#include "buffer_head_io.h"
+/*
+ * Bits on bh->b_state used by ocfs2.
+ *
+ * These MUST be after the JBD2 bits. Hence, we use BH_JBDPrivateStart.
+ */
+enum ocfs2_state_bits {
+ BH_NeedsValidate = BH_JBDPrivateStart,
+};
+
+/* Expand the magic b_state functions */
+BUFFER_FNS(NeedsValidate, needs_validate);
+
int ocfs2_write_block(struct ocfs2_super *osb, struct buffer_head *bh,
- struct inode *inode)
+ struct ocfs2_caching_info *ci)
{
int ret = 0;
- mlog_entry("(bh->b_blocknr = %llu, inode=%p)\n",
- (unsigned long long)bh->b_blocknr, inode);
+ mlog_entry("(bh->b_blocknr = %llu, ci=%p)\n",
+ (unsigned long long)bh->b_blocknr, ci);
BUG_ON(bh->b_blocknr < OCFS2_SUPER_BLOCK_BLKNO);
BUG_ON(buffer_jbd(bh));
goto out;
}
- mutex_lock(&OCFS2_I(inode)->ip_io_mutex);
+ ocfs2_metadata_cache_io_lock(ci);
lock_buffer(bh);
set_buffer_uptodate(bh);
/* remove from dirty list before I/O. */
clear_buffer_dirty(bh);
- get_bh(bh); /* for end_buffer_write_sync() */
+ get_bh(bh); /* for end_buffer_write_sync() */
bh->b_end_io = end_buffer_write_sync;
submit_bh(WRITE, bh);
wait_on_buffer(bh);
if (buffer_uptodate(bh)) {
- ocfs2_set_buffer_uptodate(inode, bh);
+ ocfs2_set_buffer_uptodate(ci, bh);
} else {
/* We don't need to remove the clustered uptodate
* information for this bh as it's not marked locally
* uptodate. */
ret = -EIO;
- brelse(bh);
+ put_bh(bh);
}
- mutex_unlock(&OCFS2_I(inode)->ip_io_mutex);
+ ocfs2_metadata_cache_io_unlock(ci);
out:
mlog_exit(ret);
return ret;
}
-int ocfs2_read_blocks(struct ocfs2_super *osb, u64 block, int nr,
+int ocfs2_read_blocks_sync(struct ocfs2_super *osb, u64 block,
+ unsigned int nr, struct buffer_head *bhs[])
+{
+ int status = 0;
+ unsigned int i;
+ struct buffer_head *bh;
+
+ if (!nr) {
+ mlog(ML_BH_IO, "No buffers will be read!\n");
+ goto bail;
+ }
+
+ for (i = 0 ; i < nr ; i++) {
+ if (bhs[i] == NULL) {
+ bhs[i] = sb_getblk(osb->sb, block++);
+ if (bhs[i] == NULL) {
+ status = -EIO;
+ mlog_errno(status);
+ goto bail;
+ }
+ }
+ bh = bhs[i];
+
+ if (buffer_jbd(bh)) {
+ mlog(ML_BH_IO,
+ "trying to sync read a jbd "
+ "managed bh (blocknr = %llu), skipping\n",
+ (unsigned long long)bh->b_blocknr);
+ continue;
+ }
+
+ if (buffer_dirty(bh)) {
+ /* This should probably be a BUG, or
+ * at least return an error. */
+ mlog(ML_ERROR,
+ "trying to sync read a dirty "
+ "buffer! (blocknr = %llu), skipping\n",
+ (unsigned long long)bh->b_blocknr);
+ continue;
+ }
+
+ lock_buffer(bh);
+ if (buffer_jbd(bh)) {
+ mlog(ML_ERROR,
+ "block %llu had the JBD bit set "
+ "while I was in lock_buffer!",
+ (unsigned long long)bh->b_blocknr);
+ BUG();
+ }
+
+ clear_buffer_uptodate(bh);
+ get_bh(bh); /* for end_buffer_read_sync() */
+ bh->b_end_io = end_buffer_read_sync;
+ submit_bh(READ, bh);
+ }
+
+ for (i = nr; i > 0; i--) {
+ bh = bhs[i - 1];
+
+ /* No need to wait on the buffer if it's managed by JBD. */
+ if (!buffer_jbd(bh))
+ wait_on_buffer(bh);
+
+ if (!buffer_uptodate(bh)) {
+ /* Status won't be cleared from here on out,
+ * so we can safely record this and loop back
+ * to cleanup the other buffers. */
+ status = -EIO;
+ put_bh(bh);
+ bhs[i - 1] = NULL;
+ }
+ }
+
+bail:
+ return status;
+}
+
+int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
struct buffer_head *bhs[], int flags,
- struct inode *inode)
+ int (*validate)(struct super_block *sb,
+ struct buffer_head *bh))
{
int status = 0;
- struct super_block *sb;
int i, ignore_cache = 0;
struct buffer_head *bh;
+ struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
- mlog_entry("(block=(%llu), nr=(%d), flags=%d, inode=%p)\n",
- (unsigned long long)block, nr, flags, inode);
+ mlog_entry("(ci=%p, block=(%llu), nr=(%d), flags=%d)\n",
+ ci, (unsigned long long)block, nr, flags);
+ BUG_ON(!ci);
BUG_ON((flags & OCFS2_BH_READAHEAD) &&
- (!inode || !(flags & OCFS2_BH_CACHED)));
+ (flags & OCFS2_BH_IGNORE_CACHE));
- if (osb == NULL || osb->sb == NULL || bhs == NULL) {
+ if (bhs == NULL) {
status = -EINVAL;
mlog_errno(status);
goto bail;
goto bail;
}
- sb = osb->sb;
-
- if (flags & OCFS2_BH_CACHED && !inode)
- flags &= ~OCFS2_BH_CACHED;
-
- if (inode)
- mutex_lock(&OCFS2_I(inode)->ip_io_mutex);
+ ocfs2_metadata_cache_io_lock(ci);
for (i = 0 ; i < nr ; i++) {
if (bhs[i] == NULL) {
bhs[i] = sb_getblk(sb, block++);
if (bhs[i] == NULL) {
- if (inode)
- mutex_unlock(&OCFS2_I(inode)->ip_io_mutex);
+ ocfs2_metadata_cache_io_unlock(ci);
status = -EIO;
mlog_errno(status);
goto bail;
}
}
bh = bhs[i];
- ignore_cache = 0;
+ ignore_cache = (flags & OCFS2_BH_IGNORE_CACHE);
/* There are three read-ahead cases here which we need to
* be concerned with. All three assume a buffer has
* before our is-it-in-flight check.
*/
- if (flags & OCFS2_BH_CACHED &&
- !ocfs2_buffer_uptodate(inode, bh)) {
+ if (!ignore_cache && !ocfs2_buffer_uptodate(ci, bh)) {
mlog(ML_UPTODATE,
- "bh (%llu), inode %llu not uptodate\n",
+ "bh (%llu), owner %llu not uptodate\n",
(unsigned long long)bh->b_blocknr,
- (unsigned long long)OCFS2_I(inode)->ip_blkno);
+ (unsigned long long)ocfs2_metadata_cache_owner(ci));
+ /* We're using ignore_cache here to say
+ * "go to disk" */
ignore_cache = 1;
}
- /* XXX: Can we ever get this and *not* have the cached
- * flag set? */
if (buffer_jbd(bh)) {
- if (!(flags & OCFS2_BH_CACHED) || ignore_cache)
+ if (ignore_cache)
mlog(ML_BH_IO, "trying to sync read a jbd "
"managed bh (blocknr = %llu)\n",
(unsigned long long)bh->b_blocknr);
continue;
}
- if (!(flags & OCFS2_BH_CACHED) || ignore_cache) {
+ if (ignore_cache) {
if (buffer_dirty(bh)) {
/* This should probably be a BUG, or
* at least return an error. */
* previously submitted request than we are
* done here. */
if ((flags & OCFS2_BH_READAHEAD)
- && ocfs2_buffer_read_ahead(inode, bh))
+ && ocfs2_buffer_read_ahead(ci, bh))
continue;
lock_buffer(bh);
* previously read-ahead buffer may have
* completed I/O while we were waiting for the
* buffer lock. */
- if ((flags & OCFS2_BH_CACHED)
+ if (!(flags & OCFS2_BH_IGNORE_CACHE)
&& !(flags & OCFS2_BH_READAHEAD)
- && ocfs2_buffer_uptodate(inode, bh)) {
+ && ocfs2_buffer_uptodate(ci, bh)) {
unlock_buffer(bh);
continue;
}
clear_buffer_uptodate(bh);
get_bh(bh); /* for end_buffer_read_sync() */
+ if (validate)
+ set_buffer_needs_validate(bh);
bh->b_end_io = end_buffer_read_sync;
submit_bh(READ, bh);
continue;
if (!(flags & OCFS2_BH_READAHEAD)) {
/* We know this can't have changed as we hold the
- * inode sem. Avoid doing any work on the bh if the
+ * owner sem. Avoid doing any work on the bh if the
* journal has it. */
if (!buffer_jbd(bh))
wait_on_buffer(bh);
* for this bh as it's not marked locally
* uptodate. */
status = -EIO;
- brelse(bh);
+ put_bh(bh);
bhs[i] = NULL;
continue;
}
+
+ if (buffer_needs_validate(bh)) {
+ /* We never set NeedsValidate if the
+ * buffer was held by the journal, so
+ * that better not have changed */
+ BUG_ON(buffer_jbd(bh));
+ clear_buffer_needs_validate(bh);
+ status = validate(sb, bh);
+ if (status) {
+ put_bh(bh);
+ bhs[i] = NULL;
+ continue;
+ }
+ }
}
/* Always set the buffer in the cache, even if it was
* a forced read, or read-ahead which hasn't yet
* completed. */
- if (inode)
- ocfs2_set_buffer_uptodate(inode, bh);
+ ocfs2_set_buffer_uptodate(ci, bh);
}
- if (inode)
- mutex_unlock(&OCFS2_I(inode)->ip_io_mutex);
+ ocfs2_metadata_cache_io_unlock(ci);
mlog(ML_BH_IO, "block=(%llu), nr=(%d), cached=%s, flags=0x%x\n",
(unsigned long long)block, nr,
- (!(flags & OCFS2_BH_CACHED) || ignore_cache) ? "no" : "yes", flags);
+ ((flags & OCFS2_BH_IGNORE_CACHE) || ignore_cache) ? "no" : "yes",
+ flags);
bail:
mlog_exit(status);
return status;
}
+
+/* Check whether the blkno is the super block or one of the backups. */
+static void ocfs2_check_super_or_backup(struct super_block *sb,
+ sector_t blkno)
+{
+ int i;
+ u64 backup_blkno;
+
+ if (blkno == OCFS2_SUPER_BLOCK_BLKNO)
+ return;
+
+ for (i = 0; i < OCFS2_MAX_BACKUP_SUPERBLOCKS; i++) {
+ backup_blkno = ocfs2_backup_super_blkno(sb, i);
+ if (backup_blkno == blkno)
+ return;
+ }
+
+ BUG();
+}
+
+/*
+ * Write super block and backups doesn't need to collaborate with journal,
+ * so we don't need to lock ip_io_mutex and ci doesn't need to bea passed
+ * into this function.
+ */
+int ocfs2_write_super_or_backup(struct ocfs2_super *osb,
+ struct buffer_head *bh)
+{
+ int ret = 0;
+
+ mlog_entry_void();
+
+ BUG_ON(buffer_jbd(bh));
+ ocfs2_check_super_or_backup(osb->sb, bh->b_blocknr);
+
+ if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) {
+ ret = -EROFS;
+ goto out;
+ }
+
+ lock_buffer(bh);
+ set_buffer_uptodate(bh);
+
+ /* remove from dirty list before I/O. */
+ clear_buffer_dirty(bh);
+
+ get_bh(bh); /* for end_buffer_write_sync() */
+ bh->b_end_io = end_buffer_write_sync;
+ submit_bh(WRITE, bh);
+
+ wait_on_buffer(bh);
+
+ if (!buffer_uptodate(bh)) {
+ ret = -EIO;
+ put_bh(bh);
+ }
+
+out:
+ mlog_exit(ret);
+ return ret;
+}
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff