long blocknr;
};
/* The revoke table is just a simple hash table of revoke records. */
-struct jbd_revoke_table_s
+struct jbd2_revoke_table_s
{
/* It is conceivable that we might want a larger hash table
* for recovery. Must be a power of two. */
@@ -100,159 +122,187 @@ struct jbd_revoke_table_s
#ifdef __KERNEL__
static void write_one_revoke_record(journal_t *, transaction_t *,
- struct journal_head **, int *,
- struct jbd_revoke_record_s *);
-static void flush_descriptor(journal_t *, struct journal_head *, int);
+ struct list_head *,
+ struct buffer_head **, int *,
+ struct jbd2_revoke_record_s *, int);
+static void flush_descriptor(journal_t *, struct buffer_head *, int, int);
#endif
/* Utility functions to maintain the revoke table */
/* Borrowed from buffer.c: this is a tried and tested block hash function */
-static inline int hash(journal_t *journal, unsigned long block)
+static inline int hash(journal_t *journal, unsigned long long block)
{
- struct jbd_revoke_table_s *table = journal->j_revoke;
+ struct jbd2_revoke_table_s *table = journal->j_revoke;
int hash_shift = table->hash_shift;
+ int hash = (int)block ^ (int)((block >> 31) >> 1);
- return ((block << (hash_shift - 6)) ^
- (block >> 13) ^
- (block << (hash_shift - 12))) & (table->hash_size - 1);
+ return ((hash << (hash_shift - 6)) ^
+ (hash >> 13) ^
+ (hash << (hash_shift - 12))) & (table->hash_size - 1);
}
-static int insert_revoke_hash(journal_t *journal, unsigned long blocknr,
+static int insert_revoke_hash(journal_t *journal, unsigned long long blocknr,
tid_t seq)
{
struct list_head *hash_list;
- struct jbd_revoke_record_s *record;
+ struct jbd2_revoke_record_s *record;
-#ifdef __KERNEL__
repeat:
-#endif
- record = kmem_cache_alloc(revoke_record_cache, GFP_NOFS);
+ record = kmem_cache_alloc(jbd2_revoke_record_cache, GFP_NOFS);
if (!record)
goto oom;
record->sequence = seq;
record->blocknr = blocknr;
hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)];
+ spin_lock(&journal->j_revoke_lock);
list_add(&record->hash, hash_list);
+ spin_unlock(&journal->j_revoke_lock);
return 0;
oom:
-#ifdef __KERNEL__
if (!journal_oom_retry)
return -ENOMEM;
- jbd_debug(1, "ENOMEM in " __FUNCTION__ ", retrying.\n");
- current->policy |= SCHED_YIELD;
- schedule();
+ jbd_debug(1, "ENOMEM in %s, retrying\n", __func__);
+ yield();
goto repeat;
-#else
- return -ENOMEM;
-#endif
}
/* Find a revoke record in the journal's hash table. */
-static struct jbd_revoke_record_s *find_revoke_record(journal_t *journal,
- unsigned long blocknr)
+static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal,
+ unsigned long long blocknr)
{
struct list_head *hash_list;
- struct jbd_revoke_record_s *record;
+ struct jbd2_revoke_record_s *record;
hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)];
- record = (struct jbd_revoke_record_s *) hash_list->next;
+ spin_lock(&journal->j_revoke_lock);
+ record = (struct jbd2_revoke_record_s *) hash_list->next;
while (&(record->hash) != hash_list) {
- if (record->blocknr == blocknr)
+ if (record->blocknr == blocknr) {
+ spin_unlock(&journal->j_revoke_lock);
return record;
- record = (struct jbd_revoke_record_s *) record->hash.next;
+ }
+ record = (struct jbd2_revoke_record_s *) record->hash.next;
}
+ spin_unlock(&journal->j_revoke_lock);
return NULL;
}
-int __init journal_init_revoke_caches(void)
+void journal_destroy_revoke_caches(void)
{
- revoke_record_cache = kmem_cache_create("revoke_record",
- sizeof(struct jbd_revoke_record_s),
- 0, SLAB_HWCACHE_ALIGN, NULL, NULL);
- if (revoke_record_cache == 0)
- return -ENOMEM;
-
- revoke_table_cache = kmem_cache_create("revoke_table",
- sizeof(struct jbd_revoke_table_s),
- 0, 0, NULL, NULL);
- if (revoke_table_cache == 0) {
- kmem_cache_destroy(revoke_record_cache);
- revoke_record_cache = NULL;
- return -ENOMEM;
+ if (jbd2_revoke_record_cache) {
+ kmem_cache_destroy(jbd2_revoke_record_cache);
+ jbd2_revoke_record_cache = NULL;
+ }
+ if (jbd2_revoke_table_cache) {
+ kmem_cache_destroy(jbd2_revoke_table_cache);
+ jbd2_revoke_table_cache = NULL;
}
- return 0;
}
-void journal_destroy_revoke_caches(void)
+int __init journal_init_revoke_caches(void)
{
- kmem_cache_destroy(revoke_record_cache);
- revoke_record_cache = 0;
- kmem_cache_destroy(revoke_table_cache);
- revoke_table_cache = 0;
+ J_ASSERT(!jbd2_revoke_record_cache);
+ J_ASSERT(!jbd2_revoke_table_cache);
+
+ jbd2_revoke_record_cache = KMEM_CACHE(jbd2_revoke_record_s,
+ SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY);
+ if (!jbd2_revoke_record_cache)
+ goto record_cache_failure;
+
+ jbd2_revoke_table_cache = KMEM_CACHE(jbd2_revoke_table_s,
+ SLAB_TEMPORARY);
+ if (!jbd2_revoke_table_cache)
+ goto table_cache_failure;
+ return 0;
+table_cache_failure:
+ journal_destroy_revoke_caches();
+record_cache_failure:
+ return -ENOMEM;
}
-/* Initialise the revoke table for a given journal to a given size. */
-
-int journal_init_revoke(journal_t *journal, int hash_size)
+static struct jbd2_revoke_table_s *journal_init_revoke_table(int hash_size)
{
- int shift, tmp;
+ int shift = 0;
+ int tmp = hash_size;
+ struct jbd2_revoke_table_s *table;
- J_ASSERT (journal->j_revoke == NULL);
-
- journal->j_revoke = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL);
- if (!journal->j_revoke)
- return -ENOMEM;
-
- /* Check that the hash_size is a power of two */
- J_ASSERT ((hash_size & (hash_size-1)) == 0);
-
- journal->j_revoke->hash_size = hash_size;
+ table = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL);
+ if (!table)
+ goto out;
- shift = 0;
- tmp = hash_size;
while((tmp >>= 1UL) != 0UL)
shift++;
- journal->j_revoke->hash_shift = shift;
- journal->j_revoke->hash_table =
+ table->hash_size = hash_size;
+ table->hash_shift = shift;
+ table->hash_table =
kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
- if (!journal->j_revoke->hash_table) {
- kmem_cache_free(revoke_table_cache, journal->j_revoke);
- journal->j_revoke = NULL;
- return -ENOMEM;
+ if (!table->hash_table) {
+ kmem_cache_free(jbd2_revoke_table_cache, table);
+ table = NULL;
+ goto out;
}
for (tmp = 0; tmp < hash_size; tmp++)
- INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]);
+ INIT_LIST_HEAD(&table->hash_table[tmp]);
- return 0;
+out:
+ return table;
}
-/* Destoy a journal's revoke table. The table must already be empty! */
-
-void journal_destroy_revoke(journal_t *journal)
+static void journal_destroy_revoke_table(struct jbd2_revoke_table_s *table)
{
- struct jbd_revoke_table_s *table;
- struct list_head *hash_list;
int i;
+ struct list_head *hash_list;
- table = journal->j_revoke;
- if (!table)
- return;
-
- for (i=0; i<table->hash_size; i++) {
+ for (i = 0; i < table->hash_size; i++) {
hash_list = &table->hash_table[i];
- J_ASSERT (list_empty(hash_list));
+ J_ASSERT(list_empty(hash_list));
}
kfree(table->hash_table);
- kmem_cache_free(revoke_table_cache, table);
+ kmem_cache_free(jbd2_revoke_table_cache, table);
+}
+
+/* Initialise the revoke table for a given journal to a given size. */
+int journal_init_revoke(journal_t *journal, int hash_size)
+{
+ J_ASSERT(journal->j_revoke_table[0] == NULL);
+ J_ASSERT(is_power_of_2(hash_size));
+
+ journal->j_revoke_table[0] = journal_init_revoke_table(hash_size);
+ if (!journal->j_revoke_table[0])
+ goto fail0;
+
+ journal->j_revoke_table[1] = journal_init_revoke_table(hash_size);
+ if (!journal->j_revoke_table[1])
+ goto fail1;
+
+ journal->j_revoke = journal->j_revoke_table[1];
+
+ spin_lock_init(&journal->j_revoke_lock);
+
+ return 0;
+
+fail1:
+ journal_destroy_revoke_table(journal->j_revoke_table[0]);
+fail0:
+ return -ENOMEM;
+}
+
+/* Destroy a journal's revoke table. The table must already be empty! */
+void journal_destroy_revoke(journal_t *journal)
+{
journal->j_revoke = NULL;
+ if (journal->j_revoke_table[0])
+ journal_destroy_revoke_table(journal->j_revoke_table[0]);
+ if (journal->j_revoke_table[1])
+ journal_destroy_revoke_table(journal->j_revoke_table[1]);
}
@@ -282,14 +332,15 @@ void journal_destroy_revoke(journal_t *journal)
* by one.
*/
-int journal_revoke(handle_t *handle, unsigned long blocknr,
+int journal_revoke(handle_t *handle, unsigned long long blocknr,
struct buffer_head *bh_in)
{
struct buffer_head *bh = NULL;
journal_t *journal;
- kdev_t dev;
+ struct block_device *bdev;
int err;
+ might_sleep();
if (bh_in)
BUFFER_TRACE(bh_in, "enter");
@@ -299,34 +350,32 @@ int journal_revoke(handle_t *handle, unsigned long blocknr,
return -EINVAL;
}
- dev = journal->j_fs_dev;
+ bdev = journal->j_fs_dev;
bh = bh_in;
if (!bh) {
- bh = get_hash_table(dev, blocknr, journal->j_blocksize);
+ bh = __find_get_block(bdev, blocknr, journal->j_blocksize);
if (bh)
BUFFER_TRACE(bh, "found on hash");
}
-#ifdef JBD_EXPENSIVE_CHECKING
+#ifdef JFS_EXPENSIVE_CHECKING
else {
struct buffer_head *bh2;
/* If there is a different buffer_head lying around in
* memory anywhere... */
- bh2 = get_hash_table(dev, blocknr, journal->j_blocksize);
+ bh2 = __find_get_block(bdev, blocknr, journal->j_blocksize);
if (bh2) {
/* ... and it has RevokeValid status... */
- if ((bh2 != bh) &&
- test_bit(BH_RevokeValid, &bh2->b_state))
+ if (bh2 != bh && buffer_revokevalid(bh2))
/* ...then it better be revoked too,
* since it's illegal to create a revoke
* record against a buffer_head which is
* not marked revoked --- that would
* risk missing a subsequent revoke
* cancel. */
- J_ASSERT_BH(bh2, test_bit(BH_Revoked, &
- bh2->b_state));
- __brelse(bh2);
+ J_ASSERT_BH(bh2, buffer_revoked(bh2));
+ put_bh(bh2);
}
}
#endif
@@ -335,9 +384,14 @@ int journal_revoke(handle_t *handle, unsigned long blocknr,
first having the revoke cancelled: it's illegal to free a
block twice without allocating it in between! */
if (bh) {
- J_ASSERT_BH(bh, !test_bit(BH_Revoked, &bh->b_state));
- set_bit(BH_Revoked, &bh->b_state);
- set_bit(BH_RevokeValid, &bh->b_state);
+ if (!J_EXPECT_BH(bh, !buffer_revoked(bh),
+ "inconsistent data on disk")) {
+ if (!bh_in)
+ brelse(bh);
+ return -EIO;
+ }
+ set_buffer_revoked(bh);
+ set_buffer_revokevalid(bh);
if (bh_in) {
BUFFER_TRACE(bh_in, "call journal_forget");
journal_forget(handle, bh_in);
@@ -347,11 +401,9 @@ int journal_revoke(handle_t *handle, unsigned long blocknr,
}
}
- lock_journal(journal);
- jbd_debug(2, "insert revoke for block %lu, bh_in=%p\n", blocknr, bh_in);
+ jbd_debug(2, "insert revoke for block %llu, bh_in=%p\n",blocknr, bh_in);
err = insert_revoke_hash(journal, blocknr,
handle->h_transaction->t_tid);
- unlock_journal(journal);
BUFFER_TRACE(bh_in, "exit");
return err;
}
@@ -360,7 +412,7 @@ int journal_revoke(handle_t *handle, unsigned long blocknr,
* Cancel an outstanding revoke. For use only internally by the
* journaling code (called from journal_get_write_access).
*
- * We trust the BH_Revoked bit on the buffer if the buffer is already
+ * We trust buffer_revoked() on the buffer if the buffer is already
* being journaled: if there is no revoke pending on the buffer, then we
* don't do anything here.
*
@@ -370,12 +422,10 @@ int journal_revoke(handle_t *handle, unsigned long blocknr,
* the second time we would still have a pending revoke to cancel. So,
* do not trust the Revoked bit on buffers unless RevokeValid is also
* set.
- *
- * The caller must have the journal locked.
*/
int journal_cancel_revoke(handle_t *handle, struct journal_head *jh)
{
- struct jbd_revoke_record_s *record;
+ struct jbd2_revoke_record_s *record;
journal_t *journal = handle->h_transaction->t_journal;
int need_cancel;
int did_revoke = 0; /* akpm: debug */
@@ -387,25 +437,27 @@ int journal_cancel_revoke(handle_t *handle, struct journal_head *jh)
* only perform the full cancel if the revoke bit is set. If
* not, we can't trust the revoke bit, and we need to do the
* full search for a revoke record. */
- if (test_and_set_bit(BH_RevokeValid, &bh->b_state))
- need_cancel = (test_and_clear_bit(BH_Revoked, &bh->b_state));
- else {
+ if (test_set_buffer_revokevalid(bh)) {
+ need_cancel = test_clear_buffer_revoked(bh);
+ } else {
need_cancel = 1;
- clear_bit(BH_Revoked, &bh->b_state);
+ clear_buffer_revoked(bh);
}
if (need_cancel) {
record = find_revoke_record(journal, bh->b_blocknr);
if (record) {
jbd_debug(4, "cancelled existing revoke on "
- "blocknr %lu\n", bh->b_blocknr);
+ "blocknr %llu\n", (unsigned long long)bh->b_blocknr);
+ spin_lock(&journal->j_revoke_lock);
list_del(&record->hash);
- kmem_cache_free(revoke_record_cache, record);
+ spin_unlock(&journal->j_revoke_lock);
+ kmem_cache_free(jbd2_revoke_record_cache, record);
did_revoke = 1;
}
}
-#ifdef JBD_EXPENSIVE_CHECKING
+#ifdef JFS_EXPENSIVE_CHECKING
/* There better not be one left behind by now! */
record = find_revoke_record(journal, bh->b_blocknr);
J_ASSERT_JH(jh, record == NULL);
@@ -415,56 +467,104 @@ int journal_cancel_revoke(handle_t *handle, struct journal_head *jh)
* buffer_head? If so, we'd better make sure we clear the
* revoked status on any hashed alias too, otherwise the revoke
* state machine will get very upset later on. */
- if (need_cancel && !bh->b_pprev) {
+ if (need_cancel) {
struct buffer_head *bh2;
- bh2 = get_hash_table(bh->b_dev, bh->b_blocknr, bh->b_size);
+ bh2 = __find_get_block(bh->b_bdev, bh->b_blocknr, bh->b_size);
if (bh2) {
- clear_bit(BH_Revoked, &bh2->b_state);
+ if (bh2 != bh)
+ clear_buffer_revoked(bh2);
__brelse(bh2);
}
}
-
return did_revoke;
}
+/*
+ * journal_clear_revoked_flag clears revoked flag of buffers in
+ * revoke table to reflect there is no revoked buffers in the next
+ * transaction which is going to be started.
+ */
+void jbd2_clear_buffer_revoked_flags(journal_t *journal)
+{
+ struct jbd2_revoke_table_s *revoke = journal->j_revoke;
+ int i = 0;
+
+ for (i = 0; i < revoke->hash_size; i++) {
+ struct list_head *hash_list;
+ struct list_head *list_entry;
+ hash_list = &revoke->hash_table[i];
+
+ list_for_each(list_entry, hash_list) {
+ struct jbd2_revoke_record_s *record;
+ struct buffer_head *bh;
+ record = (struct jbd2_revoke_record_s *)list_entry;
+ bh = __find_get_block(journal->j_fs_dev,
+ record->blocknr,
+ journal->j_blocksize);
+ if (bh) {
+ clear_buffer_revoked(bh);
+ __brelse(bh);
+ }
+ }
+ }
+}
+
+/* journal_switch_revoke table select j_revoke for next transaction
+ * we do not want to suspend any processing until all revokes are
+ * written -bzzz
+ */
+void journal_switch_revoke_table(journal_t *journal)
+{
+ int i;
+
+ if (journal->j_revoke == journal->j_revoke_table[0])
+ journal->j_revoke = journal->j_revoke_table[1];
+ else
+ journal->j_revoke = journal->j_revoke_table[0];
+
+ for (i = 0; i < journal->j_revoke->hash_size; i++)
+ INIT_LIST_HEAD(&journal->j_revoke->hash_table[i]);
+}
/*
* Write revoke records to the journal for all entries in the current
* revoke hash, deleting the entries as we go.
- *
- * Called with the journal lock held.
*/
-
void journal_write_revoke_records(journal_t *journal,
- transaction_t *transaction)
+ transaction_t *transaction,
+ struct list_head *log_bufs,
+ int write_op)
{
- struct journal_head *descriptor;
- struct jbd_revoke_record_s *record;
- struct jbd_revoke_table_s *revoke;
+ struct buffer_head *descriptor;
+ struct jbd2_revoke_record_s *record;
+ struct jbd2_revoke_table_s *revoke;
struct list_head *hash_list;
int i, offset, count;
descriptor = NULL;
offset = 0;
count = 0;
- revoke = journal->j_revoke;
+
+ /* select revoke table for committing transaction */
+ revoke = journal->j_revoke == journal->j_revoke_table[0] ?
+ journal->j_revoke_table[1] : journal->j_revoke_table[0];
for (i = 0; i < revoke->hash_size; i++) {
hash_list = &revoke->hash_table[i];
while (!list_empty(hash_list)) {
- record = (struct jbd_revoke_record_s *)
+ record = (struct jbd2_revoke_record_s *)
hash_list->next;
- write_one_revoke_record(journal, transaction,
+ write_one_revoke_record(journal, transaction, log_bufs,
&descriptor, &offset,
- record);
+ record, write_op);
count++;
list_del(&record->hash);
- kmem_cache_free(revoke_record_cache, record);
+ kmem_cache_free(jbd2_revoke_record_cache, record);
}
}
if (descriptor)
- flush_descriptor(journal, descriptor, offset);
+ flush_descriptor(journal, descriptor, offset, write_op);
jbd_debug(1, "Wrote %d revoke records\n", count);
}
@@ -475,11 +575,14 @@ void journal_write_revoke_records(journal_t *journal,
static void write_one_revoke_record(journal_t *journal,
transaction_t *transaction,
- struct journal_head **descriptorp,
+ struct list_head *log_bufs,
+ struct buffer_head **descriptorp,
int *offsetp,
- struct jbd_revoke_record_s *record)
+ struct jbd2_revoke_record_s *record,
+ int write_op)
{
- struct journal_head *descriptor;
+ int csum_size = 0;
+ struct buffer_head *descriptor;
int offset;
journal_header_t *header;
@@ -493,10 +596,14 @@ static void write_one_revoke_record(journal_t *journal,
descriptor = *descriptorp;
offset = *offsetp;
+ /* Do we need to leave space at the end for a checksum? */
+ if (JFS_HAS_INCOMPAT_FEATURE(journal, JFS_FEATURE_INCOMPAT_CSUM_V2))
+ csum_size = sizeof(struct journal_revoke_tail);
+
/* Make sure we have a descriptor with space left for the record */
if (descriptor) {
- if (offset == journal->j_blocksize) {
- flush_descriptor(journal, descriptor, offset);
+ if (offset >= journal->j_blocksize - csum_size) {
+ flush_descriptor(journal, descriptor, offset, write_op);
descriptor = NULL;
}
}
@@ -505,25 +612,48 @@ static void write_one_revoke_record(journal_t *journal,
descriptor = journal_get_descriptor_buffer(journal);
if (!descriptor)
return;
- header = (journal_header_t *) &jh2bh(descriptor)->b_data[0];
- header->h_magic = htonl(JFS_MAGIC_NUMBER);
- header->h_blocktype = htonl(JFS_REVOKE_BLOCK);
- header->h_sequence = htonl(transaction->t_tid);
+ header = (journal_header_t *)descriptor->b_data;
+ header->h_magic = ext2fs_cpu_to_be32(JFS_MAGIC_NUMBER);
+ header->h_blocktype = ext2fs_cpu_to_be32(JFS_REVOKE_BLOCK);
+ header->h_sequence = ext2fs_cpu_to_be32(transaction->t_tid);
/* Record it so that we can wait for IO completion later */
- JBUFFER_TRACE(descriptor, "file as BJ_LogCtl");
- journal_file_buffer(descriptor, transaction, BJ_LogCtl);
+ BUFFER_TRACE(descriptor, "file in log_bufs");
+ jbd2_file_log_bh(log_bufs, descriptor);
offset = sizeof(journal_revoke_header_t);
*descriptorp = descriptor;
}
- * ((unsigned int *)(&jh2bh(descriptor)->b_data[offset])) =
- htonl(record->blocknr);
- offset += 4;
+ if (JFS_HAS_INCOMPAT_FEATURE(journal, JFS_FEATURE_INCOMPAT_64BIT)) {
+ * ((__u64 *)(&descriptor->b_data[offset])) =
+ ext2fs_cpu_to_be64(record->blocknr);
+ offset += 8;
+
+ } else {
+ * ((__u32 *)(&descriptor->b_data[offset])) =
+ ext2fs_cpu_to_be32(record->blocknr);
+ offset += 4;
+ }
+
*offsetp = offset;
}
+static void jbd2_revoke_csum_set(journal_t *j, struct buffer_head *bh)
+{
+ struct journal_revoke_tail *tail;
+ __u32 csum;
+
+ if (!JFS_HAS_INCOMPAT_FEATURE(j, JFS_FEATURE_INCOMPAT_CSUM_V2))
+ return;
+
+ tail = (struct journal_revoke_tail *)(bh->b_data + j->j_blocksize -
+ sizeof(struct journal_revoke_tail));
+ tail->r_checksum = 0;
+ csum = jbd2_chksum(j, j->j_csum_seed, bh->b_data, j->j_blocksize);
+ tail->r_checksum = ext2fs_cpu_to_be32(csum);
+}
+
/*
* Flush a revoke descriptor out to the journal. If we are aborting,
* this is a noop; otherwise we are generating a buffer which needs to
@@ -532,27 +662,25 @@ static void write_one_revoke_record(journal_t *journal,
*/
static void flush_descriptor(journal_t *journal,
- struct journal_head *descriptor,
- int offset)
+ struct buffer_head *descriptor,
+ int offset, int write_op)
{
journal_revoke_header_t *header;
if (is_journal_aborted(journal)) {
- JBUFFER_TRACE(descriptor, "brelse");
- __brelse(jh2bh(descriptor));
+ put_bh(descriptor);
return;
}
- header = (journal_revoke_header_t *) jh2bh(descriptor)->b_data;
- header->r_count = htonl(offset);
- set_bit(BH_JWrite, &jh2bh(descriptor)->b_state);
- {
- struct buffer_head *bh = jh2bh(descriptor);
- BUFFER_TRACE(bh, "write");
- ll_rw_block (WRITE, 1, &bh);
- }
-}
+ header = (journal_revoke_header_t *)descriptor->b_data;
+ header->r_count = ext2fs_cpu_to_be32(offset);
+ jbd2_revoke_csum_set(journal, descriptor);
+ set_buffer_jwrite(descriptor);
+ BUFFER_TRACE(descriptor, "write");
+ set_buffer_dirty(descriptor);
+ write_dirty_buffer(descriptor, write_op);
+}
#endif
/*
@@ -578,14 +706,14 @@ static void flush_descriptor(journal_t *journal,
*/
int journal_set_revoke(journal_t *journal,
- unsigned long blocknr,
+ unsigned long long blocknr,
tid_t sequence)
{
- struct jbd_revoke_record_s *record;
+ struct jbd2_revoke_record_s *record;
record = find_revoke_record(journal, blocknr);
if (record) {
- /* If we have multiple occurences, only record the
+ /* If we have multiple occurrences, only record the
* latest sequence number in the hashed record */
if (tid_gt(sequence, record->sequence))
record->sequence = sequence;
@@ -602,10 +730,10 @@ int journal_set_revoke(journal_t *journal,
*/
int journal_test_revoke(journal_t *journal,
- unsigned long blocknr,
+ unsigned long long blocknr,
tid_t sequence)
{
- struct jbd_revoke_record_s *record;
+ struct jbd2_revoke_record_s *record;
record = find_revoke_record(journal, blocknr);
if (!record)
@@ -624,18 +752,17 @@ void journal_clear_revoke(journal_t *journal)
{
int i;
struct list_head *hash_list;
- struct jbd_revoke_record_s *record;
- struct jbd_revoke_table_s *revoke;
+ struct jbd2_revoke_record_s *record;
+ struct jbd2_revoke_table_s *revoke;
revoke = journal->j_revoke;
for (i = 0; i < revoke->hash_size; i++) {
hash_list = &revoke->hash_table[i];
while (!list_empty(hash_list)) {
- record = (struct jbd_revoke_record_s*) hash_list->next;
+ record = (struct jbd2_revoke_record_s*) hash_list->next;
list_del(&record->hash);
- kmem_cache_free(revoke_record_cache, record);
+ kmem_cache_free(jbd2_revoke_record_cache, record);
}
}
}
-
diff --git a/lib/ext2fs/jfs_compat.h b/lib/ext2fs/jfs_compat.h
index 2638c31..d92f931 100644
--- a/lib/ext2fs/jfs_compat.h
+++ b/lib/ext2fs/jfs_compat.h
@@ -42,6 +42,18 @@ static inline __u32 jbd2_chksum(journal_t *j, __u32 crc, const void *address,
return ext2fs_crc32c_le(crc, address, length);
}
#define crc32_be(x, y, z) ext2fs_crc32_be((x), (y), (z))
+#define spin_lock_init(x)
+#define spin_lock(x)
+#define spin_unlock(x)
+#define yield()
+#define SLAB_HWCACHE_ALIGN 0
+#define SLAB_TEMPORARY 0
+#define KMEM_CACHE(__struct, __flags) kmem_cache_create(#__struct,\
+ sizeof(struct __struct), __alignof__(struct __struct),\
+ (__flags), NULL)
+
+#define blkdev_issue_flush(kdev, a, b) sync_blockdev(kdev)
+#define is_power_of_2(x) ((x) != 0 && (((x) & ((x) - 1)) == 0))
struct journal_s
{
@@ -63,7 +75,8 @@ struct journal_s
tid_t j_tail_sequence;
tid_t j_transaction_sequence;
__u8 j_uuid[16];
- struct jbd_revoke_table_s *j_revoke;
+ struct jbd2_revoke_table_s *j_revoke;
+ struct jbd2_revoke_table_s *j_revoke_table[2];
tid_t j_failed_commit;
__u32 j_csum_seed;
};
diff --git a/lib/ext2fs/kernel-jbd.h b/lib/ext2fs/kernel-jbd.h
index a9cdc30..407f4a5 100644
--- a/lib/ext2fs/kernel-jbd.h
+++ b/lib/ext2fs/kernel-jbd.h
@@ -365,6 +365,7 @@ static inline struct journal_head *bh2jh(struct buffer_head *bh)
}
struct jbd_revoke_table_s;
+struct jbd2_revoke_table_s;
/* The handle_t type represents a single atomic update being performed
* by some process. All filesystem modifications made by the process go
@@ -892,8 +893,8 @@ extern void journal_destroy_revoke_caches(void);
extern int journal_init_revoke_caches(void);
/* Recovery revoke support */
-extern int journal_set_revoke(journal_t *, unsigned long, tid_t);
-extern int journal_test_revoke(journal_t *, unsigned long, tid_t);
+extern int journal_set_revoke(journal_t *, unsigned long long, tid_t);
+extern int journal_test_revoke(journal_t *, unsigned long long, tid_t);
extern void journal_clear_revoke(journal_t *);
extern void journal_brelse_array(struct buffer_head *b[], int n);
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists