* Copyright (C) 1991, 1992 Linus Torvalds
*
* Goal-directed block allocation by Stephen Tweedie
- * (sct@redhat.com), 1993, 1998
+ * (sct@redhat.com), 1993, 1998
* Big-endian to little-endian byte-swapping/bitmaps by
* David S. Miller (davem@caip.rutgers.edu), 1995
* 64-bit file support on 64-bit platforms by Jakub Jelinek
- * (jj@sunsite.ms.mff.cuni.cz)
+ * (jj@sunsite.ms.mff.cuni.cz)
*
* Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000
*/
#include <linux/writeback.h>
#include <linux/mpage.h>
#include <linux/uio.h>
-#include <linux/vserver/xid.h>
+#include <linux/bio.h>
+#include <linux/vs_tag.h>
#include "xattr.h"
#include "acl.h"
/*
* The ext3 forget function must perform a revoke if we are freeing data
* which has been journaled. Metadata (eg. indirect blocks) must be
- * revoked in all cases.
+ * revoked in all cases.
*
* "bh" may be NULL: a metadata block may have been freed from memory
* but there may still be a record of it in the journal, and that record
* still needs to be revoked.
*/
int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode,
- struct buffer_head *bh, int blocknr)
+ struct buffer_head *bh, ext3_fsblk_t blocknr)
{
int err;
* Work out how many blocks we need to proceed with the next chunk of a
* truncate transaction.
*/
-static unsigned long blocks_for_truncate(struct inode *inode)
+static unsigned long blocks_for_truncate(struct inode *inode)
{
unsigned long needed;
/* But we need to bound the transaction so we don't overflow the
* journal. */
- if (needed > EXT3_MAX_TRANS_DATA)
+ if (needed > EXT3_MAX_TRANS_DATA)
needed = EXT3_MAX_TRANS_DATA;
return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed;
}
-/*
+/*
* Truncate transactions can be complex and absolutely huge. So we need to
* be able to restart the transaction at a conventient checkpoint to make
* sure we don't overflow the journal.
* start_transaction gets us a new handle for a truncate transaction,
* and extend_transaction tries to extend the existing one a bit. If
* extend fails, we need to propagate the failure up and restart the
- * transaction in the top-level truncate loop. --sct
+ * transaction in the top-level truncate loop. --sct
*/
-static handle_t *start_transaction(struct inode *inode)
+static handle_t *start_transaction(struct inode *inode)
{
handle_t *result;
ext3_orphan_del(handle, inode);
EXT3_I(inode)->i_dtime = get_seconds();
- /*
+ /*
* One subtle ordering requirement: if anything has gone wrong
* (transaction abort, IO errors, whatever), then we can still
* do these next steps (the fs will already have been marked as
* having errors), but we can't free the inode if the mark_dirty
- * fails.
+ * fails.
*/
if (ext3_mark_inode_dirty(handle, inode))
/* If that failed, just do the required in-core inode clear. */
* + if there is a block to the left of our position - allocate near it.
* + if pointer will live in indirect block - allocate near that block.
* + if pointer will live in inode - allocate in the same
- * cylinder group.
+ * cylinder group.
*
* In the latter case we colour the starting block by the callers PID to
* prevent it from clashing with concurrent allocations for a different inode
*
* Caller must make sure that @ind is valid and will stay that way.
*/
-static unsigned long ext3_find_near(struct inode *inode, Indirect *ind)
+static ext3_fsblk_t ext3_find_near(struct inode *inode, Indirect *ind)
{
struct ext3_inode_info *ei = EXT3_I(inode);
__le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data;
__le32 *p;
- unsigned long bg_start;
- unsigned long colour;
+ ext3_fsblk_t bg_start;
+ ext3_grpblk_t colour;
/* Try to find previous block */
for (p = ind->p - 1; p >= start; p--) {
* It is going to be referred to from the inode itself? OK, just put it
* into the same cylinder group then.
*/
- bg_start = (ei->i_block_group * EXT3_BLOCKS_PER_GROUP(inode->i_sb)) +
- le32_to_cpu(EXT3_SB(inode->i_sb)->s_es->s_first_data_block);
+ bg_start = ext3_group_first_block_no(inode->i_sb, ei->i_block_group);
colour = (current->pid % 16) *
(EXT3_BLOCKS_PER_GROUP(inode->i_sb) / 16);
return bg_start + colour;
* stores it in *@goal and returns zero.
*/
-static unsigned long ext3_find_goal(struct inode *inode, long block,
+static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block,
Indirect chain[4], Indirect *partial)
{
struct ext3_block_alloc_info *block_i;
* ext3_blks_to_allocate: Look up the block map and count the number
* of direct blocks need to be allocated for the given branch.
*
- * @branch: chain of indirect blocks
+ * @branch: chain of indirect blocks
* @k: number of blocks need for indirect blocks
* @blks: number of data blocks to be mapped.
* @blocks_to_boundary: the offset in the indirect block
* direct blocks
*/
static int ext3_alloc_blocks(handle_t *handle, struct inode *inode,
- unsigned long goal, int indirect_blks, int blks,
- unsigned long long new_blocks[4], int *err)
+ ext3_fsblk_t goal, int indirect_blks, int blks,
+ ext3_fsblk_t new_blocks[4], int *err)
{
int target, i;
unsigned long count = 0;
int index = 0;
- unsigned long current_block = 0;
+ ext3_fsblk_t current_block = 0;
int ret = 0;
/*
* as described above and return 0.
*/
static int ext3_alloc_branch(handle_t *handle, struct inode *inode,
- int indirect_blks, int *blks, unsigned long goal,
+ int indirect_blks, int *blks, ext3_fsblk_t goal,
int *offsets, Indirect *branch)
{
int blocksize = inode->i_sb->s_blocksize;
int err = 0;
struct buffer_head *bh;
int num;
- unsigned long long new_blocks[4];
- unsigned long long current_block;
+ ext3_fsblk_t new_blocks[4];
+ ext3_fsblk_t current_block;
num = ext3_alloc_blocks(handle, inode, goal, indirect_blks,
*blks, new_blocks, &err);
int i;
int err = 0;
struct ext3_block_alloc_info *block_i;
- unsigned long current_block;
+ ext3_fsblk_t current_block;
block_i = EXT3_I(inode)->i_block_alloc_info;
/*
jbd_debug(5, "splicing indirect only\n");
BUFFER_TRACE(where->bh, "call ext3_journal_dirty_metadata");
err = ext3_journal_dirty_metadata(handle, where->bh);
- if (err)
+ if (err)
goto err_out;
} else {
/*
int offsets[4];
Indirect chain[4];
Indirect *partial;
- unsigned long goal;
+ ext3_fsblk_t goal;
int indirect_blks;
int blocks_to_boundary = 0;
int depth;
struct ext3_inode_info *ei = EXT3_I(inode);
int count = 0;
- unsigned long first_block = 0;
+ ext3_fsblk_t first_block = 0;
J_ASSERT(handle != NULL || create == 0);
count++;
/*map more blocks*/
while (count < maxblocks && count <= blocks_to_boundary) {
- unsigned long blk;
+ ext3_fsblk_t blk;
if (!verify_chain(chain, partial)) {
/*
set_buffer_new(bh_result);
got_it:
map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key));
- if (blocks_to_boundary == 0)
+ if (count > blocks_to_boundary)
set_buffer_boundary(bh_result);
err = count;
/* Clean up and exit */
buffer_trace_init(&dummy.b_history);
err = ext3_get_blocks_handle(handle, inode, block, 1,
&dummy, create, 1);
- if (err == 1) {
+ /*
+ * ext3_get_blocks_handle() returns number of blocks
+ * mapped. 0 in case of a HOLE.
+ */
+ if (err > 0) {
+ if (err > 1)
+ WARN_ON(1);
err = 0;
- } else if (err >= 0) {
- WARN_ON(1);
- err = -EIO;
}
*errp = err;
if (!err && buffer_mapped(&dummy)) {
return bh;
if (buffer_uptodate(bh))
return bh;
- ll_rw_block(READ, 1, &bh);
+ ll_rw_block(READ_META, 1, &bh);
wait_on_buffer(bh);
if (buffer_uptodate(bh))
return bh;
for ( bh = head, block_start = 0;
ret == 0 && (bh != head || !block_start);
- block_start = block_end, bh = next)
+ block_start = block_end, bh = next)
{
next = bh->b_this_page;
block_end = block_start + blocksize;
* So what we do is to rely on the fact that journal_stop/journal_start
* will _not_ run commit under these circumstances because handle->h_ref
* is elevated. We'll still have enough credits for the tiny quotafile
- * write.
+ * write.
*/
static int do_journal_get_write_access(handle_t *handle,
struct buffer_head *bh)
ret = PTR_ERR(handle);
goto out;
}
- if (test_opt(inode->i_sb, NOBH))
+ if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode))
ret = nobh_prepare_write(page, from, to, ext3_get_block);
else
ret = block_prepare_write(page, from, to, ext3_get_block);
if (new_i_size > EXT3_I(inode)->i_disksize)
EXT3_I(inode)->i_disksize = new_i_size;
- if (test_opt(inode->i_sb, NOBH))
+ if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode))
ret = nobh_commit_write(file, page, from, to);
else
ret = generic_commit_write(file, page, from, to);
if (inode->i_size > EXT3_I(inode)->i_disksize) {
EXT3_I(inode)->i_disksize = inode->i_size;
ret2 = ext3_mark_inode_dirty(handle, inode);
- if (!ret)
+ if (!ret)
ret = ret2;
}
ret2 = ext3_journal_stop(handle);
return ret;
}
-/*
+/*
* bmap() is special. It gets used by applications such as lilo and by
* the swapper to find the on-disk block of a specific piece of data.
*
* filesystem and enables swap, then they may get a nasty shock when the
* data getting swapped to that swapfile suddenly gets overwritten by
* the original zero's written out previously to the journal and
- * awaiting writeback in the kernel's buffer cache.
+ * awaiting writeback in the kernel's buffer cache.
*
* So, if we see any bmap calls here on a modified, data-journaled file,
- * take extra steps to flush any blocks which might be in the cache.
+ * take extra steps to flush any blocks which might be in the cache.
*/
static sector_t ext3_bmap(struct address_space *mapping, sector_t block)
{
int err;
if (EXT3_I(inode)->i_state & EXT3_STATE_JDATA) {
- /*
+ /*
* This is a REALLY heavyweight approach, but the use of
* bmap on dirty files is expected to be extremely rare:
* only if we run lilo or swapon on a freshly made file
- * do we expect this to happen.
+ * do we expect this to happen.
*
* (bmap requires CAP_SYS_RAWIO so this does not
* represent an unprivileged user DOS attack --- we'd be
* in trouble if mortal users could trigger this path at
- * will.)
+ * will.)
*
* NB. EXT3_STATE_JDATA is not set on files other than
* regular files. If somebody wants to bmap a directory
*/
/*
- * And attach them to the current transaction. But only if
+ * And attach them to the current transaction. But only if
* block_write_full_page() succeeded. Otherwise they are unmapped,
* and generally junk.
*/
goto out_fail;
}
- if (test_opt(inode->i_sb, NOBH))
+ if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode))
ret = nobh_writepage(page, ext3_get_block, wbc);
else
ret = block_write_full_page(page, ext3_get_block, wbc);
}
}
- ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
+ ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
offset, nr_segs,
ext3_get_block, NULL);
return __set_page_dirty_nobuffers(page);
}
-static struct address_space_operations ext3_ordered_aops = {
+static const struct address_space_operations ext3_ordered_aops = {
.readpage = ext3_readpage,
.readpages = ext3_readpages,
.writepage = ext3_ordered_writepage,
.migratepage = buffer_migrate_page,
};
-static struct address_space_operations ext3_writeback_aops = {
+static const struct address_space_operations ext3_writeback_aops = {
.readpage = ext3_readpage,
.readpages = ext3_readpages,
.writepage = ext3_writeback_writepage,
.migratepage = buffer_migrate_page,
};
-static struct address_space_operations ext3_journalled_aops = {
+static const struct address_space_operations ext3_journalled_aops = {
.readpage = ext3_readpage,
.readpages = ext3_readpages,
.writepage = ext3_journalled_writepage,
static int ext3_block_truncate_page(handle_t *handle, struct page *page,
struct address_space *mapping, loff_t from)
{
- unsigned long index = from >> PAGE_CACHE_SHIFT;
+ ext3_fsblk_t index = from >> PAGE_CACHE_SHIFT;
unsigned offset = from & (PAGE_CACHE_SIZE-1);
unsigned blocksize, iblock, length, pos;
struct inode *inode = mapping->host;
* than `count' because there can be holes in there.
*/
static void ext3_clear_blocks(handle_t *handle, struct inode *inode,
- struct buffer_head *bh, unsigned long block_to_free,
+ struct buffer_head *bh, ext3_fsblk_t block_to_free,
unsigned long count, __le32 *first, __le32 *last)
{
__le32 *p;
struct buffer_head *this_bh,
__le32 *first, __le32 *last)
{
- unsigned long block_to_free = 0; /* Starting block # of a run */
- unsigned long count = 0; /* Number of blocks in the run */
+ ext3_fsblk_t block_to_free = 0; /* Starting block # of a run */
+ unsigned long count = 0; /* Number of blocks in the run */
__le32 *block_to_free_p = NULL; /* Pointer into inode/ind
corresponding to
block_to_free */
- unsigned long nr; /* Current block # */
+ ext3_fsblk_t nr; /* Current block # */
__le32 *p; /* Pointer into inode/ind
for current block */
int err;
} else if (nr == block_to_free + count) {
count++;
} else {
- ext3_clear_blocks(handle, inode, this_bh,
+ ext3_clear_blocks(handle, inode, this_bh,
block_to_free,
count, block_to_free_p, p);
block_to_free = nr;
struct buffer_head *parent_bh,
__le32 *first, __le32 *last, int depth)
{
- unsigned long nr;
+ ext3_fsblk_t nr;
__le32 *p;
if (is_handle_aborted(handle))
*/
if (!bh) {
ext3_error(inode->i_sb, "ext3_free_branches",
- "Read failure, inode=%ld, block=%ld",
+ "Read failure, inode=%lu, block="E3FSBLK,
inode->i_ino, nr);
continue;
}
*p = 0;
BUFFER_TRACE(parent_bh,
"call ext3_journal_dirty_metadata");
- ext3_journal_dirty_metadata(handle,
+ ext3_journal_dirty_metadata(handle,
parent_bh);
}
}
return;
if (ext3_inode_is_fast_symlink(inode))
return;
- if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
+ if (IS_APPEND(inode) || IS_IXORUNLINK(inode))
return;
/*
ext3_journal_stop(handle);
}
-static unsigned long ext3_get_inode_block(struct super_block *sb,
+static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb,
unsigned long ino, struct ext3_iloc *iloc)
{
unsigned long desc, group_desc, block_group;
- unsigned long offset, block;
+ unsigned long offset;
+ ext3_fsblk_t block;
struct buffer_head *bh;
struct ext3_group_desc * gdp;
-
- if ((ino != EXT3_ROOT_INO && ino != EXT3_JOURNAL_INO &&
- ino != EXT3_RESIZE_INO && ino < EXT3_FIRST_INO(sb)) ||
- ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) {
- ext3_error(sb, "ext3_get_inode_block",
- "bad inode number: %lu", ino);
+ if (!ext3_valid_inum(sb, ino)) {
+ /*
+ * This error is already checked for in namei.c unless we are
+ * looking at an NFS filehandle, in which case no error
+ * report is needed
+ */
return 0;
}
+
block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb);
if (block_group >= EXT3_SB(sb)->s_groups_count) {
ext3_error(sb,"ext3_get_inode_block","group >= groups count");
static int __ext3_get_inode_loc(struct inode *inode,
struct ext3_iloc *iloc, int in_mem)
{
- unsigned long block;
+ ext3_fsblk_t block;
struct buffer_head *bh;
block = ext3_get_inode_block(inode->i_sb, inode->i_ino, iloc);
if (!bh) {
ext3_error (inode->i_sb, "ext3_get_inode_loc",
"unable to read inode block - "
- "inode=%lu, block=%lu", inode->i_ino, block);
+ "inode=%lu, block="E3FSBLK,
+ inode->i_ino, block);
return -EIO;
}
if (!buffer_uptodate(bh)) {
*/
get_bh(bh);
bh->b_end_io = end_buffer_read_sync;
- submit_bh(READ, bh);
+ submit_bh(READ_META, bh);
wait_on_buffer(bh);
if (!buffer_uptodate(bh)) {
ext3_error(inode->i_sb, "ext3_get_inode_loc",
"unable to read inode block - "
- "inode=%lu, block=%lu",
+ "inode=%lu, block="E3FSBLK,
inode->i_ino, block);
brelse(bh);
return -EIO;
int err = 0;
oldflags = EXT3_I(inode)->i_flags;
- newflags = oldflags & ~(EXT3_APPEND_FL |
- EXT3_IMMUTABLE_FL | EXT3_IUNLINK_FL |
- EXT3_BARRIER_FL | EXT3_NOATIME_FL |
- EXT3_SYNC_FL | EXT3_DIRSYNC_FL);
+ newflags = oldflags & ~(EXT3_IMMUTABLE_FL |
+ EXT3_IUNLINK_FL | EXT3_BARRIER_FL);
- if (IS_APPEND(inode))
- newflags |= EXT3_APPEND_FL;
if (IS_IMMUTABLE(inode))
newflags |= EXT3_IMMUTABLE_FL;
if (IS_IUNLINK(inode))
if (IS_BARRIER(inode))
newflags |= EXT3_BARRIER_FL;
- /* we do not want to copy superblock flags */
- if (inode->i_flags & S_NOATIME)
- newflags |= EXT3_NOATIME_FL;
- if (inode->i_flags & S_SYNC)
- newflags |= EXT3_SYNC_FL;
- if (inode->i_flags & S_DIRSYNC)
- newflags |= EXT3_DIRSYNC_FL;
-
if (oldflags ^ newflags) {
handle_t *handle;
struct ext3_iloc iloc;
uid |= le16_to_cpu(raw_inode->i_uid_high) << 16;
gid |= le16_to_cpu(raw_inode->i_gid_high) << 16;
}
- inode->i_uid = INOXID_UID(XID_TAG(inode), uid, gid);
- inode->i_gid = INOXID_GID(XID_TAG(inode), uid, gid);
- inode->i_xid = INOXID_XID(XID_TAG(inode), uid, gid,
- le16_to_cpu(raw_inode->i_raw_xid));
+ inode->i_uid = INOTAG_UID(DX_TAG(inode), uid, gid);
+ inode->i_gid = INOTAG_GID(DX_TAG(inode), uid, gid);
+ inode->i_tag = INOTAG_TAG(DX_TAG(inode), uid, gid,
+ le16_to_cpu(raw_inode->i_raw_tag));
inode->i_nlink = le16_to_cpu(raw_inode->i_links_count);
inode->i_size = le32_to_cpu(raw_inode->i_size);
* recovery code: that's fine, we're about to complete
* the process of deleting those. */
}
- inode->i_blksize = PAGE_SIZE; /* This is the optimal IO size
- * (for stat), not the fs block
- * size */
inode->i_blocks = le32_to_cpu(raw_inode->i_blocks);
ei->i_flags = le32_to_cpu(raw_inode->i_flags);
#ifdef EXT3_FRAGMENTS
if (raw_inode->i_block[0])
init_special_inode(inode, inode->i_mode,
old_decode_dev(le32_to_cpu(raw_inode->i_block[0])));
- else
+ else
init_special_inode(inode, inode->i_mode,
new_decode_dev(le32_to_cpu(raw_inode->i_block[1])));
}
*
* The caller must have write access to iloc->bh.
*/
-static int ext3_do_update_inode(handle_t *handle,
- struct inode *inode,
+static int ext3_do_update_inode(handle_t *handle,
+ struct inode *inode,
struct ext3_iloc *iloc)
{
struct ext3_inode *raw_inode = ext3_raw_inode(iloc);
struct ext3_inode_info *ei = EXT3_I(inode);
struct buffer_head *bh = iloc->bh;
- uid_t uid = XIDINO_UID(XID_TAG(inode), inode->i_uid, inode->i_xid);
- gid_t gid = XIDINO_GID(XID_TAG(inode), inode->i_gid, inode->i_xid);
+ uid_t uid = TAGINO_UID(DX_TAG(inode), inode->i_uid, inode->i_tag);
+ gid_t gid = TAGINO_GID(DX_TAG(inode), inode->i_gid, inode->i_tag);
int err = 0, rc, block;
/* For fields not not tracking in the in-memory inode,
raw_inode->i_uid_high = 0;
raw_inode->i_gid_high = 0;
}
-#ifdef CONFIG_INOXID_INTERN
- raw_inode->i_raw_xid = cpu_to_le16(inode->i_xid);
+#ifdef CONFIG_TAGGING_INTERN
+ raw_inode->i_raw_tag = cpu_to_le16(inode->i_tag);
#endif
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
raw_inode->i_size = cpu_to_le32(ei->i_disksize);
* commit will leave the blocks being flushed in an unused state on
* disk. (On recovery, the inode will get truncated and the blocks will
* be freed, so we have a strong guarantee that no future commit will
- * leave these blocks visible to the user.)
+ * leave these blocks visible to the user.)
*
* Called with inode->sem down.
*/
if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
(ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid) ||
- (ia_valid & ATTR_XID && attr->ia_xid != inode->i_xid)) {
+ (ia_valid & ATTR_TAG && attr->ia_tag != inode->i_tag)) {
handle_t *handle;
/* (user+group)*(old+new) structure, inode write (sb,
inode->i_uid = attr->ia_uid;
if (attr->ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
- if ((attr->ia_valid & ATTR_XID) && IS_TAGXID(inode))
- inode->i_xid = attr->ia_xid;
+ if ((attr->ia_valid & ATTR_TAG) && IS_TAGGED(inode))
+ inode->i_tag = attr->ia_tag;
error = ext3_mark_inode_dirty(handle, inode);
ext3_journal_stop(handle);
}
return err;
}
-/*
+/*
* On success, We end up with an outstanding reference count against
- * iloc->bh. This _must_ be cleaned up later.
+ * iloc->bh. This _must_ be cleaned up later.
*/
int
-ext3_reserve_inode_write(handle_t *handle, struct inode *inode,
+ext3_reserve_inode_write(handle_t *handle, struct inode *inode,
struct ext3_iloc *iloc)
{
int err = 0;
}
#if 0
-/*
+/*
* Bind an inode's backing buffer_head into this transaction, to prevent
* it from being flushed to disk early. Unlike
* ext3_reserve_inode_write, this leaves behind no bh reference and
BUFFER_TRACE(iloc.bh, "get_write_access");
err = journal_get_write_access(handle, iloc.bh);
if (!err)
- err = ext3_journal_dirty_metadata(handle,
+ err = ext3_journal_dirty_metadata(handle,
iloc.bh);
brelse(iloc.bh);
}
git://git.onelab.eu / linux-2.6.git / blobdiff