linux/fs/ext4/indirect.c
Zheng Liu 8cde7ad17e ext4: fix big-endian bugs which could cause fs corruptions
When an extent was zeroed out, we forgot to do convert from cpu to le16.
It could make us hit a BUG_ON when we try to write dirty pages out.  So
fix it.

[ Also fix a bug found by Dmitry Monakhov where we were missing
  le32_to_cpu() calls in the new indirect punch hole code.

  There are a number of other big endian warnings found by static code
  analyzers, but we'll wait for the next merge window to fix them all
  up.  These fixes are designed to be Obviously Correct by code
  inspection, and easy to demonstrate that it won't make any
  difference (and hence, won't introduce any bugs) on little endian
  architectures such as x86.  --tytso ]

Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reported-by: CAI Qian <caiqian@redhat.com>
Reported-by: Christian Kujau <lists@nerdbynature.de>
Cc: Dmitry Monakhov <dmonakhov@openvz.org>
2013-04-03 12:37:17 -04:00

1761 lines
51 KiB
C

/*
* linux/fs/ext4/indirect.c
*
* from
*
* linux/fs/ext4/inode.c
*
* Copyright (C) 1992, 1993, 1994, 1995
* Remy Card (card@masi.ibp.fr)
* Laboratoire MASI - Institut Blaise Pascal
* Universite Pierre et Marie Curie (Paris VI)
*
* from
*
* linux/fs/minix/inode.c
*
* Copyright (C) 1991, 1992 Linus Torvalds
*
* Goal-directed block allocation by Stephen Tweedie
* (sct@redhat.com), 1993, 1998
*/
#include "ext4_jbd2.h"
#include "truncate.h"
#include "ext4_extents.h" /* Needed for EXT_MAX_BLOCKS */
#include <trace/events/ext4.h>
typedef struct {
__le32 *p;
__le32 key;
struct buffer_head *bh;
} Indirect;
static inline void add_chain(Indirect *p, struct buffer_head *bh, __le32 *v)
{
p->key = *(p->p = v);
p->bh = bh;
}
/**
* ext4_block_to_path - parse the block number into array of offsets
* @inode: inode in question (we are only interested in its superblock)
* @i_block: block number to be parsed
* @offsets: array to store the offsets in
* @boundary: set this non-zero if the referred-to block is likely to be
* followed (on disk) by an indirect block.
*
* To store the locations of file's data ext4 uses a data structure common
* for UNIX filesystems - tree of pointers anchored in the inode, with
* data blocks at leaves and indirect blocks in intermediate nodes.
* This function translates the block number into path in that tree -
* return value is the path length and @offsets[n] is the offset of
* pointer to (n+1)th node in the nth one. If @block is out of range
* (negative or too large) warning is printed and zero returned.
*
* Note: function doesn't find node addresses, so no IO is needed. All
* we need to know is the capacity of indirect blocks (taken from the
* inode->i_sb).
*/
/*
* Portability note: the last comparison (check that we fit into triple
* indirect block) is spelled differently, because otherwise on an
* architecture with 32-bit longs and 8Kb pages we might get into trouble
* if our filesystem had 8Kb blocks. We might use long long, but that would
* kill us on x86. Oh, well, at least the sign propagation does not matter -
* i_block would have to be negative in the very beginning, so we would not
* get there at all.
*/
static int ext4_block_to_path(struct inode *inode,
ext4_lblk_t i_block,
ext4_lblk_t offsets[4], int *boundary)
{
int ptrs = EXT4_ADDR_PER_BLOCK(inode->i_sb);
int ptrs_bits = EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb);
const long direct_blocks = EXT4_NDIR_BLOCKS,
indirect_blocks = ptrs,
double_blocks = (1 << (ptrs_bits * 2));
int n = 0;
int final = 0;
if (i_block < direct_blocks) {
offsets[n++] = i_block;
final = direct_blocks;
} else if ((i_block -= direct_blocks) < indirect_blocks) {
offsets[n++] = EXT4_IND_BLOCK;
offsets[n++] = i_block;
final = ptrs;
} else if ((i_block -= indirect_blocks) < double_blocks) {
offsets[n++] = EXT4_DIND_BLOCK;
offsets[n++] = i_block >> ptrs_bits;
offsets[n++] = i_block & (ptrs - 1);
final = ptrs;
} else if (((i_block -= double_blocks) >> (ptrs_bits * 2)) < ptrs) {
offsets[n++] = EXT4_TIND_BLOCK;
offsets[n++] = i_block >> (ptrs_bits * 2);
offsets[n++] = (i_block >> ptrs_bits) & (ptrs - 1);
offsets[n++] = i_block & (ptrs - 1);
final = ptrs;
} else {
ext4_warning(inode->i_sb, "block %lu > max in inode %lu",
i_block + direct_blocks +
indirect_blocks + double_blocks, inode->i_ino);
}
if (boundary)
*boundary = final - 1 - (i_block & (ptrs - 1));
return n;
}
/**
* ext4_get_branch - read the chain of indirect blocks leading to data
* @inode: inode in question
* @depth: depth of the chain (1 - direct pointer, etc.)
* @offsets: offsets of pointers in inode/indirect blocks
* @chain: place to store the result
* @err: here we store the error value
*
* Function fills the array of triples <key, p, bh> and returns %NULL
* if everything went OK or the pointer to the last filled triple
* (incomplete one) otherwise. Upon the return chain[i].key contains
* the number of (i+1)-th block in the chain (as it is stored in memory,
* i.e. little-endian 32-bit), chain[i].p contains the address of that
* number (it points into struct inode for i==0 and into the bh->b_data
* for i>0) and chain[i].bh points to the buffer_head of i-th indirect
* block for i>0 and NULL for i==0. In other words, it holds the block
* numbers of the chain, addresses they were taken from (and where we can
* verify that chain did not change) and buffer_heads hosting these
* numbers.
*
* Function stops when it stumbles upon zero pointer (absent block)
* (pointer to last triple returned, *@err == 0)
* or when it gets an IO error reading an indirect block
* (ditto, *@err == -EIO)
* or when it reads all @depth-1 indirect blocks successfully and finds
* the whole chain, all way to the data (returns %NULL, *err == 0).
*
* Need to be called with
* down_read(&EXT4_I(inode)->i_data_sem)
*/
static Indirect *ext4_get_branch(struct inode *inode, int depth,
ext4_lblk_t *offsets,
Indirect chain[4], int *err)
{
struct super_block *sb = inode->i_sb;
Indirect *p = chain;
struct buffer_head *bh;
int ret = -EIO;
*err = 0;
/* i_data is not going away, no lock needed */
add_chain(chain, NULL, EXT4_I(inode)->i_data + *offsets);
if (!p->key)
goto no_block;
while (--depth) {
bh = sb_getblk(sb, le32_to_cpu(p->key));
if (unlikely(!bh)) {
ret = -ENOMEM;
goto failure;
}
if (!bh_uptodate_or_lock(bh)) {
if (bh_submit_read(bh) < 0) {
put_bh(bh);
goto failure;
}
/* validate block references */
if (ext4_check_indirect_blockref(inode, bh)) {
put_bh(bh);
goto failure;
}
}
add_chain(++p, bh, (__le32 *)bh->b_data + *++offsets);
/* Reader: end */
if (!p->key)
goto no_block;
}
return NULL;
failure:
*err = ret;
no_block:
return p;
}
/**
* ext4_find_near - find a place for allocation with sufficient locality
* @inode: owner
* @ind: descriptor of indirect block.
*
* This function returns the preferred place for block allocation.
* It is used when heuristic for sequential allocation fails.
* Rules are:
* + 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.
*
* 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
* in the same block group. The PID is used here so that functionally related
* files will be close-by on-disk.
*
* Caller must make sure that @ind is valid and will stay that way.
*/
static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind)
{
struct ext4_inode_info *ei = EXT4_I(inode);
__le32 *start = ind->bh ? (__le32 *) ind->bh->b_data : ei->i_data;
__le32 *p;
/* Try to find previous block */
for (p = ind->p - 1; p >= start; p--) {
if (*p)
return le32_to_cpu(*p);
}
/* No such thing, so let's try location of indirect block */
if (ind->bh)
return ind->bh->b_blocknr;
/*
* It is going to be referred to from the inode itself? OK, just put it
* into the same cylinder group then.
*/
return ext4_inode_to_goal_block(inode);
}
/**
* ext4_find_goal - find a preferred place for allocation.
* @inode: owner
* @block: block we want
* @partial: pointer to the last triple within a chain
*
* Normally this function find the preferred place for block allocation,
* returns it.
* Because this is only used for non-extent files, we limit the block nr
* to 32 bits.
*/
static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t block,
Indirect *partial)
{
ext4_fsblk_t goal;
/*
* XXX need to get goal block from mballoc's data structures
*/
goal = ext4_find_near(inode, partial);
goal = goal & EXT4_MAX_BLOCK_FILE_PHYS;
return goal;
}
/**
* ext4_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
* @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
*
* return the total number of blocks to be allocate, including the
* direct and indirect blocks.
*/
static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned int blks,
int blocks_to_boundary)
{
unsigned int count = 0;
/*
* Simple case, [t,d]Indirect block(s) has not allocated yet
* then it's clear blocks on that path have not allocated
*/
if (k > 0) {
/* right now we don't handle cross boundary allocation */
if (blks < blocks_to_boundary + 1)
count += blks;
else
count += blocks_to_boundary + 1;
return count;
}
count++;
while (count < blks && count <= blocks_to_boundary &&
le32_to_cpu(*(branch[0].p + count)) == 0) {
count++;
}
return count;
}
/**
* ext4_alloc_blocks: multiple allocate blocks needed for a branch
* @handle: handle for this transaction
* @inode: inode which needs allocated blocks
* @iblock: the logical block to start allocated at
* @goal: preferred physical block of allocation
* @indirect_blks: the number of blocks need to allocate for indirect
* blocks
* @blks: number of desired blocks
* @new_blocks: on return it will store the new block numbers for
* the indirect blocks(if needed) and the first direct block,
* @err: on return it will store the error code
*
* This function will return the number of blocks allocated as
* requested by the passed-in parameters.
*/
static int ext4_alloc_blocks(handle_t *handle, struct inode *inode,
ext4_lblk_t iblock, ext4_fsblk_t goal,
int indirect_blks, int blks,
ext4_fsblk_t new_blocks[4], int *err)
{
struct ext4_allocation_request ar;
int target, i;
unsigned long count = 0, blk_allocated = 0;
int index = 0;
ext4_fsblk_t current_block = 0;
int ret = 0;
/*
* Here we try to allocate the requested multiple blocks at once,
* on a best-effort basis.
* To build a branch, we should allocate blocks for
* the indirect blocks(if not allocated yet), and at least
* the first direct block of this branch. That's the
* minimum number of blocks need to allocate(required)
*/
/* first we try to allocate the indirect blocks */
target = indirect_blks;
while (target > 0) {
count = target;
/* allocating blocks for indirect blocks and direct blocks */
current_block = ext4_new_meta_blocks(handle, inode, goal,
0, &count, err);
if (*err)
goto failed_out;
if (unlikely(current_block + count > EXT4_MAX_BLOCK_FILE_PHYS)) {
EXT4_ERROR_INODE(inode,
"current_block %llu + count %lu > %d!",
current_block, count,
EXT4_MAX_BLOCK_FILE_PHYS);
*err = -EIO;
goto failed_out;
}
target -= count;
/* allocate blocks for indirect blocks */
while (index < indirect_blks && count) {
new_blocks[index++] = current_block++;
count--;
}
if (count > 0) {
/*
* save the new block number
* for the first direct block
*/
new_blocks[index] = current_block;
WARN(1, KERN_INFO "%s returned more blocks than "
"requested\n", __func__);
break;
}
}
target = blks - count ;
blk_allocated = count;
if (!target)
goto allocated;
/* Now allocate data blocks */
memset(&ar, 0, sizeof(ar));
ar.inode = inode;
ar.goal = goal;
ar.len = target;
ar.logical = iblock;
if (S_ISREG(inode->i_mode))
/* enable in-core preallocation only for regular files */
ar.flags = EXT4_MB_HINT_DATA;
current_block = ext4_mb_new_blocks(handle, &ar, err);
if (unlikely(current_block + ar.len > EXT4_MAX_BLOCK_FILE_PHYS)) {
EXT4_ERROR_INODE(inode,
"current_block %llu + ar.len %d > %d!",
current_block, ar.len,
EXT4_MAX_BLOCK_FILE_PHYS);
*err = -EIO;
goto failed_out;
}
if (*err && (target == blks)) {
/*
* if the allocation failed and we didn't allocate
* any blocks before
*/
goto failed_out;
}
if (!*err) {
if (target == blks) {
/*
* save the new block number
* for the first direct block
*/
new_blocks[index] = current_block;
}
blk_allocated += ar.len;
}
allocated:
/* total number of blocks allocated for direct blocks */
ret = blk_allocated;
*err = 0;
return ret;
failed_out:
for (i = 0; i < index; i++)
ext4_free_blocks(handle, inode, NULL, new_blocks[i], 1, 0);
return ret;
}
/**
* ext4_alloc_branch - allocate and set up a chain of blocks.
* @handle: handle for this transaction
* @inode: owner
* @indirect_blks: number of allocated indirect blocks
* @blks: number of allocated direct blocks
* @goal: preferred place for allocation
* @offsets: offsets (in the blocks) to store the pointers to next.
* @branch: place to store the chain in.
*
* This function allocates blocks, zeroes out all but the last one,
* links them into chain and (if we are synchronous) writes them to disk.
* In other words, it prepares a branch that can be spliced onto the
* inode. It stores the information about that chain in the branch[], in
* the same format as ext4_get_branch() would do. We are calling it after
* we had read the existing part of chain and partial points to the last
* triple of that (one with zero ->key). Upon the exit we have the same
* picture as after the successful ext4_get_block(), except that in one
* place chain is disconnected - *branch->p is still zero (we did not
* set the last link), but branch->key contains the number that should
* be placed into *branch->p to fill that gap.
*
* If allocation fails we free all blocks we've allocated (and forget
* their buffer_heads) and return the error value the from failed
* ext4_alloc_block() (normally -ENOSPC). Otherwise we set the chain
* as described above and return 0.
*/
static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
ext4_lblk_t iblock, int indirect_blks,
int *blks, ext4_fsblk_t goal,
ext4_lblk_t *offsets, Indirect *branch)
{
int blocksize = inode->i_sb->s_blocksize;
int i, n = 0;
int err = 0;
struct buffer_head *bh;
int num;
ext4_fsblk_t new_blocks[4];
ext4_fsblk_t current_block;
num = ext4_alloc_blocks(handle, inode, iblock, goal, indirect_blks,
*blks, new_blocks, &err);
if (err)
return err;
branch[0].key = cpu_to_le32(new_blocks[0]);
/*
* metadata blocks and data blocks are allocated.
*/
for (n = 1; n <= indirect_blks; n++) {
/*
* Get buffer_head for parent block, zero it out
* and set the pointer to new one, then send
* parent to disk.
*/
bh = sb_getblk(inode->i_sb, new_blocks[n-1]);
if (unlikely(!bh)) {
err = -ENOMEM;
goto failed;
}
branch[n].bh = bh;
lock_buffer(bh);
BUFFER_TRACE(bh, "call get_create_access");
err = ext4_journal_get_create_access(handle, bh);
if (err) {
/* Don't brelse(bh) here; it's done in
* ext4_journal_forget() below */
unlock_buffer(bh);
goto failed;
}
memset(bh->b_data, 0, blocksize);
branch[n].p = (__le32 *) bh->b_data + offsets[n];
branch[n].key = cpu_to_le32(new_blocks[n]);
*branch[n].p = branch[n].key;
if (n == indirect_blks) {
current_block = new_blocks[n];
/*
* End of chain, update the last new metablock of
* the chain to point to the new allocated
* data blocks numbers
*/
for (i = 1; i < num; i++)
*(branch[n].p + i) = cpu_to_le32(++current_block);
}
BUFFER_TRACE(bh, "marking uptodate");
set_buffer_uptodate(bh);
unlock_buffer(bh);
BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle, inode, bh);
if (err)
goto failed;
}
*blks = num;
return err;
failed:
/* Allocation failed, free what we already allocated */
ext4_free_blocks(handle, inode, NULL, new_blocks[0], 1, 0);
for (i = 1; i <= n ; i++) {
/*
* branch[i].bh is newly allocated, so there is no
* need to revoke the block, which is why we don't
* need to set EXT4_FREE_BLOCKS_METADATA.
*/
ext4_free_blocks(handle, inode, NULL, new_blocks[i], 1,
EXT4_FREE_BLOCKS_FORGET);
}
for (i = n+1; i < indirect_blks; i++)
ext4_free_blocks(handle, inode, NULL, new_blocks[i], 1, 0);
ext4_free_blocks(handle, inode, NULL, new_blocks[i], num, 0);
return err;
}
/**
* ext4_splice_branch - splice the allocated branch onto inode.
* @handle: handle for this transaction
* @inode: owner
* @block: (logical) number of block we are adding
* @chain: chain of indirect blocks (with a missing link - see
* ext4_alloc_branch)
* @where: location of missing link
* @num: number of indirect blocks we are adding
* @blks: number of direct blocks we are adding
*
* This function fills the missing link and does all housekeeping needed in
* inode (->i_blocks, etc.). In case of success we end up with the full
* chain to new block and return 0.
*/
static int ext4_splice_branch(handle_t *handle, struct inode *inode,
ext4_lblk_t block, Indirect *where, int num,
int blks)
{
int i;
int err = 0;
ext4_fsblk_t current_block;
/*
* If we're splicing into a [td]indirect block (as opposed to the
* inode) then we need to get write access to the [td]indirect block
* before the splice.
*/
if (where->bh) {
BUFFER_TRACE(where->bh, "get_write_access");
err = ext4_journal_get_write_access(handle, where->bh);
if (err)
goto err_out;
}
/* That's it */
*where->p = where->key;
/*
* Update the host buffer_head or inode to point to more just allocated
* direct blocks blocks
*/
if (num == 0 && blks > 1) {
current_block = le32_to_cpu(where->key) + 1;
for (i = 1; i < blks; i++)
*(where->p + i) = cpu_to_le32(current_block++);
}
/* We are done with atomic stuff, now do the rest of housekeeping */
/* had we spliced it onto indirect block? */
if (where->bh) {
/*
* If we spliced it onto an indirect block, we haven't
* altered the inode. Note however that if it is being spliced
* onto an indirect block at the very end of the file (the
* file is growing) then we *will* alter the inode to reflect
* the new i_size. But that is not done here - it is done in
* generic_commit_write->__mark_inode_dirty->ext4_dirty_inode.
*/
jbd_debug(5, "splicing indirect only\n");
BUFFER_TRACE(where->bh, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle, inode, where->bh);
if (err)
goto err_out;
} else {
/*
* OK, we spliced it into the inode itself on a direct block.
*/
ext4_mark_inode_dirty(handle, inode);
jbd_debug(5, "splicing direct\n");
}
return err;
err_out:
for (i = 1; i <= num; i++) {
/*
* branch[i].bh is newly allocated, so there is no
* need to revoke the block, which is why we don't
* need to set EXT4_FREE_BLOCKS_METADATA.
*/
ext4_free_blocks(handle, inode, where[i].bh, 0, 1,
EXT4_FREE_BLOCKS_FORGET);
}
ext4_free_blocks(handle, inode, NULL, le32_to_cpu(where[num].key),
blks, 0);
return err;
}
/*
* The ext4_ind_map_blocks() function handles non-extents inodes
* (i.e., using the traditional indirect/double-indirect i_blocks
* scheme) for ext4_map_blocks().
*
* Allocation strategy is simple: if we have to allocate something, we will
* have to go the whole way to leaf. So let's do it before attaching anything
* to tree, set linkage between the newborn blocks, write them if sync is
* required, recheck the path, free and repeat if check fails, otherwise
* set the last missing link (that will protect us from any truncate-generated
* removals - all blocks on the path are immune now) and possibly force the
* write on the parent block.
* That has a nice additional property: no special recovery from the failed
* allocations is needed - we simply release blocks and do not touch anything
* reachable from inode.
*
* `handle' can be NULL if create == 0.
*
* return > 0, # of blocks mapped or allocated.
* return = 0, if plain lookup failed.
* return < 0, error case.
*
* The ext4_ind_get_blocks() function should be called with
* down_write(&EXT4_I(inode)->i_data_sem) if allocating filesystem
* blocks (i.e., flags has EXT4_GET_BLOCKS_CREATE set) or
* down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system
* blocks.
*/
int ext4_ind_map_blocks(handle_t *handle, struct inode *inode,
struct ext4_map_blocks *map,
int flags)
{
int err = -EIO;
ext4_lblk_t offsets[4];
Indirect chain[4];
Indirect *partial;
ext4_fsblk_t goal;
int indirect_blks;
int blocks_to_boundary = 0;
int depth;
int count = 0;
ext4_fsblk_t first_block = 0;
trace_ext4_ind_map_blocks_enter(inode, map->m_lblk, map->m_len, flags);
J_ASSERT(!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)));
J_ASSERT(handle != NULL || (flags & EXT4_GET_BLOCKS_CREATE) == 0);
depth = ext4_block_to_path(inode, map->m_lblk, offsets,
&blocks_to_boundary);
if (depth == 0)
goto out;
partial = ext4_get_branch(inode, depth, offsets, chain, &err);
/* Simplest case - block found, no allocation needed */
if (!partial) {
first_block = le32_to_cpu(chain[depth - 1].key);
count++;
/*map more blocks*/
while (count < map->m_len && count <= blocks_to_boundary) {
ext4_fsblk_t blk;
blk = le32_to_cpu(*(chain[depth-1].p + count));
if (blk == first_block + count)
count++;
else
break;
}
goto got_it;
}
/* Next simple case - plain lookup or failed read of indirect block */
if ((flags & EXT4_GET_BLOCKS_CREATE) == 0 || err == -EIO)
goto cleanup;
/*
* Okay, we need to do block allocation.
*/
if (EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
EXT4_FEATURE_RO_COMPAT_BIGALLOC)) {
EXT4_ERROR_INODE(inode, "Can't allocate blocks for "
"non-extent mapped inodes with bigalloc");
return -ENOSPC;
}
goal = ext4_find_goal(inode, map->m_lblk, partial);
/* the number of blocks need to allocate for [d,t]indirect blocks */
indirect_blks = (chain + depth) - partial - 1;
/*
* Next look up the indirect map to count the totoal number of
* direct blocks to allocate for this branch.
*/
count = ext4_blks_to_allocate(partial, indirect_blks,
map->m_len, blocks_to_boundary);
/*
* Block out ext4_truncate while we alter the tree
*/
err = ext4_alloc_branch(handle, inode, map->m_lblk, indirect_blks,
&count, goal,
offsets + (partial - chain), partial);
/*
* The ext4_splice_branch call will free and forget any buffers
* on the new chain if there is a failure, but that risks using
* up transaction credits, especially for bitmaps where the
* credits cannot be returned. Can we handle this somehow? We
* may need to return -EAGAIN upwards in the worst case. --sct
*/
if (!err)
err = ext4_splice_branch(handle, inode, map->m_lblk,
partial, indirect_blks, count);
if (err)
goto cleanup;
map->m_flags |= EXT4_MAP_NEW;
ext4_update_inode_fsync_trans(handle, inode, 1);
got_it:
map->m_flags |= EXT4_MAP_MAPPED;
map->m_pblk = le32_to_cpu(chain[depth-1].key);
map->m_len = count;
if (count > blocks_to_boundary)
map->m_flags |= EXT4_MAP_BOUNDARY;
err = count;
/* Clean up and exit */
partial = chain + depth - 1; /* the whole chain */
cleanup:
while (partial > chain) {
BUFFER_TRACE(partial->bh, "call brelse");
brelse(partial->bh);
partial--;
}
out:
trace_ext4_ind_map_blocks_exit(inode, map, err);
return err;
}
/*
* O_DIRECT for ext3 (or indirect map) based files
*
* If the O_DIRECT write will extend the file then add this inode to the
* orphan list. So recovery will truncate it back to the original size
* if the machine crashes during the write.
*
* If the O_DIRECT write is intantiating holes inside i_size and the machine
* crashes then stale disk data _may_ be exposed inside the file. But current
* VFS code falls back into buffered path in that case so we are safe.
*/
ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb,
const struct iovec *iov, loff_t offset,
unsigned long nr_segs)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
struct ext4_inode_info *ei = EXT4_I(inode);
handle_t *handle;
ssize_t ret;
int orphan = 0;
size_t count = iov_length(iov, nr_segs);
int retries = 0;
if (rw == WRITE) {
loff_t final_size = offset + count;
if (final_size > inode->i_size) {
/* Credits for sb + inode write */
handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
goto out;
}
ret = ext4_orphan_add(handle, inode);
if (ret) {
ext4_journal_stop(handle);
goto out;
}
orphan = 1;
ei->i_disksize = inode->i_size;
ext4_journal_stop(handle);
}
}
retry:
if (rw == READ && ext4_should_dioread_nolock(inode)) {
if (unlikely(atomic_read(&EXT4_I(inode)->i_unwritten))) {
mutex_lock(&inode->i_mutex);
ext4_flush_unwritten_io(inode);
mutex_unlock(&inode->i_mutex);
}
/*
* Nolock dioread optimization may be dynamically disabled
* via ext4_inode_block_unlocked_dio(). Check inode's state
* while holding extra i_dio_count ref.
*/
atomic_inc(&inode->i_dio_count);
smp_mb();
if (unlikely(ext4_test_inode_state(inode,
EXT4_STATE_DIOREAD_LOCK))) {
inode_dio_done(inode);
goto locked;
}
ret = __blockdev_direct_IO(rw, iocb, inode,
inode->i_sb->s_bdev, iov,
offset, nr_segs,
ext4_get_block, NULL, NULL, 0);
inode_dio_done(inode);
} else {
locked:
ret = blockdev_direct_IO(rw, iocb, inode, iov,
offset, nr_segs, ext4_get_block);
if (unlikely((rw & WRITE) && ret < 0)) {
loff_t isize = i_size_read(inode);
loff_t end = offset + iov_length(iov, nr_segs);
if (end > isize)
ext4_truncate_failed_write(inode);
}
}
if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
goto retry;
if (orphan) {
int err;
/* Credits for sb + inode write */
handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
if (IS_ERR(handle)) {
/* This is really bad luck. We've written the data
* but cannot extend i_size. Bail out and pretend
* the write failed... */
ret = PTR_ERR(handle);
if (inode->i_nlink)
ext4_orphan_del(NULL, inode);
goto out;
}
if (inode->i_nlink)
ext4_orphan_del(handle, inode);
if (ret > 0) {
loff_t end = offset + ret;
if (end > inode->i_size) {
ei->i_disksize = end;
i_size_write(inode, end);
/*
* We're going to return a positive `ret'
* here due to non-zero-length I/O, so there's
* no way of reporting error returns from
* ext4_mark_inode_dirty() to userspace. So
* ignore it.
*/
ext4_mark_inode_dirty(handle, inode);
}
}
err = ext4_journal_stop(handle);
if (ret == 0)
ret = err;
}
out:
return ret;
}
/*
* Calculate the number of metadata blocks need to reserve
* to allocate a new block at @lblocks for non extent file based file
*/
int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock)
{
struct ext4_inode_info *ei = EXT4_I(inode);
sector_t dind_mask = ~((sector_t)EXT4_ADDR_PER_BLOCK(inode->i_sb) - 1);
int blk_bits;
if (lblock < EXT4_NDIR_BLOCKS)
return 0;
lblock -= EXT4_NDIR_BLOCKS;
if (ei->i_da_metadata_calc_len &&
(lblock & dind_mask) == ei->i_da_metadata_calc_last_lblock) {
ei->i_da_metadata_calc_len++;
return 0;
}
ei->i_da_metadata_calc_last_lblock = lblock & dind_mask;
ei->i_da_metadata_calc_len = 1;
blk_bits = order_base_2(lblock);
return (blk_bits / EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb)) + 1;
}
int ext4_ind_trans_blocks(struct inode *inode, int nrblocks, int chunk)
{
int indirects;
/* if nrblocks are contiguous */
if (chunk) {
/*
* With N contiguous data blocks, we need at most
* N/EXT4_ADDR_PER_BLOCK(inode->i_sb) + 1 indirect blocks,
* 2 dindirect blocks, and 1 tindirect block
*/
return DIV_ROUND_UP(nrblocks,
EXT4_ADDR_PER_BLOCK(inode->i_sb)) + 4;
}
/*
* if nrblocks are not contiguous, worse case, each block touch
* a indirect block, and each indirect block touch a double indirect
* block, plus a triple indirect block
*/
indirects = nrblocks * 2 + 1;
return indirects;
}
/*
* 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
*/
static handle_t *start_transaction(struct inode *inode)
{
handle_t *result;
result = ext4_journal_start(inode, EXT4_HT_TRUNCATE,
ext4_blocks_for_truncate(inode));
if (!IS_ERR(result))
return result;
ext4_std_error(inode->i_sb, PTR_ERR(result));
return result;
}
/*
* Try to extend this transaction for the purposes of truncation.
*
* Returns 0 if we managed to create more room. If we can't create more
* room, and the transaction must be restarted we return 1.
*/
static int try_to_extend_transaction(handle_t *handle, struct inode *inode)
{
if (!ext4_handle_valid(handle))
return 0;
if (ext4_handle_has_enough_credits(handle, EXT4_RESERVE_TRANS_BLOCKS+1))
return 0;
if (!ext4_journal_extend(handle, ext4_blocks_for_truncate(inode)))
return 0;
return 1;
}
/*
* Probably it should be a library function... search for first non-zero word
* or memcmp with zero_page, whatever is better for particular architecture.
* Linus?
*/
static inline int all_zeroes(__le32 *p, __le32 *q)
{
while (p < q)
if (*p++)
return 0;
return 1;
}
/**
* ext4_find_shared - find the indirect blocks for partial truncation.
* @inode: inode in question
* @depth: depth of the affected branch
* @offsets: offsets of pointers in that branch (see ext4_block_to_path)
* @chain: place to store the pointers to partial indirect blocks
* @top: place to the (detached) top of branch
*
* This is a helper function used by ext4_truncate().
*
* When we do truncate() we may have to clean the ends of several
* indirect blocks but leave the blocks themselves alive. Block is
* partially truncated if some data below the new i_size is referred
* from it (and it is on the path to the first completely truncated
* data block, indeed). We have to free the top of that path along
* with everything to the right of the path. Since no allocation
* past the truncation point is possible until ext4_truncate()
* finishes, we may safely do the latter, but top of branch may
* require special attention - pageout below the truncation point
* might try to populate it.
*
* We atomically detach the top of branch from the tree, store the
* block number of its root in *@top, pointers to buffer_heads of
* partially truncated blocks - in @chain[].bh and pointers to
* their last elements that should not be removed - in
* @chain[].p. Return value is the pointer to last filled element
* of @chain.
*
* The work left to caller to do the actual freeing of subtrees:
* a) free the subtree starting from *@top
* b) free the subtrees whose roots are stored in
* (@chain[i].p+1 .. end of @chain[i].bh->b_data)
* c) free the subtrees growing from the inode past the @chain[0].
* (no partially truncated stuff there). */
static Indirect *ext4_find_shared(struct inode *inode, int depth,
ext4_lblk_t offsets[4], Indirect chain[4],
__le32 *top)
{
Indirect *partial, *p;
int k, err;
*top = 0;
/* Make k index the deepest non-null offset + 1 */
for (k = depth; k > 1 && !offsets[k-1]; k--)
;
partial = ext4_get_branch(inode, k, offsets, chain, &err);
/* Writer: pointers */
if (!partial)
partial = chain + k-1;
/*
* If the branch acquired continuation since we've looked at it -
* fine, it should all survive and (new) top doesn't belong to us.
*/
if (!partial->key && *partial->p)
/* Writer: end */
goto no_top;
for (p = partial; (p > chain) && all_zeroes((__le32 *) p->bh->b_data, p->p); p--)
;
/*
* OK, we've found the last block that must survive. The rest of our
* branch should be detached before unlocking. However, if that rest
* of branch is all ours and does not grow immediately from the inode
* it's easier to cheat and just decrement partial->p.
*/
if (p == chain + k - 1 && p > chain) {
p->p--;
} else {
*top = *p->p;
/* Nope, don't do this in ext4. Must leave the tree intact */
#if 0
*p->p = 0;
#endif
}
/* Writer: end */
while (partial > p) {
brelse(partial->bh);
partial--;
}
no_top:
return partial;
}
/*
* Zero a number of block pointers in either an inode or an indirect block.
* If we restart the transaction we must again get write access to the
* indirect block for further modification.
*
* We release `count' blocks on disk, but (last - first) may be greater
* than `count' because there can be holes in there.
*
* Return 0 on success, 1 on invalid block range
* and < 0 on fatal error.
*/
static int ext4_clear_blocks(handle_t *handle, struct inode *inode,
struct buffer_head *bh,
ext4_fsblk_t block_to_free,
unsigned long count, __le32 *first,
__le32 *last)
{
__le32 *p;
int flags = EXT4_FREE_BLOCKS_FORGET | EXT4_FREE_BLOCKS_VALIDATED;
int err;
if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
flags |= EXT4_FREE_BLOCKS_METADATA;
if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), block_to_free,
count)) {
EXT4_ERROR_INODE(inode, "attempt to clear invalid "
"blocks %llu len %lu",
(unsigned long long) block_to_free, count);
return 1;
}
if (try_to_extend_transaction(handle, inode)) {
if (bh) {
BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle, inode, bh);
if (unlikely(err))
goto out_err;
}
err = ext4_mark_inode_dirty(handle, inode);
if (unlikely(err))
goto out_err;
err = ext4_truncate_restart_trans(handle, inode,
ext4_blocks_for_truncate(inode));
if (unlikely(err))
goto out_err;
if (bh) {
BUFFER_TRACE(bh, "retaking write access");
err = ext4_journal_get_write_access(handle, bh);
if (unlikely(err))
goto out_err;
}
}
for (p = first; p < last; p++)
*p = 0;
ext4_free_blocks(handle, inode, NULL, block_to_free, count, flags);
return 0;
out_err:
ext4_std_error(inode->i_sb, err);
return err;
}
/**
* ext4_free_data - free a list of data blocks
* @handle: handle for this transaction
* @inode: inode we are dealing with
* @this_bh: indirect buffer_head which contains *@first and *@last
* @first: array of block numbers
* @last: points immediately past the end of array
*
* We are freeing all blocks referred from that array (numbers are stored as
* little-endian 32-bit) and updating @inode->i_blocks appropriately.
*
* We accumulate contiguous runs of blocks to free. Conveniently, if these
* blocks are contiguous then releasing them at one time will only affect one
* or two bitmap blocks (+ group descriptor(s) and superblock) and we won't
* actually use a lot of journal space.
*
* @this_bh will be %NULL if @first and @last point into the inode's direct
* block pointers.
*/
static void ext4_free_data(handle_t *handle, struct inode *inode,
struct buffer_head *this_bh,
__le32 *first, __le32 *last)
{
ext4_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 */
ext4_fsblk_t nr; /* Current block # */
__le32 *p; /* Pointer into inode/ind
for current block */
int err = 0;
if (this_bh) { /* For indirect block */
BUFFER_TRACE(this_bh, "get_write_access");
err = ext4_journal_get_write_access(handle, this_bh);
/* Important: if we can't update the indirect pointers
* to the blocks, we can't free them. */
if (err)
return;
}
for (p = first; p < last; p++) {
nr = le32_to_cpu(*p);
if (nr) {
/* accumulate blocks to free if they're contiguous */
if (count == 0) {
block_to_free = nr;
block_to_free_p = p;
count = 1;
} else if (nr == block_to_free + count) {
count++;
} else {
err = ext4_clear_blocks(handle, inode, this_bh,
block_to_free, count,
block_to_free_p, p);
if (err)
break;
block_to_free = nr;
block_to_free_p = p;
count = 1;
}
}
}
if (!err && count > 0)
err = ext4_clear_blocks(handle, inode, this_bh, block_to_free,
count, block_to_free_p, p);
if (err < 0)
/* fatal error */
return;
if (this_bh) {
BUFFER_TRACE(this_bh, "call ext4_handle_dirty_metadata");
/*
* The buffer head should have an attached journal head at this
* point. However, if the data is corrupted and an indirect
* block pointed to itself, it would have been detached when
* the block was cleared. Check for this instead of OOPSing.
*/
if ((EXT4_JOURNAL(inode) == NULL) || bh2jh(this_bh))
ext4_handle_dirty_metadata(handle, inode, this_bh);
else
EXT4_ERROR_INODE(inode,
"circular indirect block detected at "
"block %llu",
(unsigned long long) this_bh->b_blocknr);
}
}
/**
* ext4_free_branches - free an array of branches
* @handle: JBD handle for this transaction
* @inode: inode we are dealing with
* @parent_bh: the buffer_head which contains *@first and *@last
* @first: array of block numbers
* @last: pointer immediately past the end of array
* @depth: depth of the branches to free
*
* We are freeing all blocks referred from these branches (numbers are
* stored as little-endian 32-bit) and updating @inode->i_blocks
* appropriately.
*/
static void ext4_free_branches(handle_t *handle, struct inode *inode,
struct buffer_head *parent_bh,
__le32 *first, __le32 *last, int depth)
{
ext4_fsblk_t nr;
__le32 *p;
if (ext4_handle_is_aborted(handle))
return;
if (depth--) {
struct buffer_head *bh;
int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb);
p = last;
while (--p >= first) {
nr = le32_to_cpu(*p);
if (!nr)
continue; /* A hole */
if (!ext4_data_block_valid(EXT4_SB(inode->i_sb),
nr, 1)) {
EXT4_ERROR_INODE(inode,
"invalid indirect mapped "
"block %lu (level %d)",
(unsigned long) nr, depth);
break;
}
/* Go read the buffer for the next level down */
bh = sb_bread(inode->i_sb, nr);
/*
* A read failure? Report error and clear slot
* (should be rare).
*/
if (!bh) {
EXT4_ERROR_INODE_BLOCK(inode, nr,
"Read failure");
continue;
}
/* This zaps the entire block. Bottom up. */
BUFFER_TRACE(bh, "free child branches");
ext4_free_branches(handle, inode, bh,
(__le32 *) bh->b_data,
(__le32 *) bh->b_data + addr_per_block,
depth);
brelse(bh);
/*
* Everything below this this pointer has been
* released. Now let this top-of-subtree go.
*
* We want the freeing of this indirect block to be
* atomic in the journal with the updating of the
* bitmap block which owns it. So make some room in
* the journal.
*
* We zero the parent pointer *after* freeing its
* pointee in the bitmaps, so if extend_transaction()
* for some reason fails to put the bitmap changes and
* the release into the same transaction, recovery
* will merely complain about releasing a free block,
* rather than leaking blocks.
*/
if (ext4_handle_is_aborted(handle))
return;
if (try_to_extend_transaction(handle, inode)) {
ext4_mark_inode_dirty(handle, inode);
ext4_truncate_restart_trans(handle, inode,
ext4_blocks_for_truncate(inode));
}
/*
* The forget flag here is critical because if
* we are journaling (and not doing data
* journaling), we have to make sure a revoke
* record is written to prevent the journal
* replay from overwriting the (former)
* indirect block if it gets reallocated as a
* data block. This must happen in the same
* transaction where the data blocks are
* actually freed.
*/
ext4_free_blocks(handle, inode, NULL, nr, 1,
EXT4_FREE_BLOCKS_METADATA|
EXT4_FREE_BLOCKS_FORGET);
if (parent_bh) {
/*
* The block which we have just freed is
* pointed to by an indirect block: journal it
*/
BUFFER_TRACE(parent_bh, "get_write_access");
if (!ext4_journal_get_write_access(handle,
parent_bh)){
*p = 0;
BUFFER_TRACE(parent_bh,
"call ext4_handle_dirty_metadata");
ext4_handle_dirty_metadata(handle,
inode,
parent_bh);
}
}
}
} else {
/* We have reached the bottom of the tree. */
BUFFER_TRACE(parent_bh, "free data blocks");
ext4_free_data(handle, inode, parent_bh, first, last);
}
}
void ext4_ind_truncate(struct inode *inode)
{
handle_t *handle;
struct ext4_inode_info *ei = EXT4_I(inode);
__le32 *i_data = ei->i_data;
int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb);
struct address_space *mapping = inode->i_mapping;
ext4_lblk_t offsets[4];
Indirect chain[4];
Indirect *partial;
__le32 nr = 0;
int n = 0;
ext4_lblk_t last_block, max_block;
loff_t page_len;
unsigned blocksize = inode->i_sb->s_blocksize;
int err;
handle = start_transaction(inode);
if (IS_ERR(handle))
return; /* AKPM: return what? */
last_block = (inode->i_size + blocksize-1)
>> EXT4_BLOCK_SIZE_BITS(inode->i_sb);
max_block = (EXT4_SB(inode->i_sb)->s_bitmap_maxbytes + blocksize-1)
>> EXT4_BLOCK_SIZE_BITS(inode->i_sb);
if (inode->i_size % PAGE_CACHE_SIZE != 0) {
page_len = PAGE_CACHE_SIZE -
(inode->i_size & (PAGE_CACHE_SIZE - 1));
err = ext4_discard_partial_page_buffers(handle,
mapping, inode->i_size, page_len, 0);
if (err)
goto out_stop;
}
if (last_block != max_block) {
n = ext4_block_to_path(inode, last_block, offsets, NULL);
if (n == 0)
goto out_stop; /* error */
}
/*
* OK. This truncate is going to happen. We add the inode to the
* orphan list, so that if this truncate spans multiple transactions,
* and we crash, we will resume the truncate when the filesystem
* recovers. It also marks the inode dirty, to catch the new size.
*
* Implication: the file must always be in a sane, consistent
* truncatable state while each transaction commits.
*/
if (ext4_orphan_add(handle, inode))
goto out_stop;
/*
* From here we block out all ext4_get_block() callers who want to
* modify the block allocation tree.
*/
down_write(&ei->i_data_sem);
ext4_discard_preallocations(inode);
ext4_es_remove_extent(inode, last_block, EXT_MAX_BLOCKS - last_block);
/*
* The orphan list entry will now protect us from any crash which
* occurs before the truncate completes, so it is now safe to propagate
* the new, shorter inode size (held for now in i_size) into the
* on-disk inode. We do this via i_disksize, which is the value which
* ext4 *really* writes onto the disk inode.
*/
ei->i_disksize = inode->i_size;
if (last_block == max_block) {
/*
* It is unnecessary to free any data blocks if last_block is
* equal to the indirect block limit.
*/
goto out_unlock;
} else if (n == 1) { /* direct blocks */
ext4_free_data(handle, inode, NULL, i_data+offsets[0],
i_data + EXT4_NDIR_BLOCKS);
goto do_indirects;
}
partial = ext4_find_shared(inode, n, offsets, chain, &nr);
/* Kill the top of shared branch (not detached) */
if (nr) {
if (partial == chain) {
/* Shared branch grows from the inode */
ext4_free_branches(handle, inode, NULL,
&nr, &nr+1, (chain+n-1) - partial);
*partial->p = 0;
/*
* We mark the inode dirty prior to restart,
* and prior to stop. No need for it here.
*/
} else {
/* Shared branch grows from an indirect block */
BUFFER_TRACE(partial->bh, "get_write_access");
ext4_free_branches(handle, inode, partial->bh,
partial->p,
partial->p+1, (chain+n-1) - partial);
}
}
/* Clear the ends of indirect blocks on the shared branch */
while (partial > chain) {
ext4_free_branches(handle, inode, partial->bh, partial->p + 1,
(__le32*)partial->bh->b_data+addr_per_block,
(chain+n-1) - partial);
BUFFER_TRACE(partial->bh, "call brelse");
brelse(partial->bh);
partial--;
}
do_indirects:
/* Kill the remaining (whole) subtrees */
switch (offsets[0]) {
default:
nr = i_data[EXT4_IND_BLOCK];
if (nr) {
ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 1);
i_data[EXT4_IND_BLOCK] = 0;
}
case EXT4_IND_BLOCK:
nr = i_data[EXT4_DIND_BLOCK];
if (nr) {
ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 2);
i_data[EXT4_DIND_BLOCK] = 0;
}
case EXT4_DIND_BLOCK:
nr = i_data[EXT4_TIND_BLOCK];
if (nr) {
ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 3);
i_data[EXT4_TIND_BLOCK] = 0;
}
case EXT4_TIND_BLOCK:
;
}
out_unlock:
up_write(&ei->i_data_sem);
inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
ext4_mark_inode_dirty(handle, inode);
/*
* In a multi-transaction truncate, we only make the final transaction
* synchronous
*/
if (IS_SYNC(inode))
ext4_handle_sync(handle);
out_stop:
/*
* If this was a simple ftruncate(), and the file will remain alive
* then we need to clear up the orphan record which we created above.
* However, if this was a real unlink then we were called by
* ext4_delete_inode(), and we allow that function to clean up the
* orphan info for us.
*/
if (inode->i_nlink)
ext4_orphan_del(handle, inode);
ext4_journal_stop(handle);
trace_ext4_truncate_exit(inode);
}
static int free_hole_blocks(handle_t *handle, struct inode *inode,
struct buffer_head *parent_bh, __le32 *i_data,
int level, ext4_lblk_t first,
ext4_lblk_t count, int max)
{
struct buffer_head *bh = NULL;
int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb);
int ret = 0;
int i, inc;
ext4_lblk_t offset;
__le32 blk;
inc = 1 << ((EXT4_BLOCK_SIZE_BITS(inode->i_sb) - 2) * level);
for (i = 0, offset = 0; i < max; i++, i_data++, offset += inc) {
if (offset >= count + first)
break;
if (*i_data == 0 || (offset + inc) <= first)
continue;
blk = *i_data;
if (level > 0) {
ext4_lblk_t first2;
bh = sb_bread(inode->i_sb, le32_to_cpu(blk));
if (!bh) {
EXT4_ERROR_INODE_BLOCK(inode, le32_to_cpu(blk),
"Read failure");
return -EIO;
}
first2 = (first > offset) ? first - offset : 0;
ret = free_hole_blocks(handle, inode, bh,
(__le32 *)bh->b_data, level - 1,
first2, count - offset,
inode->i_sb->s_blocksize >> 2);
if (ret) {
brelse(bh);
goto err;
}
}
if (level == 0 ||
(bh && all_zeroes((__le32 *)bh->b_data,
(__le32 *)bh->b_data + addr_per_block))) {
ext4_free_data(handle, inode, parent_bh, &blk, &blk+1);
*i_data = 0;
}
brelse(bh);
bh = NULL;
}
err:
return ret;
}
static int ext4_free_hole_blocks(handle_t *handle, struct inode *inode,
ext4_lblk_t first, ext4_lblk_t stop)
{
int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb);
int level, ret = 0;
int num = EXT4_NDIR_BLOCKS;
ext4_lblk_t count, max = EXT4_NDIR_BLOCKS;
__le32 *i_data = EXT4_I(inode)->i_data;
count = stop - first;
for (level = 0; level < 4; level++, max *= addr_per_block) {
if (first < max) {
ret = free_hole_blocks(handle, inode, NULL, i_data,
level, first, count, num);
if (ret)
goto err;
if (count > max - first)
count -= max - first;
else
break;
first = 0;
} else {
first -= max;
}
i_data += num;
if (level == 0) {
num = 1;
max = 1;
}
}
err:
return ret;
}
int ext4_ind_punch_hole(struct file *file, loff_t offset, loff_t length)
{
struct inode *inode = file_inode(file);
struct super_block *sb = inode->i_sb;
ext4_lblk_t first_block, stop_block;
struct address_space *mapping = inode->i_mapping;
handle_t *handle = NULL;
loff_t first_page, last_page, page_len;
loff_t first_page_offset, last_page_offset;
int err = 0;
/*
* Write out all dirty pages to avoid race conditions
* Then release them.
*/
if (mapping->nrpages && mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
err = filemap_write_and_wait_range(mapping,
offset, offset + length - 1);
if (err)
return err;
}
mutex_lock(&inode->i_mutex);
/* It's not possible punch hole on append only file */
if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) {
err = -EPERM;
goto out_mutex;
}
if (IS_SWAPFILE(inode)) {
err = -ETXTBSY;
goto out_mutex;
}
/* No need to punch hole beyond i_size */
if (offset >= inode->i_size)
goto out_mutex;
/*
* If the hole extents beyond i_size, set the hole
* to end after the page that contains i_size
*/
if (offset + length > inode->i_size) {
length = inode->i_size +
PAGE_CACHE_SIZE - (inode->i_size & (PAGE_CACHE_SIZE - 1)) -
offset;
}
first_page = (offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
last_page = (offset + length) >> PAGE_CACHE_SHIFT;
first_page_offset = first_page << PAGE_CACHE_SHIFT;
last_page_offset = last_page << PAGE_CACHE_SHIFT;
/* Now release the pages */
if (last_page_offset > first_page_offset) {
truncate_pagecache_range(inode, first_page_offset,
last_page_offset - 1);
}
/* Wait all existing dio works, newcomers will block on i_mutex */
inode_dio_wait(inode);
handle = start_transaction(inode);
if (IS_ERR(handle))
goto out_mutex;
/*
* Now we need to zero out the non-page-aligned data in the
* pages at the start and tail of the hole, and unmap the buffer
* heads for the block aligned regions of the page that were
* completely zerod.
*/
if (first_page > last_page) {
/*
* If the file space being truncated is contained within a page
* just zero out and unmap the middle of that page
*/
err = ext4_discard_partial_page_buffers(handle,
mapping, offset, length, 0);
if (err)
goto out;
} else {
/*
* Zero out and unmap the paritial page that contains
* the start of the hole
*/
page_len = first_page_offset - offset;
if (page_len > 0) {
err = ext4_discard_partial_page_buffers(handle, mapping,
offset, page_len, 0);
if (err)
goto out;
}
/*
* Zero out and unmap the partial page that contains
* the end of the hole
*/
page_len = offset + length - last_page_offset;
if (page_len > 0) {
err = ext4_discard_partial_page_buffers(handle, mapping,
last_page_offset, page_len, 0);
if (err)
goto out;
}
}
/*
* If i_size contained in the last page, we need to
* unmap and zero the paritial page after i_size
*/
if (inode->i_size >> PAGE_CACHE_SHIFT == last_page &&
inode->i_size % PAGE_CACHE_SIZE != 0) {
page_len = PAGE_CACHE_SIZE -
(inode->i_size & (PAGE_CACHE_SIZE - 1));
if (page_len > 0) {
err = ext4_discard_partial_page_buffers(handle,
mapping, inode->i_size, page_len, 0);
if (err)
goto out;
}
}
first_block = (offset + sb->s_blocksize - 1) >>
EXT4_BLOCK_SIZE_BITS(sb);
stop_block = (offset + length) >> EXT4_BLOCK_SIZE_BITS(sb);
if (first_block >= stop_block)
goto out;
down_write(&EXT4_I(inode)->i_data_sem);
ext4_discard_preallocations(inode);
err = ext4_es_remove_extent(inode, first_block,
stop_block - first_block);
err = ext4_free_hole_blocks(handle, inode, first_block, stop_block);
ext4_discard_preallocations(inode);
if (IS_SYNC(inode))
ext4_handle_sync(handle);
up_write(&EXT4_I(inode)->i_data_sem);
out:
inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
ext4_mark_inode_dirty(handle, inode);
ext4_journal_stop(handle);
out_mutex:
mutex_unlock(&inode->i_mutex);
return err;
}