X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=fs%2Fext4%2Finode.c;h=ac97348f85b5730c8157ea19913485236c9b764a;hb=af901ca181d92aac3a7dc265144a9081a86d8f39;hp=0080999d2cd4dff2a51d5b21e6841d797bf392d3;hpb=2b2d6d019724de6e51ac5bcf22b5ef969daefa8b;p=safe%2Fjmp%2Flinux-2.6 diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 0080999..0282ec7 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -34,18 +34,27 @@ #include #include #include +#include #include #include +#include + #include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" #include "ext4_extents.h" +#include + +#define MPAGE_DA_EXTENT_TAIL 0x01 + static inline int ext4_begin_ordered_truncate(struct inode *inode, loff_t new_size) { - return jbd2_journal_begin_ordered_truncate(&EXT4_I(inode)->jinode, - new_size); + return jbd2_journal_begin_ordered_truncate( + EXT4_SB(inode->i_sb)->s_journal, + &EXT4_I(inode)->jinode, + new_size); } static void ext4_invalidatepage(struct page *page, unsigned long offset); @@ -69,9 +78,12 @@ static int ext4_inode_is_fast_symlink(struct inode *inode) * "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. + * + * If the handle isn't valid we're not journaling, but we still need to + * call into ext4_journal_revoke() to put the buffer head. */ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, - struct buffer_head *bh, ext4_fsblk_t blocknr) + struct buffer_head *bh, ext4_fsblk_t blocknr) { int err; @@ -80,7 +92,7 @@ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, BUFFER_TRACE(bh, "enter"); jbd_debug(4, "forgetting bh %p: is_metadata = %d, mode %o, " - "data mode %lx\n", + "data mode %x\n", bh, is_metadata, inode->i_mode, test_opt(inode->i_sb, DATA_FLAGS)); @@ -167,7 +179,9 @@ static handle_t *start_transaction(struct inode *inode) */ static int try_to_extend_transaction(handle_t *handle, struct inode *inode) { - if (handle->h_buffer_credits > EXT4_RESERVE_TRANS_BLOCKS) + 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, blocks_for_truncate(inode))) return 0; @@ -179,16 +193,30 @@ static int try_to_extend_transaction(handle_t *handle, struct inode *inode) * so before we call here everything must be consistently dirtied against * this transaction. */ -static int ext4_journal_test_restart(handle_t *handle, struct inode *inode) + int ext4_truncate_restart_trans(handle_t *handle, struct inode *inode, + int nblocks) { + int ret; + + /* + * Drop i_data_sem to avoid deadlock with ext4_get_blocks At this + * moment, get_block can be called only for blocks inside i_size since + * page cache has been already dropped and writes are blocked by + * i_mutex. So we can safely drop the i_data_sem here. + */ + BUG_ON(EXT4_JOURNAL(inode) == NULL); jbd_debug(2, "restarting handle %p\n", handle); - return ext4_journal_restart(handle, blocks_for_truncate(inode)); + up_write(&EXT4_I(inode)->i_data_sem); + ret = ext4_journal_restart(handle, blocks_for_truncate(inode)); + down_write(&EXT4_I(inode)->i_data_sem); + + return ret; } /* * Called at the last iput() if i_nlink is zero. */ -void ext4_delete_inode (struct inode * inode) +void ext4_delete_inode(struct inode *inode) { handle_t *handle; int err; @@ -213,7 +241,7 @@ void ext4_delete_inode (struct inode * inode) } if (IS_SYNC(inode)) - handle->h_sync = 1; + ext4_handle_sync(handle); inode->i_size = 0; err = ext4_mark_inode_dirty(handle, inode); if (err) { @@ -230,7 +258,7 @@ void ext4_delete_inode (struct inode * inode) * enough credits left in the handle to remove the inode from * the orphan list and set the dtime field. */ - if (handle->h_buffer_credits < 3) { + if (!ext4_handle_has_enough_credits(handle, 3)) { err = ext4_journal_extend(handle, 3); if (err > 0) err = ext4_journal_restart(handle, 3); @@ -316,8 +344,8 @@ static inline void add_chain(Indirect *p, struct buffer_head *bh, __le32 *v) */ static int ext4_block_to_path(struct inode *inode, - ext4_lblk_t i_block, - ext4_lblk_t offsets[4], int *boundary) + 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); @@ -327,12 +355,10 @@ static int ext4_block_to_path(struct inode *inode, int n = 0; int final = 0; - if (i_block < 0) { - ext4_warning (inode->i_sb, "ext4_block_to_path", "block < 0"); - } else if (i_block < direct_blocks) { + if (i_block < direct_blocks) { offsets[n++] = i_block; final = direct_blocks; - } else if ( (i_block -= direct_blocks) < indirect_blocks) { + } else if ((i_block -= direct_blocks) < indirect_blocks) { offsets[n++] = EXT4_IND_BLOCK; offsets[n++] = i_block; final = ptrs; @@ -349,15 +375,44 @@ static int ext4_block_to_path(struct inode *inode, final = ptrs; } else { ext4_warning(inode->i_sb, "ext4_block_to_path", - "block %lu > max", - i_block + direct_blocks + - indirect_blocks + double_blocks); + "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; } +static int __ext4_check_blockref(const char *function, struct inode *inode, + __le32 *p, unsigned int max) +{ + __le32 *bref = p; + unsigned int blk; + + while (bref < p+max) { + blk = le32_to_cpu(*bref++); + if (blk && + unlikely(!ext4_data_block_valid(EXT4_SB(inode->i_sb), + blk, 1))) { + ext4_error(inode->i_sb, function, + "invalid block reference %u " + "in inode #%lu", blk, inode->i_ino); + return -EIO; + } + } + return 0; +} + + +#define ext4_check_indirect_blockref(inode, bh) \ + __ext4_check_blockref(__func__, inode, (__le32 *)(bh)->b_data, \ + EXT4_ADDR_PER_BLOCK((inode)->i_sb)) + +#define ext4_check_inode_blockref(inode) \ + __ext4_check_blockref(__func__, inode, EXT4_I(inode)->i_data, \ + EXT4_NDIR_BLOCKS) + /** * ext4_get_branch - read the chain of indirect blocks leading to data * @inode: inode in question @@ -398,14 +453,27 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth, *err = 0; /* i_data is not going away, no lock needed */ - add_chain (chain, NULL, EXT4_I(inode)->i_data + *offsets); + add_chain(chain, NULL, EXT4_I(inode)->i_data + *offsets); if (!p->key) goto no_block; while (--depth) { - bh = sb_bread(sb, le32_to_cpu(p->key)); - if (!bh) + bh = sb_getblk(sb, le32_to_cpu(p->key)); + if (unlikely(!bh)) goto failure; - add_chain(++p, bh, (__le32*)bh->b_data + *++offsets); + + 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; @@ -441,11 +509,13 @@ no_block: 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 *start = ind->bh ? (__le32 *) ind->bh->b_data : ei->i_data; __le32 *p; ext4_fsblk_t bg_start; ext4_fsblk_t last_block; ext4_grpblk_t colour; + ext4_group_t block_group; + int flex_size = ext4_flex_bg_size(EXT4_SB(inode->i_sb)); /* Try to find previous block */ for (p = ind->p - 1; p >= start; p--) { @@ -461,9 +531,22 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) * It is going to be referred to from the inode itself? OK, just put it * into the same cylinder group then. */ - bg_start = ext4_group_first_block_no(inode->i_sb, ei->i_block_group); + block_group = ei->i_block_group; + if (flex_size >= EXT4_FLEX_SIZE_DIR_ALLOC_SCHEME) { + block_group &= ~(flex_size-1); + if (S_ISREG(inode->i_mode)) + block_group++; + } + bg_start = ext4_group_first_block_no(inode->i_sb, block_group); last_block = ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es) - 1; + /* + * If we are doing delayed allocation, we don't need take + * colour into account. + */ + if (test_opt(inode->i_sb, DELALLOC)) + return bg_start; + if (bg_start + EXT4_BLOCKS_PER_GROUP(inode->i_sb) <= last_block) colour = (current->pid % 16) * (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16); @@ -480,24 +563,21 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) * * 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) + Indirect *partial) { - struct ext4_block_alloc_info *block_i; - - block_i = EXT4_I(inode)->i_block_alloc_info; + ext4_fsblk_t goal; /* - * try the heuristic for sequential allocation, - * failing that at least try to get decent locality. + * XXX need to get goal block from mballoc's data structures */ - if (block_i && (block == block_i->last_alloc_logical_block + 1) - && (block_i->last_alloc_physical_block != 0)) { - return block_i->last_alloc_physical_block + 1; - } - return ext4_find_near(inode, partial); + goal = ext4_find_near(inode, partial); + goal = goal & EXT4_MAX_BLOCK_FILE_PHYS; + return goal; } /** @@ -512,10 +592,10 @@ static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t 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 long blks, - int blocks_to_boundary) +static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned int blks, + int blocks_to_boundary) { - unsigned long count = 0; + unsigned int count = 0; /* * Simple case, [t,d]Indirect block(s) has not allocated yet @@ -549,10 +629,11 @@ static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned long blks, * direct blocks */ 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) + 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; @@ -577,6 +658,8 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, if (*err) goto failed_out; + BUG_ON(current_block + count > EXT4_MAX_BLOCK_FILE_PHYS); + target -= count; /* allocate blocks for indirect blocks */ while (index < indirect_blks && count) { @@ -601,10 +684,18 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, if (!target) goto allocated; /* Now allocate data blocks */ - count = target; - /* allocating blocks for data blocks */ - current_block = ext4_new_blocks(handle, inode, iblock, - goal, &count, err); + 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); + BUG_ON(current_block + ar.len > EXT4_MAX_BLOCK_FILE_PHYS); + if (*err && (target == blks)) { /* * if the allocation failed and we didn't allocate @@ -614,13 +705,13 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, } if (!*err) { if (target == blks) { - /* - * save the new block number - * for the first direct block - */ + /* + * save the new block number + * for the first direct block + */ new_blocks[index] = current_block; } - blk_allocated += count; + blk_allocated += ar.len; } allocated: /* total number of blocks allocated for direct blocks */ @@ -628,7 +719,7 @@ allocated: *err = 0; return ret; failed_out: - for (i = 0; i i_sb->s_blocksize; int i, n = 0; @@ -692,8 +783,9 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, 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); - brelse(bh); goto failed; } @@ -701,22 +793,22 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, 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) { + 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++) + 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_journal_dirty_metadata"); - err = ext4_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); + err = ext4_handle_dirty_metadata(handle, inode, bh); if (err) goto failed; } @@ -728,7 +820,7 @@ failed: BUFFER_TRACE(branch[i].bh, "call jbd2_journal_forget"); ext4_journal_forget(handle, branch[i].bh); } - for (i = 0; i i_block_alloc_info; /* * 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 @@ -781,25 +872,10 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode, 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++); - } - - /* - * update the most recently allocated logical & physical block - * in i_block_alloc_info, to assist find the proper goal block for next - * allocation - */ - if (block_i) { - block_i->last_alloc_logical_block = block + blks - 1; - block_i->last_alloc_physical_block = - le32_to_cpu(where[num].key) + blks - 1; + *(where->p + i) = cpu_to_le32(current_block++); } /* We are done with atomic stuff, now do the rest of housekeeping */ - - inode->i_ctime = ext4_current_time(inode); - ext4_mark_inode_dirty(handle, inode); - /* had we spliced it onto indirect block? */ if (where->bh) { /* @@ -811,15 +887,15 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode, * generic_commit_write->__mark_inode_dirty->ext4_dirty_inode. */ jbd_debug(5, "splicing indirect only\n"); - BUFFER_TRACE(where->bh, "call ext4_journal_dirty_metadata"); - err = ext4_journal_dirty_metadata(handle, where->bh); + 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. - * Inode was dirtied above. */ + ext4_mark_inode_dirty(handle, inode); jbd_debug(5, "splicing direct\n"); } return err; @@ -837,6 +913,10 @@ err_out: } /* + * The ext4_ind_get_blocks() function handles non-extents inodes + * (i.e., using the traditional indirect/double-indirect i_blocks + * scheme) for ext4_get_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 @@ -854,15 +934,16 @@ err_out: * return = 0, if plain lookup failed. * return < 0, error case. * - * - * Need to be called with - * down_read(&EXT4_I(inode)->i_data_sem) if not allocating file system block - * (ie, create is zero). Otherwise down_write(&EXT4_I(inode)->i_data_sem) + * 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_get_blocks_handle(handle_t *handle, struct inode *inode, - ext4_lblk_t iblock, unsigned long maxblocks, - struct buffer_head *bh_result, - int create, int extend_disksize) +static int ext4_ind_get_blocks(handle_t *handle, struct inode *inode, + ext4_lblk_t iblock, unsigned int maxblocks, + struct buffer_head *bh_result, + int flags) { int err = -EIO; ext4_lblk_t offsets[4]; @@ -872,16 +953,13 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, int indirect_blks; int blocks_to_boundary = 0; int depth; - struct ext4_inode_info *ei = EXT4_I(inode); int count = 0; ext4_fsblk_t first_block = 0; - loff_t disksize; - J_ASSERT(!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)); - J_ASSERT(handle != NULL || create == 0); + J_ASSERT(handle != NULL || (flags & EXT4_GET_BLOCKS_CREATE) == 0); depth = ext4_block_to_path(inode, iblock, offsets, - &blocks_to_boundary); + &blocks_to_boundary); if (depth == 0) goto out; @@ -908,16 +986,12 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, } /* Next simple case - plain lookup or failed read of indirect block */ - if (!create || err == -EIO) + if ((flags & EXT4_GET_BLOCKS_CREATE) == 0 || err == -EIO) goto cleanup; /* - * Okay, we need to do block allocation. Lazily initialize the block - * allocation info here if necessary + * Okay, we need to do block allocation. */ - if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) - ext4_init_block_alloc_info(inode); - goal = ext4_find_goal(inode, iblock, partial); /* the number of blocks need to allocate for [d,t]indirect blocks */ @@ -933,8 +1007,8 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, * Block out ext4_truncate while we alter the tree */ err = ext4_alloc_branch(handle, inode, iblock, indirect_blks, - &count, goal, - offsets + (partial - chain), partial); + &count, goal, + offsets + (partial - chain), partial); /* * The ext4_splice_branch call will free and forget any buffers @@ -945,20 +1019,8 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, */ if (!err) err = ext4_splice_branch(handle, inode, iblock, - partial, indirect_blks, count); - /* - * i_disksize growing is protected by i_data_sem. Don't forget to - * protect it if you're about to implement concurrent - * ext4_get_block() -bzzz - */ - if (!err && extend_disksize) { - disksize = ((loff_t) iblock + count) << inode->i_blkbits; - if (disksize > i_size_read(inode)) - disksize = i_size_read(inode); - if (disksize > ei->i_disksize) - ei->i_disksize = disksize; - } - if (err) + partial, indirect_blks, count); + else goto cleanup; set_buffer_new(bh_result); @@ -980,6 +1042,17 @@ out: return err; } +qsize_t ext4_get_reserved_space(struct inode *inode) +{ + unsigned long long total; + + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + total = EXT4_I(inode)->i_reserved_data_blocks + + EXT4_I(inode)->i_reserved_meta_blocks; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + + return total; +} /* * Calculate the number of metadata blocks need to reserve * to allocate @blocks for non extent file based file @@ -1005,6 +1078,9 @@ static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks) */ static int ext4_calc_metadata_amount(struct inode *inode, int blocks) { + if (!blocks) + return 0; + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) return ext4_ext_calc_metadata_amount(inode, blocks); @@ -1025,36 +1101,110 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; - /* Account for allocated meta_blocks */ - mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; + if (mdb_free) { + /* Account for allocated meta_blocks */ + mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; - /* update fs free blocks counter for truncate case */ - percpu_counter_add(&sbi->s_freeblocks_counter, mdb_free); + /* update fs dirty blocks counter */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); + EXT4_I(inode)->i_allocated_meta_blocks = 0; + EXT4_I(inode)->i_reserved_meta_blocks = mdb; + } /* update per-inode reservations */ BUG_ON(used > EXT4_I(inode)->i_reserved_data_blocks); EXT4_I(inode)->i_reserved_data_blocks -= used; - - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - EXT4_I(inode)->i_reserved_meta_blocks = mdb; - EXT4_I(inode)->i_allocated_meta_blocks = 0; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + + /* + * free those over-booking quota for metadata blocks + */ + if (mdb_free) + vfs_dq_release_reservation_block(inode, mdb_free); + + /* + * If we have done all the pending block allocations and if + * there aren't any writers on the inode, we can discard the + * inode's preallocations. + */ + if (!total && (atomic_read(&inode->i_writecount) == 0)) + ext4_discard_preallocations(inode); +} + +static int check_block_validity(struct inode *inode, const char *msg, + sector_t logical, sector_t phys, int len) +{ + if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), phys, len)) { + ext4_error(inode->i_sb, msg, + "inode #%lu logical block %llu mapped to %llu " + "(size %d)", inode->i_ino, + (unsigned long long) logical, + (unsigned long long) phys, len); + return -EIO; + } + return 0; } -/* Maximum number of blocks we map for direct IO at once. */ -#define DIO_MAX_BLOCKS 4096 /* - * Number of credits we need for writing DIO_MAX_BLOCKS: - * We need sb + group descriptor + bitmap + inode -> 4 - * For B blocks with A block pointers per block we need: - * 1 (triple ind.) + (B/A/A + 2) (doubly ind.) + (B/A + 2) (indirect). - * If we plug in 4096 for B and 256 for A (for 1KB block size), we get 25. + * Return the number of contiguous dirty pages in a given inode + * starting at page frame idx. */ -#define DIO_CREDITS 25 +static pgoff_t ext4_num_dirty_pages(struct inode *inode, pgoff_t idx, + unsigned int max_pages) +{ + struct address_space *mapping = inode->i_mapping; + pgoff_t index; + struct pagevec pvec; + pgoff_t num = 0; + int i, nr_pages, done = 0; + if (max_pages == 0) + return 0; + pagevec_init(&pvec, 0); + while (!done) { + index = idx; + nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, + PAGECACHE_TAG_DIRTY, + (pgoff_t)PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + struct buffer_head *bh, *head; + + lock_page(page); + if (unlikely(page->mapping != mapping) || + !PageDirty(page) || + PageWriteback(page) || + page->index != idx) { + done = 1; + unlock_page(page); + break; + } + if (page_has_buffers(page)) { + bh = head = page_buffers(page); + do { + if (!buffer_delay(bh) && + !buffer_unwritten(bh)) + done = 1; + bh = bh->b_this_page; + } while (!done && (bh != head)); + } + unlock_page(page); + if (done) + break; + idx++; + num++; + if (num >= max_pages) + break; + } + pagevec_release(&pvec); + } + return num; +} /* - * The ext4_get_blocks_wrap() function try to look up the requested blocks, + * The ext4_get_blocks() function tries to look up the requested blocks, * and returns if the blocks are already mapped. * * Otherwise it takes the write lock of the i_data_sem and allocate blocks @@ -1062,7 +1212,7 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) * mapped. * * If file type is extents based, it will call ext4_ext_get_blocks(), - * Otherwise, call with ext4_get_blocks_handle() to handle indirect mapping + * Otherwise, call with ext4_ind_get_blocks() to handle indirect mapping * based files * * On success, it returns the number of blocks being mapped or allocate. @@ -1075,30 +1225,41 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) * * It returns the error in case of allocation failure. */ -int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, - unsigned long max_blocks, struct buffer_head *bh, - int create, int extend_disksize, int flag) +int ext4_get_blocks(handle_t *handle, struct inode *inode, sector_t block, + unsigned int max_blocks, struct buffer_head *bh, + int flags) { int retval; clear_buffer_mapped(bh); + clear_buffer_unwritten(bh); + ext_debug("ext4_get_blocks(): inode %lu, flag %d, max_blocks %u," + "logical block %lu\n", inode->i_ino, flags, max_blocks, + (unsigned long)block); /* - * Try to see if we can get the block without requesting - * for new file system block. + * Try to see if we can get the block without requesting a new + * file system block. */ down_read((&EXT4_I(inode)->i_data_sem)); if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { retval = ext4_ext_get_blocks(handle, inode, block, max_blocks, - bh, 0, 0); + bh, 0); } else { - retval = ext4_get_blocks_handle(handle, - inode, block, max_blocks, bh, 0, 0); + retval = ext4_ind_get_blocks(handle, inode, block, max_blocks, + bh, 0); } up_read((&EXT4_I(inode)->i_data_sem)); + if (retval > 0 && buffer_mapped(bh)) { + int ret = check_block_validity(inode, "file system corruption", + block, bh->b_blocknr, retval); + if (ret != 0) + return ret; + } + /* If it is only a block(s) look up */ - if (!create) + if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) return retval; /* @@ -1112,6 +1273,18 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, return retval; /* + * When we call get_blocks without the create flag, the + * BH_Unwritten flag could have gotten set if the blocks + * requested were part of a uninitialized extent. We need to + * clear this flag now that we are committed to convert all or + * part of the uninitialized extent to be an initialized + * extent. This is because we need to avoid the combination + * of BH_Unwritten and BH_Mapped flags being simultaneously + * set on the buffer_head. + */ + clear_buffer_unwritten(bh); + + /* * New blocks allocate and/or writing to uninitialized extent * will possibly result in updating i_data, so we take * the write lock of i_data_sem, and call get_blocks() @@ -1125,7 +1298,7 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, * let the underlying get_block() function know to * avoid double accounting */ - if (flag) + if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) EXT4_I(inode)->i_delalloc_reserved_flag = 1; /* * We need to check for EXT4 here because migrate @@ -1133,10 +1306,10 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, */ if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { retval = ext4_ext_get_blocks(handle, inode, block, max_blocks, - bh, create, extend_disksize); + bh, flags); } else { - retval = ext4_get_blocks_handle(handle, inode, block, - max_blocks, bh, create, extend_disksize); + retval = ext4_ind_get_blocks(handle, inode, block, + max_blocks, bh, flags); if (retval > 0 && buffer_new(bh)) { /* @@ -1144,39 +1317,48 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, * i_data's format changing. Force the migrate * to fail by clearing migrate flags */ - EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags & - ~EXT4_EXT_MIGRATE; + EXT4_I(inode)->i_state &= ~EXT4_STATE_EXT_MIGRATE; } } - if (flag) { + if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) EXT4_I(inode)->i_delalloc_reserved_flag = 0; - /* - * Update reserved blocks/metadata blocks - * after successful block allocation - * which were deferred till now - */ - if ((retval > 0) && buffer_delay(bh)) - ext4_da_update_reserve_space(inode, retval); - } + + /* + * Update reserved blocks/metadata blocks after successful + * block allocation which had been deferred till now. + */ + if ((retval > 0) && (flags & EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE)) + ext4_da_update_reserve_space(inode, retval); up_write((&EXT4_I(inode)->i_data_sem)); + if (retval > 0 && buffer_mapped(bh)) { + int ret = check_block_validity(inode, "file system " + "corruption after allocation", + block, bh->b_blocknr, retval); + if (ret != 0) + return ret; + } return retval; } -static int ext4_get_block(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) +/* Maximum number of blocks we map for direct IO at once. */ +#define DIO_MAX_BLOCKS 4096 + +int ext4_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) { handle_t *handle = ext4_journal_current_handle(); int ret = 0, started = 0; unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + int dio_credits; if (create && !handle) { /* Direct IO write... */ if (max_blocks > DIO_MAX_BLOCKS) max_blocks = DIO_MAX_BLOCKS; - handle = ext4_journal_start(inode, DIO_CREDITS + - 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb)); + dio_credits = ext4_chunk_trans_blocks(inode, max_blocks); + handle = ext4_journal_start(inode, dio_credits); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; @@ -1184,8 +1366,8 @@ static int ext4_get_block(struct inode *inode, sector_t iblock, started = 1; } - ret = ext4_get_blocks_wrap(handle, inode, iblock, - max_blocks, bh_result, create, 0, 0); + ret = ext4_get_blocks(handle, inode, iblock, max_blocks, bh_result, + create ? EXT4_GET_BLOCKS_CREATE : 0); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); ret = 0; @@ -1204,17 +1386,19 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, { struct buffer_head dummy; int fatal = 0, err; + int flags = 0; J_ASSERT(handle != NULL || create == 0); dummy.b_state = 0; dummy.b_blocknr = -1000; buffer_trace_init(&dummy.b_history); - err = ext4_get_blocks_wrap(handle, inode, block, 1, - &dummy, create, 1, 0); + if (create) + flags |= EXT4_GET_BLOCKS_CREATE; + err = ext4_get_blocks(handle, inode, block, 1, &dummy, flags); /* - * ext4_get_blocks_handle() returns number of blocks - * mapped. 0 in case of a HOLE. + * ext4_get_blocks() returns number of blocks mapped. 0 in + * case of a HOLE. */ if (err > 0) { if (err > 1) @@ -1244,12 +1428,12 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, BUFFER_TRACE(bh, "call get_create_access"); fatal = ext4_journal_get_create_access(handle, bh); if (!fatal && !buffer_uptodate(bh)) { - memset(bh->b_data,0,inode->i_sb->s_blocksize); + memset(bh->b_data, 0, inode->i_sb->s_blocksize); set_buffer_uptodate(bh); } unlock_buffer(bh); - BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); - err = ext4_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); + err = ext4_handle_dirty_metadata(handle, inode, bh); if (!fatal) fatal = err; } else { @@ -1269,7 +1453,7 @@ err: struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, ext4_lblk_t block, int create, int *err) { - struct buffer_head * bh; + struct buffer_head *bh; bh = ext4_getblk(handle, inode, block, create, err); if (!bh) @@ -1285,13 +1469,13 @@ struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, return NULL; } -static int walk_page_buffers( handle_t *handle, - struct buffer_head *head, - unsigned from, - unsigned to, - int *partial, - int (*fn)( handle_t *handle, - struct buffer_head *bh)) +static int walk_page_buffers(handle_t *handle, + struct buffer_head *head, + unsigned from, + unsigned to, + int *partial, + int (*fn)(handle_t *handle, + struct buffer_head *bh)) { struct buffer_head *bh; unsigned block_start, block_end; @@ -1299,10 +1483,9 @@ static int walk_page_buffers( handle_t *handle, int err, ret = 0; struct buffer_head *next; - for ( bh = head, block_start = 0; - ret == 0 && (bh != head || !block_start); - block_start = block_end, bh = next) - { + for (bh = head, block_start = 0; + ret == 0 && (bh != head || !block_start); + block_start = block_end, bh = next) { next = bh->b_this_page; block_end = block_start + blocksize; if (block_end <= from || block_start >= to) { @@ -1343,7 +1526,7 @@ static int walk_page_buffers( handle_t *handle, * write. */ static int do_journal_get_write_access(handle_t *handle, - struct buffer_head *bh) + struct buffer_head *bh) { if (!buffer_mapped(bh) || buffer_freed(bh)) return 0; @@ -1351,29 +1534,39 @@ static int do_journal_get_write_access(handle_t *handle, } static int ext4_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned flags, - struct page **pagep, void **fsdata) + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) { - struct inode *inode = mapping->host; - int ret, needed_blocks = ext4_writepage_trans_blocks(inode); + struct inode *inode = mapping->host; + int ret, needed_blocks; handle_t *handle; int retries = 0; - struct page *page; - pgoff_t index; - unsigned from, to; + struct page *page; + pgoff_t index; + unsigned from, to; - index = pos >> PAGE_CACHE_SHIFT; - from = pos & (PAGE_CACHE_SIZE - 1); - to = from + len; + trace_ext4_write_begin(inode, pos, len, flags); + /* + * Reserve one block more for addition to orphan list in case + * we allocate blocks but write fails for some reason + */ + needed_blocks = ext4_writepage_trans_blocks(inode) + 1; + index = pos >> PAGE_CACHE_SHIFT; + from = pos & (PAGE_CACHE_SIZE - 1); + to = from + len; retry: - handle = ext4_journal_start(inode, needed_blocks); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; + handle = ext4_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; } - page = __grab_cache_page(mapping, index); + /* We cannot recurse into the filesystem as the transaction is already + * started */ + flags |= AOP_FLAG_NOFS; + + page = grab_cache_page_write_begin(mapping, index, flags); if (!page) { ext4_journal_stop(handle); ret = -ENOMEM; @@ -1382,7 +1575,7 @@ retry: *pagep = page; ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, - ext4_get_block); + ext4_get_block); if (!ret && ext4_should_journal_data(inode)) { ret = walk_page_buffers(handle, page_buffers(page), @@ -1390,9 +1583,31 @@ retry: } if (ret) { - unlock_page(page); + unlock_page(page); + page_cache_release(page); + /* + * block_write_begin may have instantiated a few blocks + * outside i_size. Trim these off again. Don't need + * i_size_read because we hold i_mutex. + * + * Add inode to orphan list in case we crash before + * truncate finishes + */ + if (pos + len > inode->i_size && ext4_can_truncate(inode)) + ext4_orphan_add(handle, inode); + ext4_journal_stop(handle); - page_cache_release(page); + if (pos + len > inode->i_size) { + ext4_truncate(inode); + /* + * If truncate failed early the inode might + * still be on the orphan list; we need to + * make sure the inode is removed from the + * orphan list in that case. + */ + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + } } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -1407,7 +1622,53 @@ static int write_end_fn(handle_t *handle, struct buffer_head *bh) if (!buffer_mapped(bh) || buffer_freed(bh)) return 0; set_buffer_uptodate(bh); - return ext4_journal_dirty_metadata(handle, bh); + return ext4_handle_dirty_metadata(handle, NULL, bh); +} + +static int ext4_generic_write_end(struct file *file, + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + int i_size_changed = 0; + struct inode *inode = mapping->host; + handle_t *handle = ext4_journal_current_handle(); + + copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); + + /* + * No need to use i_size_read() here, the i_size + * cannot change under us because we hold i_mutex. + * + * But it's important to update i_size while still holding page lock: + * page writeout could otherwise come in and zero beyond i_size. + */ + if (pos + copied > inode->i_size) { + i_size_write(inode, pos + copied); + i_size_changed = 1; + } + + if (pos + copied > EXT4_I(inode)->i_disksize) { + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_update_i_disksize(inode, (pos + copied)); + i_size_changed = 1; + } + unlock_page(page); + page_cache_release(page); + + /* + * Don't mark the inode dirty under page lock. First, it unnecessarily + * makes the holding time of page lock longer. Second, it forces lock + * ordering of page lock and transaction start for journaling + * filesystems. + */ + if (i_size_changed) + ext4_mark_inode_dirty(handle, inode); + + return copied; } /* @@ -1418,34 +1679,27 @@ static int write_end_fn(handle_t *handle, struct buffer_head *bh) * buffers are managed internally. */ static int ext4_ordered_write_end(struct file *file, - struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) { handle_t *handle = ext4_journal_current_handle(); struct inode *inode = mapping->host; - unsigned from, to; int ret = 0, ret2; - from = pos & (PAGE_CACHE_SIZE - 1); - to = from + len; - + trace_ext4_ordered_write_end(inode, pos, len, copied); ret = ext4_jbd2_file_inode(handle, inode); if (ret == 0) { - /* - * generic_write_end() will run mark_inode_dirty() if i_size - * changes. So let's piggyback the i_disksize mark_inode_dirty - * into that. - */ - loff_t new_i_size; - - new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = new_i_size; - ret2 = generic_write_end(file, mapping, pos, len, copied, + ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; + if (pos + len > inode->i_size && ext4_can_truncate(inode)) + /* if we have allocated more blocks and copied + * less. We will have blocks allocated outside + * inode->i_size. So truncate them + */ + ext4_orphan_add(handle, inode); if (ret2 < 0) ret = ret2; } @@ -1453,26 +1707,41 @@ static int ext4_ordered_write_end(struct file *file, if (!ret) ret = ret2; + if (pos + len > inode->i_size) { + ext4_truncate(inode); + /* + * If truncate failed early the inode might still be + * on the orphan list; we need to make sure the inode + * is removed from the orphan list in that case. + */ + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + } + + return ret ? ret : copied; } static int ext4_writeback_write_end(struct file *file, - struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) { handle_t *handle = ext4_journal_current_handle(); struct inode *inode = mapping->host; int ret = 0, ret2; - loff_t new_i_size; - - new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = new_i_size; - ret2 = generic_write_end(file, mapping, pos, len, copied, + trace_ext4_writeback_write_end(inode, pos, len, copied); + ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; + if (pos + len > inode->i_size && ext4_can_truncate(inode)) + /* if we have allocated more blocks and copied + * less. We will have blocks allocated outside + * inode->i_size. So truncate them + */ + ext4_orphan_add(handle, inode); + if (ret2 < 0) ret = ret2; @@ -1480,20 +1749,33 @@ static int ext4_writeback_write_end(struct file *file, if (!ret) ret = ret2; + if (pos + len > inode->i_size) { + ext4_truncate(inode); + /* + * If truncate failed early the inode might still be + * on the orphan list; we need to make sure the inode + * is removed from the orphan list in that case. + */ + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + } + return ret ? ret : copied; } static int ext4_journalled_write_end(struct file *file, - struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) { handle_t *handle = ext4_journal_current_handle(); struct inode *inode = mapping->host; int ret = 0, ret2; int partial = 0; unsigned from, to; + loff_t new_i_size; + trace_ext4_journalled_write_end(inode, pos, len, copied); from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; @@ -1507,35 +1789,55 @@ static int ext4_journalled_write_end(struct file *file, to, &partial, write_end_fn); if (!partial) SetPageUptodate(page); - if (pos+copied > inode->i_size) + new_i_size = pos + copied; + if (new_i_size > inode->i_size) i_size_write(inode, pos+copied); EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; - if (inode->i_size > EXT4_I(inode)->i_disksize) { - EXT4_I(inode)->i_disksize = inode->i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, new_i_size); ret2 = ext4_mark_inode_dirty(handle, inode); if (!ret) ret = ret2; } unlock_page(page); + page_cache_release(page); + if (pos + len > inode->i_size && ext4_can_truncate(inode)) + /* if we have allocated more blocks and copied + * less. We will have blocks allocated outside + * inode->i_size. So truncate them + */ + ext4_orphan_add(handle, inode); + ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; - page_cache_release(page); + if (pos + len > inode->i_size) { + ext4_truncate(inode); + /* + * If truncate failed early the inode might still be + * on the orphan list; we need to make sure the inode + * is removed from the orphan list in that case. + */ + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + } return ret ? ret : copied; } static int ext4_da_reserve_space(struct inode *inode, int nrblocks) { - struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - unsigned long md_needed, mdblocks, total = 0; + int retries = 0; + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + unsigned long md_needed, mdblocks, total = 0; /* * recalculate the amount of metadata blocks to reserve * in order to allocate nrblocks * worse case is one extent per block */ +repeat: spin_lock(&EXT4_I(inode)->i_block_reservation_lock); total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks; mdblocks = ext4_calc_metadata_amount(inode, total); @@ -1544,13 +1846,25 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; total = md_needed + nrblocks; - if (ext4_has_free_blocks(sbi, total) < total) { + /* + * Make quota reservation here to prevent quota overflow + * later. Real quota accounting is done at pages writeout + * time. + */ + if (vfs_dq_reserve_block(inode, total)) { spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - return -ENOSPC; + return -EDQUOT; } - /* reduce fs free blocks counter */ - percpu_counter_sub(&sbi->s_freeblocks_counter, total); + if (ext4_claim_free_blocks(sbi, total)) { + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + vfs_dq_release_reservation_block(inode, total); + if (ext4_should_retry_alloc(inode->i_sb, &retries)) { + yield(); + goto repeat; + } + return -ENOSPC; + } EXT4_I(inode)->i_reserved_data_blocks += nrblocks; EXT4_I(inode)->i_reserved_meta_blocks = mdblocks; @@ -1563,7 +1877,25 @@ static void ext4_da_release_space(struct inode *inode, int to_free) struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); int total, mdb, mdb_free, release; + if (!to_free) + return; /* Nothing to release, exit */ + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + + if (!EXT4_I(inode)->i_reserved_data_blocks) { + /* + * if there is no reserved blocks, but we try to free some + * then the counter is messed up somewhere. + * but since this function is called from invalidate + * page, it's harmless to return without any action + */ + printk(KERN_INFO "ext4 delalloc try to release %d reserved " + "blocks for inode %lu, but there is no reserved " + "data blocks\n", to_free, inode->i_ino); + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return; + } + /* recalculate the number of metablocks still need to be reserved */ total = EXT4_I(inode)->i_reserved_data_blocks - to_free; mdb = ext4_calc_metadata_amount(inode, total); @@ -1574,8 +1906,8 @@ static void ext4_da_release_space(struct inode *inode, int to_free) release = to_free + mdb_free; - /* update fs free blocks counter for truncate case */ - percpu_counter_add(&sbi->s_freeblocks_counter, release); + /* update fs dirty blocks counter for truncate case */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, release); /* update per-inode reservations */ BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks); @@ -1584,10 +1916,12 @@ static void ext4_da_release_space(struct inode *inode, int to_free) BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); EXT4_I(inode)->i_reserved_meta_blocks = mdb; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + + vfs_dq_release_reservation_block(inode, release); } static void ext4_da_page_release_reservation(struct page *page, - unsigned long offset) + unsigned long offset) { int to_release = 0; struct buffer_head *head, *bh; @@ -1611,22 +1945,13 @@ static void ext4_da_page_release_reservation(struct page *page, * Delayed allocation stuff */ -struct mpage_da_data { - struct inode *inode; - struct buffer_head lbh; /* extent of blocks */ - unsigned long first_page, next_page; /* extent of pages */ - get_block_t *get_block; - struct writeback_control *wbc; -}; - /* * mpage_da_submit_io - walks through extent of pages and try to write - * them with __mpage_writepage() + * them with writepage() call back * * @mpd->inode: inode * @mpd->first_page: first page of the extent * @mpd->next_page: page after the last page of the extent - * @mpd->get_block: the filesystem's block mapper function * * By the time mpage_da_submit_io() is called we expect all blocks * to be allocated. this may be wrong if allocation failed. @@ -1635,25 +1960,25 @@ struct mpage_da_data { */ static int mpage_da_submit_io(struct mpage_da_data *mpd) { - struct address_space *mapping = mpd->inode->i_mapping; - struct mpage_data mpd_pp = { - .bio = NULL, - .last_block_in_bio = 0, - .get_block = mpd->get_block, - .use_writepage = 1, - }; - int ret = 0, err, nr_pages, i; - unsigned long index, end; + long pages_skipped; struct pagevec pvec; + unsigned long index, end; + int ret = 0, err, nr_pages, i; + struct inode *inode = mpd->inode; + struct address_space *mapping = inode->i_mapping; BUG_ON(mpd->next_page <= mpd->first_page); - - pagevec_init(&pvec, 0); + /* + * We need to start from the first_page to the next_page - 1 + * to make sure we also write the mapped dirty buffer_heads. + * If we look at mpd->b_blocknr we would only be looking + * at the currently mapped buffer_heads. + */ index = mpd->first_page; end = mpd->next_page - 1; + pagevec_init(&pvec, 0); while (index <= end) { - /* XXX: optimize tail */ nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); if (nr_pages == 0) break; @@ -1665,8 +1990,17 @@ static int mpage_da_submit_io(struct mpage_da_data *mpd) break; index++; - err = __mpage_writepage(page, mpd->wbc, &mpd_pp); + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + pages_skipped = mpd->wbc->pages_skipped; + err = mapping->a_ops->writepage(page, mpd->wbc); + if (!err && (pages_skipped == mpd->wbc->pages_skipped)) + /* + * have successfully written the page + * without skipping the same + */ + mpd->pages_written++; /* * In error case, we have to continue because * remaining pages are still locked @@ -1677,9 +2011,6 @@ static int mpage_da_submit_io(struct mpage_da_data *mpd) } pagevec_release(&pvec); } - if (mpd_pp.bio) - mpage_bio_submit(WRITE, mpd_pp.bio); - return ret; } @@ -1692,7 +2023,7 @@ static int mpage_da_submit_io(struct mpage_da_data *mpd) * @logical - first logical block to start assignment with * * the function goes through all passed space and put actual disk - * block numbers into buffer heads, dropping BH_Delay + * block numbers into buffer heads, dropping BH_Delay and BH_Unwritten */ static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, struct buffer_head *exbh) @@ -1702,7 +2033,7 @@ static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, int blocks = exbh->b_size >> inode->i_blkbits; sector_t pblock = exbh->b_blocknr, cur_logical; struct buffer_head *head, *bh; - unsigned long index, end; + pgoff_t index, end; struct pagevec pvec; int nr_pages, i; @@ -1742,9 +2073,24 @@ static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, do { if (cur_logical >= logical + blocks) break; - if (buffer_delay(bh)) { - bh->b_blocknr = pblock; - clear_buffer_delay(bh); + + if (buffer_delay(bh) || + buffer_unwritten(bh)) { + + BUG_ON(bh->b_bdev != inode->i_sb->s_bdev); + + if (buffer_delay(bh)) { + clear_buffer_delay(bh); + bh->b_blocknr = pblock; + } else { + /* + * unwritten already should have + * blocknr assigned. Verify that + */ + clear_buffer_unwritten(bh); + BUG_ON(bh->b_blocknr != pblock); + } + } else if (buffer_mapped(bh)) BUG_ON(bh->b_blocknr != pblock); @@ -1772,62 +2118,192 @@ static inline void __unmap_underlying_blocks(struct inode *inode, unmap_underlying_metadata(bdev, bh->b_blocknr + i); } -/* - * mpage_da_map_blocks - go through given space - * - * @mpd->lbh - bh describing space - * @mpd->get_block - the filesystem's block mapper function +static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd, + sector_t logical, long blk_cnt) +{ + int nr_pages, i; + pgoff_t index, end; + struct pagevec pvec; + struct inode *inode = mpd->inode; + struct address_space *mapping = inode->i_mapping; + + index = logical >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + end = (logical + blk_cnt - 1) >> + (PAGE_CACHE_SHIFT - inode->i_blkbits); + while (index <= end) { + nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + index = page->index; + if (index > end) + break; + index++; + + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + block_invalidatepage(page, 0); + ClearPageUptodate(page); + unlock_page(page); + } + } + return; +} + +static void ext4_print_free_blocks(struct inode *inode) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + printk(KERN_CRIT "Total free blocks count %lld\n", + ext4_count_free_blocks(inode->i_sb)); + printk(KERN_CRIT "Free/Dirty block details\n"); + printk(KERN_CRIT "free_blocks=%lld\n", + (long long) percpu_counter_sum(&sbi->s_freeblocks_counter)); + printk(KERN_CRIT "dirty_blocks=%lld\n", + (long long) percpu_counter_sum(&sbi->s_dirtyblocks_counter)); + printk(KERN_CRIT "Block reservation details\n"); + printk(KERN_CRIT "i_reserved_data_blocks=%u\n", + EXT4_I(inode)->i_reserved_data_blocks); + printk(KERN_CRIT "i_reserved_meta_blocks=%u\n", + EXT4_I(inode)->i_reserved_meta_blocks); + return; +} + +/* + * mpage_da_map_blocks - go through given space + * + * @mpd - bh describing space * * The function skips space we know is already mapped to disk blocks. * - * The function ignores errors ->get_block() returns, thus real - * error handling is postponed to __mpage_writepage() */ -static void mpage_da_map_blocks(struct mpage_da_data *mpd) +static int mpage_da_map_blocks(struct mpage_da_data *mpd) { - struct buffer_head *lbh = &mpd->lbh; - int err = 0, remain = lbh->b_size; - sector_t next = lbh->b_blocknr; + int err, blks, get_blocks_flags; struct buffer_head new; + sector_t next = mpd->b_blocknr; + unsigned max_blocks = mpd->b_size >> mpd->inode->i_blkbits; + loff_t disksize = EXT4_I(mpd->inode)->i_disksize; + handle_t *handle = NULL; /* * We consider only non-mapped and non-allocated blocks */ - if (buffer_mapped(lbh) && !buffer_delay(lbh)) - return; + if ((mpd->b_state & (1 << BH_Mapped)) && + !(mpd->b_state & (1 << BH_Delay)) && + !(mpd->b_state & (1 << BH_Unwritten))) + return 0; - while (remain) { - new.b_state = lbh->b_state; - new.b_blocknr = 0; - new.b_size = remain; - err = mpd->get_block(mpd->inode, next, &new, 1); - if (err) { - /* - * Rather than implement own error handling - * here, we just leave remaining blocks - * unallocated and try again with ->writepage() - */ - break; - } - BUG_ON(new.b_size == 0); + /* + * If we didn't accumulate anything to write simply return + */ + if (!mpd->b_size) + return 0; + + handle = ext4_journal_current_handle(); + BUG_ON(!handle); + + /* + * Call ext4_get_blocks() to allocate any delayed allocation + * blocks, or to convert an uninitialized extent to be + * initialized (in the case where we have written into + * one or more preallocated blocks). + * + * We pass in the magic EXT4_GET_BLOCKS_DELALLOC_RESERVE to + * indicate that we are on the delayed allocation path. This + * affects functions in many different parts of the allocation + * call path. This flag exists primarily because we don't + * want to change *many* call functions, so ext4_get_blocks() + * will set the magic i_delalloc_reserved_flag once the + * inode's allocation semaphore is taken. + * + * If the blocks in questions were delalloc blocks, set + * EXT4_GET_BLOCKS_DELALLOC_RESERVE so the delalloc accounting + * variables are updated after the blocks have been allocated. + */ + new.b_state = 0; + get_blocks_flags = (EXT4_GET_BLOCKS_CREATE | + EXT4_GET_BLOCKS_DELALLOC_RESERVE); + if (mpd->b_state & (1 << BH_Delay)) + get_blocks_flags |= EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE; + blks = ext4_get_blocks(handle, mpd->inode, next, max_blocks, + &new, get_blocks_flags); + if (blks < 0) { + err = blks; + /* + * If get block returns with error we simply + * return. Later writepage will redirty the page and + * writepages will find the dirty page again + */ + if (err == -EAGAIN) + return 0; - if (buffer_new(&new)) - __unmap_underlying_blocks(mpd->inode, &new); + if (err == -ENOSPC && + ext4_count_free_blocks(mpd->inode->i_sb)) { + mpd->retval = err; + return 0; + } /* - * If blocks are delayed marked, we need to - * put actual blocknr and drop delayed bit + * get block failure will cause us to loop in + * writepages, because a_ops->writepage won't be able + * to make progress. The page will be redirtied by + * writepage and writepages will again try to write + * the same. */ - if (buffer_delay(lbh)) - mpage_put_bnr_to_bhs(mpd, next, &new); + ext4_msg(mpd->inode->i_sb, KERN_CRIT, + "delayed block allocation failed for inode %lu at " + "logical offset %llu with max blocks %zd with " + "error %d\n", mpd->inode->i_ino, + (unsigned long long) next, + mpd->b_size >> mpd->inode->i_blkbits, err); + printk(KERN_CRIT "This should not happen!! " + "Data will be lost\n"); + if (err == -ENOSPC) { + ext4_print_free_blocks(mpd->inode); + } + /* invalidate all the pages */ + ext4_da_block_invalidatepages(mpd, next, + mpd->b_size >> mpd->inode->i_blkbits); + return err; + } + BUG_ON(blks == 0); + + new.b_size = (blks << mpd->inode->i_blkbits); + + if (buffer_new(&new)) + __unmap_underlying_blocks(mpd->inode, &new); + + /* + * If blocks are delayed marked, we need to + * put actual blocknr and drop delayed bit + */ + if ((mpd->b_state & (1 << BH_Delay)) || + (mpd->b_state & (1 << BH_Unwritten))) + mpage_put_bnr_to_bhs(mpd, next, &new); + + if (ext4_should_order_data(mpd->inode)) { + err = ext4_jbd2_file_inode(handle, mpd->inode); + if (err) + return err; + } - /* go for the remaining blocks */ - next += new.b_size >> mpd->inode->i_blkbits; - remain -= new.b_size; + /* + * Update on-disk size along with block allocation. + */ + disksize = ((loff_t) next + blks) << mpd->inode->i_blkbits; + if (disksize > i_size_read(mpd->inode)) + disksize = i_size_read(mpd->inode); + if (disksize > EXT4_I(mpd->inode)->i_disksize) { + ext4_update_i_disksize(mpd->inode, disksize); + return ext4_mark_inode_dirty(handle, mpd->inode); } + + return 0; } -#define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | (1 << BH_Delay)) +#define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | \ + (1 << BH_Delay) | (1 << BH_Unwritten)) /* * mpage_add_bh_to_extent - try to add one more block to extent of blocks @@ -1839,43 +2315,67 @@ static void mpage_da_map_blocks(struct mpage_da_data *mpd) * the function is used to collect contig. blocks in same state */ static void mpage_add_bh_to_extent(struct mpage_da_data *mpd, - sector_t logical, struct buffer_head *bh) + sector_t logical, size_t b_size, + unsigned long b_state) { - struct buffer_head *lbh = &mpd->lbh; sector_t next; + int nrblocks = mpd->b_size >> mpd->inode->i_blkbits; - next = lbh->b_blocknr + (lbh->b_size >> mpd->inode->i_blkbits); - + /* check if thereserved journal credits might overflow */ + if (!(EXT4_I(mpd->inode)->i_flags & EXT4_EXTENTS_FL)) { + if (nrblocks >= EXT4_MAX_TRANS_DATA) { + /* + * With non-extent format we are limited by the journal + * credit available. Total credit needed to insert + * nrblocks contiguous blocks is dependent on the + * nrblocks. So limit nrblocks. + */ + goto flush_it; + } else if ((nrblocks + (b_size >> mpd->inode->i_blkbits)) > + EXT4_MAX_TRANS_DATA) { + /* + * Adding the new buffer_head would make it cross the + * allowed limit for which we have journal credit + * reserved. So limit the new bh->b_size + */ + b_size = (EXT4_MAX_TRANS_DATA - nrblocks) << + mpd->inode->i_blkbits; + /* we will do mpage_da_submit_io in the next loop */ + } + } /* * First block in the extent */ - if (lbh->b_size == 0) { - lbh->b_blocknr = logical; - lbh->b_size = bh->b_size; - lbh->b_state = bh->b_state & BH_FLAGS; + if (mpd->b_size == 0) { + mpd->b_blocknr = logical; + mpd->b_size = b_size; + mpd->b_state = b_state & BH_FLAGS; return; } + next = mpd->b_blocknr + nrblocks; /* * Can we merge the block to our big extent? */ - if (logical == next && (bh->b_state & BH_FLAGS) == lbh->b_state) { - lbh->b_size += bh->b_size; + if (logical == next && (b_state & BH_FLAGS) == mpd->b_state) { + mpd->b_size += b_size; return; } +flush_it: /* * We couldn't merge the block to our extent, so we * need to flush current extent and start new one */ - mpage_da_map_blocks(mpd); + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); + mpd->io_done = 1; + return; +} - /* - * Now start a new extent - */ - lbh->b_size = bh->b_size; - lbh->b_state = bh->b_state & BH_FLAGS; - lbh->b_blocknr = logical; +static int ext4_bh_delay_or_unwritten(handle_t *handle, struct buffer_head *bh) +{ + return (buffer_delay(bh) || buffer_unwritten(bh)) && buffer_dirty(bh); } /* @@ -1892,20 +2392,38 @@ static int __mpage_da_writepage(struct page *page, { struct mpage_da_data *mpd = data; struct inode *inode = mpd->inode; - struct buffer_head *bh, *head, fake; + struct buffer_head *bh, *head; sector_t logical; + if (mpd->io_done) { + /* + * Rest of the page in the page_vec + * redirty then and skip then. We will + * try to write them again after + * starting a new transaction + */ + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return MPAGE_DA_EXTENT_TAIL; + } /* * Can we merge this page to current extent? */ if (mpd->next_page != page->index) { /* * Nope, we can't. So, we map non-allocated blocks - * and start IO on them using __mpage_writepage() + * and start IO on them using writepage() */ if (mpd->next_page != mpd->first_page) { - mpage_da_map_blocks(mpd); - mpage_da_submit_io(mpd); + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); + /* + * skip rest of the page in the page_vec + */ + mpd->io_done = 1; + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return MPAGE_DA_EXTENT_TAIL; } /* @@ -1916,9 +2434,9 @@ static int __mpage_da_writepage(struct page *page, /* * ... and blocks */ - mpd->lbh.b_size = 0; - mpd->lbh.b_state = 0; - mpd->lbh.b_blocknr = 0; + mpd->b_size = 0; + mpd->b_state = 0; + mpd->b_blocknr = 0; } mpd->next_page = page->index + 1; @@ -1926,16 +2444,10 @@ static int __mpage_da_writepage(struct page *page, (PAGE_CACHE_SHIFT - inode->i_blkbits); if (!page_has_buffers(page)) { - /* - * There is no attached buffer heads yet (mmap?) - * we treat the page asfull of dirty blocks - */ - bh = &fake; - bh->b_size = PAGE_CACHE_SIZE; - bh->b_state = 0; - set_buffer_dirty(bh); - set_buffer_uptodate(bh); - mpage_add_bh_to_extent(mpd, logical, bh); + mpage_add_bh_to_extent(mpd, logical, PAGE_CACHE_SIZE, + (1 << BH_Dirty) | (1 << BH_Uptodate)); + if (mpd->io_done) + return MPAGE_DA_EXTENT_TAIL; } else { /* * Page with regular buffer heads, just add all dirty ones @@ -1944,8 +2456,30 @@ static int __mpage_da_writepage(struct page *page, bh = head; do { BUG_ON(buffer_locked(bh)); - if (buffer_dirty(bh)) - mpage_add_bh_to_extent(mpd, logical, bh); + /* + * We need to try to allocate + * unmapped blocks in the same page. + * Otherwise we won't make progress + * with the page in ext4_writepage + */ + if (ext4_bh_delay_or_unwritten(NULL, bh)) { + mpage_add_bh_to_extent(mpd, logical, + bh->b_size, + bh->b_state); + if (mpd->io_done) + return MPAGE_DA_EXTENT_TAIL; + } else if (buffer_dirty(bh) && (buffer_mapped(bh))) { + /* + * mapped dirty buffer. We need to update + * the b_state because we look at + * b_state in mpage_da_map_blocks. We don't + * update b_size because if we find an + * unmapped buffer_head later we need to + * use the b_state flag of that buffer_head. + */ + if (mpd->b_size == 0) + mpd->b_state = bh->b_state & BH_FLAGS; + } logical++; } while ((bh = bh->b_this_page) != head); } @@ -1954,67 +2488,25 @@ static int __mpage_da_writepage(struct page *page, } /* - * mpage_da_writepages - walk the list of dirty pages of the given - * address space, allocates non-allocated blocks, maps newly-allocated - * blocks to existing bhs and issue IO them + * This is a special get_blocks_t callback which is used by + * ext4_da_write_begin(). It will either return mapped block or + * reserve space for a single block. * - * @mapping: address space structure to write - * @wbc: subtract the number of written pages from *@wbc->nr_to_write - * @get_block: the filesystem's block mapper function. + * For delayed buffer_head we have BH_Mapped, BH_New, BH_Delay set. + * We also have b_blocknr = -1 and b_bdev initialized properly * - * This is a library function, which implements the writepages() - * address_space_operation. - * - * In order to avoid duplication of logic that deals with partial pages, - * multiple bio per page, etc, we find non-allocated blocks, allocate - * them with minimal calls to ->get_block() and re-use __mpage_writepage() - * - * It's important that we call __mpage_writepage() only once for each - * involved page, otherwise we'd have to implement more complicated logic - * to deal with pages w/o PG_lock or w/ PG_writeback and so on. - * - * See comments to mpage_writepages() - */ -static int mpage_da_writepages(struct address_space *mapping, - struct writeback_control *wbc, - get_block_t get_block) -{ - struct mpage_da_data mpd; - int ret; - - if (!get_block) - return generic_writepages(mapping, wbc); - - mpd.wbc = wbc; - mpd.inode = mapping->host; - mpd.lbh.b_size = 0; - mpd.lbh.b_state = 0; - mpd.lbh.b_blocknr = 0; - mpd.first_page = 0; - mpd.next_page = 0; - mpd.get_block = get_block; - - ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, &mpd); - - /* - * Handle last extent of pages - */ - if (mpd.next_page != mpd.first_page) { - mpage_da_map_blocks(&mpd); - mpage_da_submit_io(&mpd); - } - - return ret; -} - -/* - * this is a special callback for ->write_begin() only - * it's intention is to return mapped block or reserve space + * For unwritten buffer_head we have BH_Mapped, BH_New, BH_Unwritten set. + * We also have b_blocknr = physicalblock mapping unwritten extent and b_bdev + * initialized properly. */ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { int ret = 0; + sector_t invalid_block = ~((sector_t) 0xffff); + + if (invalid_block < ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es)) + invalid_block = ~0; BUG_ON(create == 0); BUG_ON(bh_result->b_size != inode->i_sb->s_blocksize); @@ -2024,7 +2516,7 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, * preallocated blocks are unmapped but should treated * the same as allocated blocks. */ - ret = ext4_get_blocks_wrap(NULL, inode, iblock, 1, bh_result, 0, 0, 0); + ret = ext4_get_blocks(NULL, inode, iblock, 1, bh_result, 0); if ((ret == 0) && !buffer_delay(bh_result)) { /* the block isn't (pre)allocated yet, let's reserve space */ /* @@ -2036,108 +2528,168 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, /* not enough space to reserve */ return ret; - map_bh(bh_result, inode->i_sb, 0); + map_bh(bh_result, inode->i_sb, invalid_block); set_buffer_new(bh_result); set_buffer_delay(bh_result); } else if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); + if (buffer_unwritten(bh_result)) { + /* A delayed write to unwritten bh should + * be marked new and mapped. Mapped ensures + * that we don't do get_block multiple times + * when we write to the same offset and new + * ensures that we do proper zero out for + * partial write. + */ + set_buffer_new(bh_result); + set_buffer_mapped(bh_result); + } ret = 0; } return ret; } -#define EXT4_DELALLOC_RSVED 1 -static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, + +/* + * This function is used as a standard get_block_t calback function + * when there is no desire to allocate any blocks. It is used as a + * callback function for block_prepare_write(), nobh_writepage(), and + * block_write_full_page(). These functions should only try to map a + * single block at a time. + * + * Since this function doesn't do block allocations even if the caller + * requests it by passing in create=1, it is critically important that + * any caller checks to make sure that any buffer heads are returned + * by this function are either all already mapped or marked for + * delayed allocation before calling nobh_writepage() or + * block_write_full_page(). Otherwise, b_blocknr could be left + * unitialized, and the page write functions will be taken by + * surprise. + */ +static int noalloc_get_block_write(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { - int ret; + int ret = 0; unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; - loff_t disksize = EXT4_I(inode)->i_disksize; - handle_t *handle = NULL; - handle = ext4_journal_current_handle(); - if (!handle) { - ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, - bh_result, 0, 0, 0); - BUG_ON(!ret); - } else { - ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, - bh_result, create, 0, EXT4_DELALLOC_RSVED); - } + BUG_ON(bh_result->b_size != inode->i_sb->s_blocksize); + /* + * we don't want to do block allocation in writepage + * so call get_block_wrap with create = 0 + */ + ret = ext4_get_blocks(NULL, inode, iblock, max_blocks, bh_result, 0); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); - - /* - * Update on-disk size along with block allocation - * we don't use 'extend_disksize' as size may change - * within already allocated block -bzzz - */ - disksize = ((loff_t) iblock + ret) << inode->i_blkbits; - if (disksize > i_size_read(inode)) - disksize = i_size_read(inode); - if (disksize > EXT4_I(inode)->i_disksize) { - /* - * XXX: replace with spinlock if seen contended -bzzz - */ - down_write(&EXT4_I(inode)->i_data_sem); - if (disksize > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = disksize; - up_write(&EXT4_I(inode)->i_data_sem); - - if (EXT4_I(inode)->i_disksize == disksize) { - ret = ext4_mark_inode_dirty(handle, inode); - return ret; - } - } ret = 0; } return ret; } -static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) +static int bget_one(handle_t *handle, struct buffer_head *bh) { - /* - * unmapped buffer is possible for holes. - * delay buffer is possible with delayed allocation - */ - return ((!buffer_mapped(bh) || buffer_delay(bh)) && buffer_dirty(bh)); + get_bh(bh); + return 0; } -static int ext4_normal_get_block_write(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) +static int bput_one(handle_t *handle, struct buffer_head *bh) +{ + put_bh(bh); + return 0; +} + +static int __ext4_journalled_writepage(struct page *page, + struct writeback_control *wbc, + unsigned int len) { + struct address_space *mapping = page->mapping; + struct inode *inode = mapping->host; + struct buffer_head *page_bufs; + handle_t *handle = NULL; int ret = 0; - unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + int err; - /* - * we don't want to do block allocation in writepage - * so call get_block_wrap with create = 0 - */ - ret = ext4_get_blocks_wrap(NULL, inode, iblock, max_blocks, - bh_result, 0, 0, 0); - if (ret > 0) { - bh_result->b_size = (ret << inode->i_blkbits); - ret = 0; + page_bufs = page_buffers(page); + BUG_ON(!page_bufs); + walk_page_buffers(handle, page_bufs, 0, len, NULL, bget_one); + /* As soon as we unlock the page, it can go away, but we have + * references to buffers so we are safe */ + unlock_page(page); + + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; } + + ret = walk_page_buffers(handle, page_bufs, 0, len, NULL, + do_journal_get_write_access); + + err = walk_page_buffers(handle, page_bufs, 0, len, NULL, + write_end_fn); + if (ret == 0) + ret = err; + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + + walk_page_buffers(handle, page_bufs, 0, len, NULL, bput_one); + EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; +out: return ret; } /* - * get called vi ext4_da_writepages after taking page lock (have journal handle) - * get called via journal_submit_inode_data_buffers (no journal handle) - * get called via shrink_page_list via pdflush (no journal handle) - * or grab_page_cache when doing write_begin (have journal handle) + * Note that we don't need to start a transaction unless we're journaling data + * because we should have holes filled from ext4_page_mkwrite(). We even don't + * need to file the inode to the transaction's list in ordered mode because if + * we are writing back data added by write(), the inode is already there and if + * we are writing back data modified via mmap(), noone guarantees in which + * transaction the data will hit the disk. In case we are journaling data, we + * cannot start transaction directly because transaction start ranks above page + * lock so we have to do some magic. + * + * This function can get called via... + * - ext4_da_writepages after taking page lock (have journal handle) + * - journal_submit_inode_data_buffers (no journal handle) + * - shrink_page_list via pdflush (no journal handle) + * - grab_page_cache when doing write_begin (have journal handle) + * + * We don't do any block allocation in this function. If we have page with + * multiple blocks we need to write those buffer_heads that are mapped. This + * is important for mmaped based write. So if we do with blocksize 1K + * truncate(f, 1024); + * a = mmap(f, 0, 4096); + * a[0] = 'a'; + * truncate(f, 4096); + * we have in the page first buffer_head mapped via page_mkwrite call back + * but other bufer_heads would be unmapped but dirty(dirty done via the + * do_wp_page). So writepage should write the first block. If we modify + * the mmap area beyond 1024 we will again get a page_fault and the + * page_mkwrite callback will do the block allocation and mark the + * buffer_heads mapped. + * + * We redirty the page if we have any buffer_heads that is either delay or + * unwritten in the page. + * + * We can get recursively called as show below. + * + * ext4_writepage() -> kmalloc() -> __alloc_pages() -> page_launder() -> + * ext4_writepage() + * + * But since we don't do any block allocation we should not deadlock. + * Page also have the dirty flag cleared so we don't get recurive page_lock. */ -static int ext4_da_writepage(struct page *page, - struct writeback_control *wbc) +static int ext4_writepage(struct page *page, + struct writeback_control *wbc) { int ret = 0; loff_t size; - unsigned long len; + unsigned int len; struct buffer_head *page_bufs; struct inode *inode = page->mapping->host; + trace_ext4_writepage(inode, page); size = i_size_read(inode); if (page->index == size >> PAGE_CACHE_SHIFT) len = size & ~PAGE_CACHE_MASK; @@ -2147,7 +2699,7 @@ static int ext4_da_writepage(struct page *page, if (page_has_buffers(page)) { page_bufs = page_buffers(page); if (walk_page_buffers(NULL, page_bufs, 0, len, NULL, - ext4_bh_unmapped_or_delay)) { + ext4_bh_delay_or_unwritten)) { /* * We don't want to do block allocation * So redirty the page and return @@ -2174,13 +2726,13 @@ static int ext4_da_writepage(struct page *page, * all are mapped and non delay. We don't want to * do block allocation here. */ - ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, - ext4_normal_get_block_write); + ret = block_prepare_write(page, 0, len, + noalloc_get_block_write); if (!ret) { page_bufs = page_buffers(page); /* check whether all are mapped and non delay */ if (walk_page_buffers(NULL, page_bufs, 0, len, NULL, - ext4_bh_unmapped_or_delay)) { + ext4_bh_delay_or_unwritten)) { redirty_page_for_writepage(wbc, page); unlock_page(page); return 0; @@ -2195,118 +2747,294 @@ static int ext4_da_writepage(struct page *page, unlock_page(page); return 0; } + /* now mark the buffer_heads as dirty and uptodate */ + block_commit_write(page, 0, len); + } + + if (PageChecked(page) && ext4_should_journal_data(inode)) { + /* + * It's mmapped pagecache. Add buffers and journal it. There + * doesn't seem much point in redirtying the page here. + */ + ClearPageChecked(page); + return __ext4_journalled_writepage(page, wbc, len); } if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) - ret = nobh_writepage(page, ext4_normal_get_block_write, wbc); + ret = nobh_writepage(page, noalloc_get_block_write, wbc); else - ret = block_write_full_page(page, - ext4_normal_get_block_write, - wbc); + ret = block_write_full_page(page, noalloc_get_block_write, + wbc); return ret; } /* - * For now just follow the DIO way to estimate the max credits - * needed to write out EXT4_MAX_WRITEBACK_PAGES. - * todo: need to calculate the max credits need for - * extent based files, currently the DIO credits is based on - * indirect-blocks mapping way. - * - * Probably should have a generic way to calculate credits - * for DIO, writepages, and truncate + * This is called via ext4_da_writepages() to + * calulate the total number of credits to reserve to fit + * a single extent allocation into a single transaction, + * ext4_da_writpeages() will loop calling this before + * the block allocation. */ -#define EXT4_MAX_WRITEBACK_PAGES DIO_MAX_BLOCKS -#define EXT4_MAX_WRITEBACK_CREDITS DIO_CREDITS + +static int ext4_da_writepages_trans_blocks(struct inode *inode) +{ + int max_blocks = EXT4_I(inode)->i_reserved_data_blocks; + + /* + * With non-extent format the journal credit needed to + * insert nrblocks contiguous block is dependent on + * number of contiguous block. So we will limit + * number of contiguous block to a sane value + */ + if (!(inode->i_flags & EXT4_EXTENTS_FL) && + (max_blocks > EXT4_MAX_TRANS_DATA)) + max_blocks = EXT4_MAX_TRANS_DATA; + + return ext4_chunk_trans_blocks(inode, max_blocks); +} static int ext4_da_writepages(struct address_space *mapping, - struct writeback_control *wbc) + struct writeback_control *wbc) { - struct inode *inode = mapping->host; + pgoff_t index; + int range_whole = 0; handle_t *handle = NULL; - int needed_blocks; - int ret = 0; - long to_write; - loff_t range_start = 0; + struct mpage_da_data mpd; + struct inode *inode = mapping->host; + int no_nrwrite_index_update; + int pages_written = 0; + long pages_skipped; + unsigned int max_pages; + int range_cyclic, cycled = 1, io_done = 0; + int needed_blocks, ret = 0; + long desired_nr_to_write, nr_to_writebump = 0; + loff_t range_start = wbc->range_start; + struct ext4_sb_info *sbi = EXT4_SB(mapping->host->i_sb); + + trace_ext4_da_writepages(inode, wbc); /* * No pages to write? This is mainly a kludge to avoid starting * a transaction for special inodes like journal inode on last iput() * because that could violate lock ordering on umount */ - if (!mapping->nrpages) + if (!mapping->nrpages || !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) return 0; /* - * Estimate the worse case needed credits to write out - * EXT4_MAX_BUF_BLOCKS pages + * If the filesystem has aborted, it is read-only, so return + * right away instead of dumping stack traces later on that + * will obscure the real source of the problem. We test + * EXT4_MF_FS_ABORTED instead of sb->s_flag's MS_RDONLY because + * the latter could be true if the filesystem is mounted + * read-only, and in that case, ext4_da_writepages should + * *never* be called, so if that ever happens, we would want + * the stack trace. + */ + if (unlikely(sbi->s_mount_flags & EXT4_MF_FS_ABORTED)) + return -EROFS; + + if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) + range_whole = 1; + + range_cyclic = wbc->range_cyclic; + if (wbc->range_cyclic) { + index = mapping->writeback_index; + if (index) + cycled = 0; + wbc->range_start = index << PAGE_CACHE_SHIFT; + wbc->range_end = LLONG_MAX; + wbc->range_cyclic = 0; + } else + index = wbc->range_start >> PAGE_CACHE_SHIFT; + + /* + * This works around two forms of stupidity. The first is in + * the writeback code, which caps the maximum number of pages + * written to be 1024 pages. This is wrong on multiple + * levels; different architectues have a different page size, + * which changes the maximum amount of data which gets + * written. Secondly, 4 megabytes is way too small. XFS + * forces this value to be 16 megabytes by multiplying + * nr_to_write parameter by four, and then relies on its + * allocator to allocate larger extents to make them + * contiguous. Unfortunately this brings us to the second + * stupidity, which is that ext4's mballoc code only allocates + * at most 2048 blocks. So we force contiguous writes up to + * the number of dirty blocks in the inode, or + * sbi->max_writeback_mb_bump whichever is smaller. + */ + max_pages = sbi->s_max_writeback_mb_bump << (20 - PAGE_CACHE_SHIFT); + if (!range_cyclic && range_whole) + desired_nr_to_write = wbc->nr_to_write * 8; + else + desired_nr_to_write = ext4_num_dirty_pages(inode, index, + max_pages); + if (desired_nr_to_write > max_pages) + desired_nr_to_write = max_pages; + + if (wbc->nr_to_write < desired_nr_to_write) { + nr_to_writebump = desired_nr_to_write - wbc->nr_to_write; + wbc->nr_to_write = desired_nr_to_write; + } + + mpd.wbc = wbc; + mpd.inode = mapping->host; + + /* + * we don't want write_cache_pages to update + * nr_to_write and writeback_index */ - needed_blocks = EXT4_MAX_WRITEBACK_CREDITS; + no_nrwrite_index_update = wbc->no_nrwrite_index_update; + wbc->no_nrwrite_index_update = 1; + pages_skipped = wbc->pages_skipped; + +retry: + while (!ret && wbc->nr_to_write > 0) { - to_write = wbc->nr_to_write; - if (!wbc->range_cyclic) { /* - * If range_cyclic is not set force range_cont - * and save the old writeback_index + * we insert one extent at a time. So we need + * credit needed for single extent allocation. + * journalled mode is currently not supported + * by delalloc */ - wbc->range_cont = 1; - range_start = wbc->range_start; - } + BUG_ON(ext4_should_journal_data(inode)); + needed_blocks = ext4_da_writepages_trans_blocks(inode); - while (!ret && to_write) { /* start a new transaction*/ handle = ext4_journal_start(inode, needed_blocks); if (IS_ERR(handle)) { ret = PTR_ERR(handle); + ext4_msg(inode->i_sb, KERN_CRIT, "%s: jbd2_start: " + "%ld pages, ino %lu; err %d\n", __func__, + wbc->nr_to_write, inode->i_ino, ret); goto out_writepages; } - if (ext4_should_order_data(inode)) { - /* - * With ordered mode we need to add - * the inode to the journal handle - * when we do block allocation. - */ - ret = ext4_jbd2_file_inode(handle, inode); - if (ret) { - ext4_journal_stop(handle); - goto out_writepages; - } - } /* - * set the max dirty pages could be write at a time - * to fit into the reserved transaction credits + * Now call __mpage_da_writepage to find the next + * contiguous region of logical blocks that need + * blocks to be allocated by ext4. We don't actually + * submit the blocks for I/O here, even though + * write_cache_pages thinks it will, and will set the + * pages as clean for write before calling + * __mpage_da_writepage(). */ - if (wbc->nr_to_write > EXT4_MAX_WRITEBACK_PAGES) - wbc->nr_to_write = EXT4_MAX_WRITEBACK_PAGES; - - to_write -= wbc->nr_to_write; - ret = mpage_da_writepages(mapping, wbc, - ext4_da_get_block_write); - ext4_journal_stop(handle); - if (wbc->nr_to_write) { + mpd.b_size = 0; + mpd.b_state = 0; + mpd.b_blocknr = 0; + mpd.first_page = 0; + mpd.next_page = 0; + mpd.io_done = 0; + mpd.pages_written = 0; + mpd.retval = 0; + ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, + &mpd); + /* + * If we have a contiguous extent of pages and we + * haven't done the I/O yet, map the blocks and submit + * them for I/O. + */ + if (!mpd.io_done && mpd.next_page != mpd.first_page) { + if (mpage_da_map_blocks(&mpd) == 0) + mpage_da_submit_io(&mpd); + mpd.io_done = 1; + ret = MPAGE_DA_EXTENT_TAIL; + } + trace_ext4_da_write_pages(inode, &mpd); + wbc->nr_to_write -= mpd.pages_written; + + ext4_journal_stop(handle); + + if ((mpd.retval == -ENOSPC) && sbi->s_journal) { + /* commit the transaction which would + * free blocks released in the transaction + * and try again + */ + jbd2_journal_force_commit_nested(sbi->s_journal); + wbc->pages_skipped = pages_skipped; + ret = 0; + } else if (ret == MPAGE_DA_EXTENT_TAIL) { + /* + * got one extent now try with + * rest of the pages + */ + pages_written += mpd.pages_written; + wbc->pages_skipped = pages_skipped; + ret = 0; + io_done = 1; + } else if (wbc->nr_to_write) /* * There is no more writeout needed * or we requested for a noblocking writeout * and we found the device congested */ - to_write += wbc->nr_to_write; break; - } - wbc->nr_to_write = to_write; } + if (!io_done && !cycled) { + cycled = 1; + index = 0; + wbc->range_start = index << PAGE_CACHE_SHIFT; + wbc->range_end = mapping->writeback_index - 1; + goto retry; + } + if (pages_skipped != wbc->pages_skipped) + ext4_msg(inode->i_sb, KERN_CRIT, + "This should not happen leaving %s " + "with nr_to_write = %ld ret = %d\n", + __func__, wbc->nr_to_write, ret); + + /* Update index */ + index += pages_written; + wbc->range_cyclic = range_cyclic; + if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0)) + /* + * set the writeback_index so that range_cyclic + * mode will write it back later + */ + mapping->writeback_index = index; out_writepages: - wbc->nr_to_write = to_write; - if (range_start) - wbc->range_start = range_start; + if (!no_nrwrite_index_update) + wbc->no_nrwrite_index_update = 0; + if (wbc->nr_to_write > nr_to_writebump) + wbc->nr_to_write -= nr_to_writebump; + wbc->range_start = range_start; + trace_ext4_da_writepages_result(inode, wbc, ret, pages_written); return ret; } +#define FALL_BACK_TO_NONDELALLOC 1 +static int ext4_nonda_switch(struct super_block *sb) +{ + s64 free_blocks, dirty_blocks; + struct ext4_sb_info *sbi = EXT4_SB(sb); + + /* + * switch to non delalloc mode if we are running low + * on free block. The free block accounting via percpu + * counters can get slightly wrong with percpu_counter_batch getting + * accumulated on each CPU without updating global counters + * Delalloc need an accurate free block accounting. So switch + * to non delalloc when we are near to error range. + */ + free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + dirty_blocks = percpu_counter_read_positive(&sbi->s_dirtyblocks_counter); + if (2 * free_blocks < 3 * dirty_blocks || + free_blocks < (dirty_blocks + EXT4_FREEBLOCKS_WATERMARK)) { + /* + * free block count is less that 150% of dirty blocks + * or free blocks is less that watermark + */ + return 1; + } + return 0; +} + static int ext4_da_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned flags, - struct page **pagep, void **fsdata) + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) { int ret, retries = 0; struct page *page; @@ -2319,6 +3047,13 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; + if (ext4_nonda_switch(inode->i_sb)) { + *fsdata = (void *)FALL_BACK_TO_NONDELALLOC; + return ext4_write_begin(file, mapping, pos, + len, flags, pagep, fsdata); + } + *fsdata = (void *)0; + trace_ext4_da_write_begin(inode, pos, len, flags); retry: /* * With delayed allocation, we don't log the i_disksize update @@ -2331,8 +3066,11 @@ retry: ret = PTR_ERR(handle); goto out; } + /* We cannot recurse into the filesystem as the transaction is already + * started */ + flags |= AOP_FLAG_NOFS; - page = __grab_cache_page(mapping, index); + page = grab_cache_page_write_begin(mapping, index, flags); if (!page) { ext4_journal_stop(handle); ret = -ENOMEM; @@ -2341,11 +3079,18 @@ retry: *pagep = page; ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, - ext4_da_get_block_prep); + ext4_da_get_block_prep); if (ret < 0) { unlock_page(page); ext4_journal_stop(handle); page_cache_release(page); + /* + * block_write_begin may have instantiated a few blocks + * outside i_size. Trim these off again. Don't need + * i_size_read because we hold i_mutex. + */ + if (pos + len > inode->i_size) + ext4_truncate(inode); } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -2359,7 +3104,7 @@ out: * when write to the end of file but not require block allocation */ static int ext4_da_should_update_i_disksize(struct page *page, - unsigned long offset) + unsigned long offset) { struct buffer_head *bh; struct inode *inode = page->mapping->host; @@ -2369,27 +3114,41 @@ static int ext4_da_should_update_i_disksize(struct page *page, bh = page_buffers(page); idx = offset >> inode->i_blkbits; - for (i=0; i < idx; i++) + for (i = 0; i < idx; i++) bh = bh->b_this_page; - if (!buffer_mapped(bh) || (buffer_delay(bh))) + if (!buffer_mapped(bh) || (buffer_delay(bh)) || buffer_unwritten(bh)) return 0; return 1; } static int ext4_da_write_end(struct file *file, - struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) { struct inode *inode = mapping->host; int ret = 0, ret2; handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; + int write_mode = (int)(unsigned long)fsdata; + + if (write_mode == FALL_BACK_TO_NONDELALLOC) { + if (ext4_should_order_data(inode)) { + return ext4_ordered_write_end(file, mapping, pos, + len, copied, page, fsdata); + } else if (ext4_should_writeback_data(inode)) { + return ext4_writeback_write_end(file, mapping, pos, + len, copied, page, fsdata); + } else { + BUG(); + } + } + trace_ext4_da_write_end(inode, pos, len, copied); start = pos & (PAGE_CACHE_SIZE - 1); - end = start + copied -1; + end = start + copied - 1; /* * generic_write_end() will run mark_inode_dirty() if i_size @@ -2413,6 +3172,11 @@ static int ext4_da_write_end(struct file *file, EXT4_I(inode)->i_disksize = new_i_size; } up_write(&EXT4_I(inode)->i_data_sem); + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_mark_inode_dirty(handle, inode); } } ret2 = generic_write_end(file, mapping, pos, len, copied, @@ -2444,6 +3208,50 @@ out: return; } +/* + * Force all delayed allocation blocks to be allocated for a given inode. + */ +int ext4_alloc_da_blocks(struct inode *inode) +{ + trace_ext4_alloc_da_blocks(inode); + + if (!EXT4_I(inode)->i_reserved_data_blocks && + !EXT4_I(inode)->i_reserved_meta_blocks) + return 0; + + /* + * We do something simple for now. The filemap_flush() will + * also start triggering a write of the data blocks, which is + * not strictly speaking necessary (and for users of + * laptop_mode, not even desirable). However, to do otherwise + * would require replicating code paths in: + * + * ext4_da_writepages() -> + * write_cache_pages() ---> (via passed in callback function) + * __mpage_da_writepage() --> + * mpage_add_bh_to_extent() + * mpage_da_map_blocks() + * + * The problem is that write_cache_pages(), located in + * mm/page-writeback.c, marks pages clean in preparation for + * doing I/O, which is not desirable if we're not planning on + * doing I/O at all. + * + * We could call write_cache_pages(), and then redirty all of + * the pages by calling redirty_page_for_writeback() but that + * would be ugly in the extreme. So instead we would need to + * replicate parts of the code in the above functions, + * simplifying them becuase we wouldn't actually intend to + * write out the pages, but rather only collect contiguous + * logical block extents, call the multi-block allocator, and + * then update the buffer heads with the block allocations. + * + * For now, though, we'll cheat by calling filemap_flush(), + * which will map the blocks, and start the I/O, but not + * actually wait for the I/O to complete. + */ + return filemap_flush(inode->i_mapping); +} /* * bmap() is special. It gets used by applications such as lilo and by @@ -2475,7 +3283,7 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) filemap_write_and_wait(mapping); } - if (EXT4_I(inode)->i_state & EXT4_STATE_JDATA) { + if (EXT4_JOURNAL(inode) && EXT4_I(inode)->i_state & EXT4_STATE_JDATA) { /* * This is a REALLY heavyweight approach, but the use of * bmap on dirty files is expected to be extremely rare: @@ -2504,342 +3312,490 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) return 0; } - return generic_block_bmap(mapping,block,ext4_get_block); + return generic_block_bmap(mapping, block, ext4_get_block); } -static int bget_one(handle_t *handle, struct buffer_head *bh) +static int ext4_readpage(struct file *file, struct page *page) { - get_bh(bh); - return 0; + return mpage_readpage(page, ext4_get_block); } -static int bput_one(handle_t *handle, struct buffer_head *bh) +static int +ext4_readpages(struct file *file, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages) { - put_bh(bh); - return 0; + return mpage_readpages(mapping, pages, nr_pages, ext4_get_block); } -/* - * Note that we don't need to start a transaction unless we're journaling data - * because we should have holes filled from ext4_page_mkwrite(). We even don't - * need to file the inode to the transaction's list in ordered mode because if - * we are writing back data added by write(), the inode is already there and if - * we are writing back data modified via mmap(), noone guarantees in which - * transaction the data will hit the disk. In case we are journaling data, we - * cannot start transaction directly because transaction start ranks above page - * lock so we have to do some magic. - * - * In all journaling modes block_write_full_page() will start the I/O. - * - * Problem: - * - * ext4_writepage() -> kmalloc() -> __alloc_pages() -> page_launder() -> - * ext4_writepage() - * - * Similar for: - * - * ext4_file_write() -> generic_file_write() -> __alloc_pages() -> ... - * - * Same applies to ext4_get_block(). We will deadlock on various things like - * lock_journal and i_data_sem - * - * Setting PF_MEMALLOC here doesn't work - too many internal memory - * allocations fail. - * - * 16May01: If we're reentered then journal_current_handle() will be - * non-zero. We simply *return*. - * - * 1 July 2001: @@@ FIXME: - * In journalled data mode, a data buffer may be metadata against the - * current transaction. But the same file is part of a shared mapping - * and someone does a writepage() on it. - * - * We will move the buffer onto the async_data list, but *after* it has - * been dirtied. So there's a small window where we have dirty data on - * BJ_Metadata. - * - * Note that this only applies to the last partial page in the file. The - * bit which block_write_full_page() uses prepare/commit for. (That's - * broken code anyway: it's wrong for msync()). - * - * It's a rare case: affects the final partial page, for journalled data - * where the file is subject to bith write() and writepage() in the same - * transction. To fix it we'll need a custom block_write_full_page(). - * We'll probably need that anyway for journalling writepage() output. - * - * We don't honour synchronous mounts for writepage(). That would be - * disastrous. Any write() or metadata operation will sync the fs for - * us. - * - */ -static int __ext4_normal_writepage(struct page *page, - struct writeback_control *wbc) +static void ext4_invalidatepage(struct page *page, unsigned long offset) { - struct inode *inode = page->mapping->host; + journal_t *journal = EXT4_JOURNAL(page->mapping->host); - if (test_opt(inode->i_sb, NOBH)) - return nobh_writepage(page, - ext4_normal_get_block_write, wbc); + /* + * If it's a full truncate we just forget about the pending dirtying + */ + if (offset == 0) + ClearPageChecked(page); + + if (journal) + jbd2_journal_invalidatepage(journal, page, offset); else - return block_write_full_page(page, - ext4_normal_get_block_write, - wbc); + block_invalidatepage(page, offset); } -static int ext4_normal_writepage(struct page *page, - struct writeback_control *wbc) +static int ext4_releasepage(struct page *page, gfp_t wait) { - struct inode *inode = page->mapping->host; - loff_t size = i_size_read(inode); - loff_t len; + journal_t *journal = EXT4_JOURNAL(page->mapping->host); - J_ASSERT(PageLocked(page)); - if (page->index == size >> PAGE_CACHE_SHIFT) - len = size & ~PAGE_CACHE_MASK; + WARN_ON(PageChecked(page)); + if (!page_has_buffers(page)) + return 0; + if (journal) + return jbd2_journal_try_to_free_buffers(journal, page, wait); else - len = PAGE_CACHE_SIZE; + return try_to_free_buffers(page); +} - if (page_has_buffers(page)) { - /* if page has buffers it should all be mapped - * and allocated. If there are not buffers attached - * to the page we know the page is dirty but it lost - * buffers. That means that at some moment in time - * after write_begin() / write_end() has been called - * all buffers have been clean and thus they must have been - * written at least once. So they are all mapped and we can - * happily proceed with mapping them and writing the page. - */ - BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, - ext4_bh_unmapped_or_delay)); +/* + * 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. + */ +static 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, 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); + } } - if (!ext4_journal_current_handle()) - return __ext4_normal_writepage(page, wbc); +retry: + ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, + offset, nr_segs, + ext4_get_block, NULL); + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + + if (orphan) { + int err; - redirty_page_for_writepage(wbc, page); - unlock_page(page); - return 0; + /* Credits for sb + inode write */ + handle = ext4_journal_start(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); + 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; } -static int __ext4_journalled_writepage(struct page *page, - struct writeback_control *wbc) +/* Maximum number of blocks we map for direct IO at once. */ + +static int ext4_get_block_dio_write(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) { - struct address_space *mapping = page->mapping; - struct inode *inode = mapping->host; - struct buffer_head *page_bufs; handle_t *handle = NULL; int ret = 0; - int err; - - ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, - ext4_normal_get_block_write); - if (ret != 0) - goto out_unlock; + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + int dio_credits; - page_bufs = page_buffers(page); - walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, - bget_one); - /* As soon as we unlock the page, it can go away, but we have - * references to buffers so we are safe */ - unlock_page(page); + ext4_debug("ext4_get_block_dio_write: inode %lu, create flag %d\n", + inode->i_ino, create); + /* + * DIO VFS code passes create = 0 flag for write to + * the middle of file. It does this to avoid block + * allocation for holes, to prevent expose stale data + * out when there is parallel buffered read (which does + * not hold the i_mutex lock) while direct IO write has + * not completed. DIO request on holes finally falls back + * to buffered IO for this reason. + * + * For ext4 extent based file, since we support fallocate, + * new allocated extent as uninitialized, for holes, we + * could fallocate blocks for holes, thus parallel + * buffered IO read will zero out the page when read on + * a hole while parallel DIO write to the hole has not completed. + * + * when we come here, we know it's a direct IO write to + * to the middle of file ( DIO_MAX_BLOCKS) + max_blocks = DIO_MAX_BLOCKS; + dio_credits = ext4_chunk_trans_blocks(inode, max_blocks); + handle = ext4_journal_start(inode, dio_credits); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; } - - ret = walk_page_buffers(handle, page_bufs, 0, - PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); - - err = walk_page_buffers(handle, page_bufs, 0, - PAGE_CACHE_SIZE, NULL, write_end_fn); - if (ret == 0) - ret = err; - err = ext4_journal_stop(handle); - if (!ret) - ret = err; - - walk_page_buffers(handle, page_bufs, 0, - PAGE_CACHE_SIZE, NULL, bput_one); - EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; - goto out; - -out_unlock: - unlock_page(page); + ret = ext4_get_blocks(handle, inode, iblock, max_blocks, bh_result, + create); + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; + } + ext4_journal_stop(handle); out: return ret; } -static int ext4_journalled_writepage(struct page *page, - struct writeback_control *wbc) +static void ext4_free_io_end(ext4_io_end_t *io) { - struct inode *inode = page->mapping->host; - loff_t size = i_size_read(inode); - loff_t len; - - J_ASSERT(PageLocked(page)); - if (page->index == size >> PAGE_CACHE_SHIFT) - len = size & ~PAGE_CACHE_MASK; - else - len = PAGE_CACHE_SIZE; + BUG_ON(!io); + iput(io->inode); + kfree(io); +} +static void dump_aio_dio_list(struct inode * inode) +{ +#ifdef EXT4_DEBUG + struct list_head *cur, *before, *after; + ext4_io_end_t *io, *io0, *io1; - if (page_has_buffers(page)) { - /* if page has buffers it should all be mapped - * and allocated. If there are not buffers attached - * to the page we know the page is dirty but it lost - * buffers. That means that at some moment in time - * after write_begin() / write_end() has been called - * all buffers have been clean and thus they must have been - * written at least once. So they are all mapped and we can - * happily proceed with mapping them and writing the page. - */ - BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, - ext4_bh_unmapped_or_delay)); + if (list_empty(&EXT4_I(inode)->i_aio_dio_complete_list)){ + ext4_debug("inode %lu aio dio list is empty\n", inode->i_ino); + return; } - if (ext4_journal_current_handle()) - goto no_write; + ext4_debug("Dump inode %lu aio_dio_completed_IO list \n", inode->i_ino); + list_for_each_entry(io, &EXT4_I(inode)->i_aio_dio_complete_list, list){ + cur = &io->list; + before = cur->prev; + io0 = container_of(before, ext4_io_end_t, list); + after = cur->next; + io1 = container_of(after, ext4_io_end_t, list); - if (PageChecked(page)) { - /* - * It's mmapped pagecache. Add buffers and journal it. There - * doesn't seem much point in redirtying the page here. - */ - ClearPageChecked(page); - return __ext4_journalled_writepage(page, wbc); - } else { - /* - * It may be a page full of checkpoint-mode buffers. We don't - * really know unless we go poke around in the buffer_heads. - * But block_write_full_page will do the right thing. - */ - return block_write_full_page(page, - ext4_normal_get_block_write, - wbc); + ext4_debug("io 0x%p from inode %lu,prev 0x%p,next 0x%p\n", + io, inode->i_ino, io0, io1); } -no_write: - redirty_page_for_writepage(wbc, page); - unlock_page(page); - return 0; +#endif } -static int ext4_readpage(struct file *file, struct page *page) +/* + * check a range of space and convert unwritten extents to written. + */ +static int ext4_end_aio_dio_nolock(ext4_io_end_t *io) { - return mpage_readpage(page, ext4_get_block); + struct inode *inode = io->inode; + loff_t offset = io->offset; + size_t size = io->size; + int ret = 0; + + ext4_debug("end_aio_dio_onlock: io 0x%p from inode %lu,list->next 0x%p," + "list->prev 0x%p\n", + io, inode->i_ino, io->list.next, io->list.prev); + + if (list_empty(&io->list)) + return ret; + + if (io->flag != DIO_AIO_UNWRITTEN) + return ret; + + if (offset + size <= i_size_read(inode)) + ret = ext4_convert_unwritten_extents(inode, offset, size); + + if (ret < 0) { + printk(KERN_EMERG "%s: failed to convert unwritten" + "extents to written extents, error is %d" + " io is still on inode %lu aio dio list\n", + __func__, ret, inode->i_ino); + return ret; + } + + /* clear the DIO AIO unwritten flag */ + io->flag = 0; + return ret; } +/* + * work on completed aio dio IO, to convert unwritten extents to extents + */ +static void ext4_end_aio_dio_work(struct work_struct *work) +{ + ext4_io_end_t *io = container_of(work, ext4_io_end_t, work); + struct inode *inode = io->inode; + int ret = 0; -static int -ext4_readpages(struct file *file, struct address_space *mapping, - struct list_head *pages, unsigned nr_pages) + mutex_lock(&inode->i_mutex); + ret = ext4_end_aio_dio_nolock(io); + if (ret >= 0) { + if (!list_empty(&io->list)) + list_del_init(&io->list); + ext4_free_io_end(io); + } + mutex_unlock(&inode->i_mutex); +} +/* + * This function is called from ext4_sync_file(). + * + * When AIO DIO IO is completed, the work to convert unwritten + * extents to written is queued on workqueue but may not get immediately + * scheduled. When fsync is called, we need to ensure the + * conversion is complete before fsync returns. + * The inode keeps track of a list of completed AIO from DIO path + * that might needs to do the conversion. This function walks through + * the list and convert the related unwritten extents to written. + */ +int flush_aio_dio_completed_IO(struct inode *inode) { - return mpage_readpages(mapping, pages, nr_pages, ext4_get_block); + ext4_io_end_t *io; + int ret = 0; + int ret2 = 0; + + if (list_empty(&EXT4_I(inode)->i_aio_dio_complete_list)) + return ret; + + dump_aio_dio_list(inode); + while (!list_empty(&EXT4_I(inode)->i_aio_dio_complete_list)){ + io = list_entry(EXT4_I(inode)->i_aio_dio_complete_list.next, + ext4_io_end_t, list); + /* + * Calling ext4_end_aio_dio_nolock() to convert completed + * IO to written. + * + * When ext4_sync_file() is called, run_queue() may already + * about to flush the work corresponding to this io structure. + * It will be upset if it founds the io structure related + * to the work-to-be schedule is freed. + * + * Thus we need to keep the io structure still valid here after + * convertion finished. The io structure has a flag to + * avoid double converting from both fsync and background work + * queue work. + */ + ret = ext4_end_aio_dio_nolock(io); + if (ret < 0) + ret2 = ret; + else + list_del_init(&io->list); + } + return (ret2 < 0) ? ret2 : 0; } -static void ext4_invalidatepage(struct page *page, unsigned long offset) +static ext4_io_end_t *ext4_init_io_end (struct inode *inode) { - journal_t *journal = EXT4_JOURNAL(page->mapping->host); - - /* - * If it's a full truncate we just forget about the pending dirtying - */ - if (offset == 0) - ClearPageChecked(page); + ext4_io_end_t *io = NULL; + + io = kmalloc(sizeof(*io), GFP_NOFS); + + if (io) { + igrab(inode); + io->inode = inode; + io->flag = 0; + io->offset = 0; + io->size = 0; + io->error = 0; + INIT_WORK(&io->work, ext4_end_aio_dio_work); + INIT_LIST_HEAD(&io->list); + } - jbd2_journal_invalidatepage(journal, page, offset); + return io; } -static int ext4_releasepage(struct page *page, gfp_t wait) +static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, + ssize_t size, void *private) { - journal_t *journal = EXT4_JOURNAL(page->mapping->host); + ext4_io_end_t *io_end = iocb->private; + struct workqueue_struct *wq; + + ext_debug("ext4_end_io_dio(): io_end 0x%p" + "for inode %lu, iocb 0x%p, offset %llu, size %llu\n", + iocb->private, io_end->inode->i_ino, iocb, offset, + size); + /* if not async direct IO or dio with 0 bytes write, just return */ + if (!io_end || !size) + return; - WARN_ON(PageChecked(page)); - if (!page_has_buffers(page)) - return 0; - return jbd2_journal_try_to_free_buffers(journal, page, wait); -} + /* if not aio dio with unwritten extents, just free io and return */ + if (io_end->flag != DIO_AIO_UNWRITTEN){ + ext4_free_io_end(io_end); + iocb->private = NULL; + return; + } + io_end->offset = offset; + io_end->size = size; + wq = EXT4_SB(io_end->inode->i_sb)->dio_unwritten_wq; + + /* queue the work to convert unwritten extents to written */ + queue_work(wq, &io_end->work); + + /* Add the io_end to per-inode completed aio dio list*/ + list_add_tail(&io_end->list, + &EXT4_I(io_end->inode)->i_aio_dio_complete_list); + iocb->private = NULL; +} /* + * For ext4 extent files, ext4 will do direct-io write to holes, + * preallocated extents, and those write extend the file, no need to + * fall back to buffered IO. + * + * For holes, we fallocate those blocks, mark them as unintialized + * If those blocks were preallocated, we mark sure they are splited, but + * still keep the range to write as unintialized. + * + * The unwrritten extents will be converted to written when DIO is completed. + * For async direct IO, since the IO may still pending when return, we + * set up an end_io call back function, which will do the convertion + * when async direct IO completed. + * * 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. */ -static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, - const struct iovec *iov, loff_t offset, - unsigned long nr_segs) +static ssize_t ext4_ext_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); - 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, 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); + loff_t final_size = offset + count; + if (rw == WRITE && final_size <= inode->i_size) { + /* + * We could direct write to holes and fallocate. + * + * Allocated blocks to fill the hole are marked as uninitialized + * to prevent paralel buffered read to expose the stale data + * before DIO complete the data IO. + * + * As to previously fallocated extents, ext4 get_block + * will just simply mark the buffer mapped but still + * keep the extents uninitialized. + * + * for non AIO case, we will convert those unwritten extents + * to written after return back from blockdev_direct_IO. + * + * for async DIO, the conversion needs to be defered when + * the IO is completed. The ext4 end_io callback function + * will be called to take care of the conversion work. + * Here for async case, we allocate an io_end structure to + * hook to the iocb. + */ + iocb->private = NULL; + EXT4_I(inode)->cur_aio_dio = NULL; + if (!is_sync_kiocb(iocb)) { + iocb->private = ext4_init_io_end(inode); + if (!iocb->private) + return -ENOMEM; + /* + * we save the io structure for current async + * direct IO, so that later ext4_get_blocks() + * could flag the io structure whether there + * is a unwritten extents needs to be converted + * when IO is completed. + */ + EXT4_I(inode)->cur_aio_dio = iocb->private; } + + ret = blockdev_direct_IO(rw, iocb, inode, + inode->i_sb->s_bdev, iov, + offset, nr_segs, + ext4_get_block_dio_write, + ext4_end_io_dio); + if (iocb->private) + EXT4_I(inode)->cur_aio_dio = NULL; + /* + * The io_end structure takes a reference to the inode, + * that structure needs to be destroyed and the + * reference to the inode need to be dropped, when IO is + * complete, even with 0 byte write, or failed. + * + * In the successful AIO DIO case, the io_end structure will be + * desctroyed and the reference to the inode will be dropped + * after the end_io call back function is called. + * + * In the case there is 0 byte write, or error case, since + * VFS direct IO won't invoke the end_io call back function, + * we need to free the end_io structure here. + */ + if (ret != -EIOCBQUEUED && ret <= 0 && iocb->private) { + ext4_free_io_end(iocb->private); + iocb->private = NULL; + } else if (ret > 0) + /* + * for non AIO case, since the IO is already + * completed, we could do the convertion right here + */ + ret = ext4_convert_unwritten_extents(inode, + offset, ret); + return ret; } - ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, - offset, nr_segs, - ext4_get_block, NULL); + /* for write the the end of file case, we fall back to old way */ + return ext4_ind_direct_IO(rw, iocb, iov, offset, nr_segs); +} - if (orphan) { - int err; +static ssize_t ext4_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; - /* Credits for sb + inode write */ - handle = ext4_journal_start(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); - 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; + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_direct_IO(rw, iocb, iov, offset, nr_segs); + + return ext4_ind_direct_IO(rw, iocb, iov, offset, nr_segs); } /* @@ -2864,7 +3820,7 @@ static int ext4_journalled_set_page_dirty(struct page *page) static const struct address_space_operations ext4_ordered_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_normal_writepage, + .writepage = ext4_writepage, .sync_page = block_sync_page, .write_begin = ext4_write_begin, .write_end = ext4_ordered_write_end, @@ -2874,12 +3830,13 @@ static const struct address_space_operations ext4_ordered_aops = { .direct_IO = ext4_direct_IO, .migratepage = buffer_migrate_page, .is_partially_uptodate = block_is_partially_uptodate, + .error_remove_page = generic_error_remove_page, }; static const struct address_space_operations ext4_writeback_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_normal_writepage, + .writepage = ext4_writepage, .sync_page = block_sync_page, .write_begin = ext4_write_begin, .write_end = ext4_writeback_write_end, @@ -2889,12 +3846,13 @@ static const struct address_space_operations ext4_writeback_aops = { .direct_IO = ext4_direct_IO, .migratepage = buffer_migrate_page, .is_partially_uptodate = block_is_partially_uptodate, + .error_remove_page = generic_error_remove_page, }; static const struct address_space_operations ext4_journalled_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_journalled_writepage, + .writepage = ext4_writepage, .sync_page = block_sync_page, .write_begin = ext4_write_begin, .write_end = ext4_journalled_write_end, @@ -2903,12 +3861,13 @@ static const struct address_space_operations ext4_journalled_aops = { .invalidatepage = ext4_invalidatepage, .releasepage = ext4_releasepage, .is_partially_uptodate = block_is_partially_uptodate, + .error_remove_page = generic_error_remove_page, }; static const struct address_space_operations ext4_da_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_da_writepage, + .writepage = ext4_writepage, .writepages = ext4_da_writepages, .sync_page = block_sync_page, .write_begin = ext4_da_write_begin, @@ -2919,6 +3878,7 @@ static const struct address_space_operations ext4_da_aops = { .direct_IO = ext4_direct_IO, .migratepage = buffer_migrate_page, .is_partially_uptodate = block_is_partially_uptodate, + .error_remove_page = generic_error_remove_page, }; void ext4_set_aops(struct inode *inode) @@ -2955,7 +3915,8 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page; int err = 0; - page = grab_cache_page(mapping, from >> PAGE_CACHE_SHIFT); + page = find_or_create_page(mapping, from >> PAGE_CACHE_SHIFT, + mapping_gfp_mask(mapping) & ~__GFP_FS); if (!page) return -EINVAL; @@ -3028,7 +3989,7 @@ int ext4_block_truncate_page(handle_t *handle, err = 0; if (ext4_should_journal_data(inode)) { - err = ext4_journal_dirty_metadata(handle, bh); + err = ext4_handle_dirty_metadata(handle, inode, bh); } else { if (ext4_should_order_data(inode)) err = ext4_jbd2_file_inode(handle, inode); @@ -3090,13 +4051,14 @@ static inline int all_zeroes(__le32 *p, __le32 *q) * (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) + 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 offest + 1 */ + /* 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); @@ -3110,7 +4072,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth, 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--) + 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 @@ -3129,7 +4091,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth, } /* Writer: end */ - while(partial > p) { + while (partial > p) { brelse(partial->bh); partial--; } @@ -3146,17 +4108,20 @@ no_top: * than `count' because there can be holes in there. */ static void 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) + struct buffer_head *bh, + ext4_fsblk_t block_to_free, + unsigned long count, __le32 *first, + __le32 *last) { __le32 *p; if (try_to_extend_transaction(handle, inode)) { if (bh) { - BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); - ext4_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); + ext4_handle_dirty_metadata(handle, inode, bh); } ext4_mark_inode_dirty(handle, inode); - ext4_journal_test_restart(handle, inode); + ext4_truncate_restart_trans(handle, inode, + blocks_for_truncate(inode)); if (bh) { BUFFER_TRACE(bh, "retaking write access"); ext4_journal_get_write_access(handle, bh); @@ -3164,10 +4129,11 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode, } /* - * Any buffers which are on the journal will be in memory. We find - * them on the hash table so jbd2_journal_revoke() will run jbd2_journal_forget() - * on them. We've already detached each block from the file, so - * bforget() in jbd2_journal_forget() should be safe. + * Any buffers which are on the journal will be in memory. We + * find them on the hash table so jbd2_journal_revoke() will + * run jbd2_journal_forget() on them. We've already detached + * each block from the file, so bforget() in + * jbd2_journal_forget() should be safe. * * AKPM: turn on bforget in jbd2_journal_forget()!!! */ @@ -3253,7 +4219,7 @@ static void ext4_free_data(handle_t *handle, struct inode *inode, count, block_to_free_p, p); if (this_bh) { - BUFFER_TRACE(this_bh, "call ext4_journal_dirty_metadata"); + BUFFER_TRACE(this_bh, "call ext4_handle_dirty_metadata"); /* * The buffer head should have an attached journal head at this @@ -3261,8 +4227,8 @@ static void ext4_free_data(handle_t *handle, struct inode *inode, * block pointed to itself, it would have been detached when * the block was cleared. Check for this instead of OOPSing. */ - if (bh2jh(this_bh)) - ext4_journal_dirty_metadata(handle, this_bh); + if ((EXT4_JOURNAL(inode) == NULL) || bh2jh(this_bh)) + ext4_handle_dirty_metadata(handle, inode, this_bh); else ext4_error(inode->i_sb, __func__, "circular indirect block detected, " @@ -3292,7 +4258,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, ext4_fsblk_t nr; __le32 *p; - if (is_handle_aborted(handle)) + if (ext4_handle_is_aborted(handle)) return; if (depth--) { @@ -3321,9 +4287,9 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, /* 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); + (__le32 *) bh->b_data, + (__le32 *) bh->b_data + addr_per_block, + depth); /* * We've probably journalled the indirect block several @@ -3362,11 +4328,12 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, * will merely complain about releasing a free block, * rather than leaking blocks. */ - if (is_handle_aborted(handle)) + if (ext4_handle_is_aborted(handle)) return; if (try_to_extend_transaction(handle, inode)) { ext4_mark_inode_dirty(handle, inode); - ext4_journal_test_restart(handle, inode); + ext4_truncate_restart_trans(handle, inode, + blocks_for_truncate(inode)); } ext4_free_blocks(handle, inode, nr, 1, 1); @@ -3381,9 +4348,10 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, parent_bh)){ *p = 0; BUFFER_TRACE(parent_bh, - "call ext4_journal_dirty_metadata"); - ext4_journal_dirty_metadata(handle, - parent_bh); + "call ext4_handle_dirty_metadata"); + ext4_handle_dirty_metadata(handle, + inode, + parent_bh); } } } @@ -3453,6 +4421,9 @@ void ext4_truncate(struct inode *inode) if (!ext4_can_truncate(inode)) return; + if (inode->i_size == 0 && !test_opt(inode->i_sb, NO_AUTO_DA_ALLOC)) + ei->i_state |= EXT4_STATE_DA_ALLOC_CLOSE; + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { ext4_ext_truncate(inode); return; @@ -3490,6 +4461,9 @@ void ext4_truncate(struct inode *inode) * modify the block allocation tree. */ down_write(&ei->i_data_sem); + + ext4_discard_preallocations(inode); + /* * The orphan list entry will now protect us from any crash which * occurs before the truncate completes, so it is now safe to propagate @@ -3531,7 +4505,7 @@ void ext4_truncate(struct inode *inode) (__le32*)partial->bh->b_data+addr_per_block, (chain+n-1) - partial); BUFFER_TRACE(partial->bh, "call brelse"); - brelse (partial->bh); + brelse(partial->bh); partial--; } do_indirects: @@ -3559,8 +4533,6 @@ do_indirects: ; } - ext4_discard_reservation(inode); - up_write(&ei->i_data_sem); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); @@ -3570,7 +4542,7 @@ do_indirects: * synchronous */ if (IS_SYNC(inode)) - handle->h_sync = 1; + ext4_handle_sync(handle); out_stop: /* * If this was a simple ftruncate(), and the file will remain alive @@ -3585,41 +4557,6 @@ out_stop: ext4_journal_stop(handle); } -static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, - unsigned long ino, struct ext4_iloc *iloc) -{ - ext4_group_t block_group; - unsigned long offset; - ext4_fsblk_t block; - struct ext4_group_desc *gdp; - - if (!ext4_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) / EXT4_INODES_PER_GROUP(sb); - gdp = ext4_get_group_desc(sb, block_group, NULL); - if (!gdp) - return 0; - - /* - * Figure out the offset within the block group inode table - */ - offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * - EXT4_INODE_SIZE(sb); - block = ext4_inode_table(sb, gdp) + - (offset >> EXT4_BLOCK_SIZE_BITS(sb)); - - iloc->block_group = block_group; - iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); - return block; -} - /* * ext4_get_inode_loc returns with an extra refcount against the inode's * underlying buffer_head on success. If 'in_mem' is true, we have all @@ -3629,19 +4566,35 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, static int __ext4_get_inode_loc(struct inode *inode, struct ext4_iloc *iloc, int in_mem) { - ext4_fsblk_t block; - struct buffer_head *bh; + struct ext4_group_desc *gdp; + struct buffer_head *bh; + struct super_block *sb = inode->i_sb; + ext4_fsblk_t block; + int inodes_per_block, inode_offset; + + iloc->bh = NULL; + if (!ext4_valid_inum(sb, inode->i_ino)) + return -EIO; - block = ext4_get_inode_block(inode->i_sb, inode->i_ino, iloc); - if (!block) + iloc->block_group = (inode->i_ino - 1) / EXT4_INODES_PER_GROUP(sb); + gdp = ext4_get_group_desc(sb, iloc->block_group, NULL); + if (!gdp) return -EIO; - bh = sb_getblk(inode->i_sb, block); + /* + * Figure out the offset within the block group inode table + */ + inodes_per_block = (EXT4_BLOCK_SIZE(sb) / EXT4_INODE_SIZE(sb)); + inode_offset = ((inode->i_ino - 1) % + EXT4_INODES_PER_GROUP(sb)); + block = ext4_inode_table(sb, gdp) + (inode_offset / inodes_per_block); + iloc->offset = (inode_offset % inodes_per_block) * EXT4_INODE_SIZE(sb); + + bh = sb_getblk(sb, block); if (!bh) { - ext4_error (inode->i_sb, "ext4_get_inode_loc", - "unable to read inode block - " - "inode=%lu, block=%llu", - inode->i_ino, block); + ext4_error(sb, "ext4_get_inode_loc", "unable to read " + "inode block - inode=%lu, block=%llu", + inode->i_ino, block); return -EIO; } if (!buffer_uptodate(bh)) { @@ -3669,28 +4622,12 @@ static int __ext4_get_inode_loc(struct inode *inode, */ if (in_mem) { struct buffer_head *bitmap_bh; - struct ext4_group_desc *desc; - int inodes_per_buffer; - int inode_offset, i; - ext4_group_t block_group; - int start; - - block_group = (inode->i_ino - 1) / - EXT4_INODES_PER_GROUP(inode->i_sb); - inodes_per_buffer = bh->b_size / - EXT4_INODE_SIZE(inode->i_sb); - inode_offset = ((inode->i_ino - 1) % - EXT4_INODES_PER_GROUP(inode->i_sb)); - start = inode_offset & ~(inodes_per_buffer - 1); + int i, start; - /* Is the inode bitmap in cache? */ - desc = ext4_get_group_desc(inode->i_sb, - block_group, NULL); - if (!desc) - goto make_io; + start = inode_offset & ~(inodes_per_block - 1); - bitmap_bh = sb_getblk(inode->i_sb, - ext4_inode_bitmap(inode->i_sb, desc)); + /* Is the inode bitmap in cache? */ + bitmap_bh = sb_getblk(sb, ext4_inode_bitmap(sb, gdp)); if (!bitmap_bh) goto make_io; @@ -3703,14 +4640,14 @@ static int __ext4_get_inode_loc(struct inode *inode, brelse(bitmap_bh); goto make_io; } - for (i = start; i < start + inodes_per_buffer; i++) { + for (i = start; i < start + inodes_per_block; i++) { if (i == inode_offset) continue; if (ext4_test_bit(i, bitmap_bh->b_data)) break; } brelse(bitmap_bh); - if (i == start + inodes_per_buffer) { + if (i == start + inodes_per_block) { /* all other inodes are free, so skip I/O */ memset(bh->b_data, 0, bh->b_size); set_buffer_uptodate(bh); @@ -3721,6 +4658,31 @@ static int __ext4_get_inode_loc(struct inode *inode, make_io: /* + * If we need to do any I/O, try to pre-readahead extra + * blocks from the inode table. + */ + if (EXT4_SB(sb)->s_inode_readahead_blks) { + ext4_fsblk_t b, end, table; + unsigned num; + + table = ext4_inode_table(sb, gdp); + /* s_inode_readahead_blks is always a power of 2 */ + b = block & ~(EXT4_SB(sb)->s_inode_readahead_blks-1); + if (table > b) + b = table; + end = b + EXT4_SB(sb)->s_inode_readahead_blks; + num = EXT4_INODES_PER_GROUP(sb); + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) + num -= ext4_itable_unused_count(sb, gdp); + table += num / inodes_per_block; + if (end > table) + end = table; + while (b <= end) + sb_breadahead(sb, b++); + } + + /* * There are other valid inodes in the buffer, this inode * has in-inode xattrs, or we don't have this inode in memory. * Read the block from disk. @@ -3730,10 +4692,9 @@ make_io: submit_bh(READ_META, bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) { - ext4_error(inode->i_sb, "ext4_get_inode_loc", - "unable to read inode block - " - "inode=%lu, block=%llu", - inode->i_ino, block); + ext4_error(sb, __func__, + "unable to read inode block - inode=%lu, " + "block=%llu", inode->i_ino, block); brelse(bh); return -EIO; } @@ -3785,8 +4746,9 @@ void ext4_get_inode_flags(struct ext4_inode_info *ei) if (flags & S_DIRSYNC) ei->i_flags |= EXT4_DIRSYNC_FL; } + static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode, - struct ext4_inode_info *ei) + struct ext4_inode_info *ei) { blkcnt_t i_blocks ; struct inode *inode = &(ei->vfs_inode); @@ -3825,11 +4787,6 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) return inode; ei = EXT4_I(inode); -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL - ei->i_acl = EXT4_ACL_NOT_CACHED; - ei->i_default_acl = EXT4_ACL_NOT_CACHED; -#endif - ei->i_block_alloc_info = NULL; ret = __ext4_get_inode_loc(inode, &iloc, 0); if (ret < 0) @@ -3839,7 +4796,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) inode->i_mode = le16_to_cpu(raw_inode->i_mode); inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low); inode->i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low); - if(!(test_opt (inode->i_sb, NO_UID32))) { + if (!(test_opt(inode->i_sb, NO_UID32))) { inode->i_uid |= le16_to_cpu(raw_inode->i_uid_high) << 16; inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16; } @@ -3857,7 +4814,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) if (inode->i_mode == 0 || !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) { /* this inode is deleted */ - brelse (bh); + brelse(bh); ret = -ESTALE; goto bad_inode; } @@ -3869,15 +4826,14 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ei->i_flags = le32_to_cpu(raw_inode->i_flags); inode->i_blocks = ext4_inode_blocks(raw_inode, ei); ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo); - if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != - cpu_to_le32(EXT4_OS_HURD)) { + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) ei->i_file_acl |= ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; - } inode->i_size = ext4_isize(raw_inode); ei->i_disksize = inode->i_size; inode->i_generation = le32_to_cpu(raw_inode->i_generation); ei->i_block_group = iloc.block_group; + ei->i_last_alloc_group = ~0; /* * NOTE! The in-memory inode i_data array is in little-endian order * even on big-endian machines: we do NOT byteswap the block numbers! @@ -3890,7 +4846,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > EXT4_INODE_SIZE(inode->i_sb)) { - brelse (bh); + brelse(bh); ret = -EIO; goto bad_inode; } @@ -3903,7 +4859,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize; if (*magic == cpu_to_le32(EXT4_XATTR_MAGIC)) - ei->i_state |= EXT4_STATE_XATTR; + ei->i_state |= EXT4_STATE_XATTR; } } else ei->i_extra_isize = 0; @@ -3920,6 +4876,34 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) (__u64)(le32_to_cpu(raw_inode->i_version_hi)) << 32; } + ret = 0; + if (ei->i_file_acl && + ((ei->i_file_acl < + (le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block) + + EXT4_SB(sb)->s_gdb_count)) || + (ei->i_file_acl >= ext4_blocks_count(EXT4_SB(sb)->s_es)))) { + ext4_error(sb, __func__, + "bad extended attribute block %llu in inode #%lu", + ei->i_file_acl, inode->i_ino); + ret = -EIO; + goto bad_inode; + } else if (ei->i_flags & EXT4_EXTENTS_FL) { + if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + (S_ISLNK(inode->i_mode) && + !ext4_inode_is_fast_symlink(inode))) + /* Validate extent which is part of inode */ + ret = ext4_ext_check_inode(inode); + } else if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + (S_ISLNK(inode->i_mode) && + !ext4_inode_is_fast_symlink(inode))) { + /* Validate block references which are part of inode */ + ret = ext4_check_inode_blockref(inode); + } + if (ret) { + brelse(bh); + goto bad_inode; + } + if (S_ISREG(inode->i_mode)) { inode->i_op = &ext4_file_inode_operations; inode->i_fop = &ext4_file_operations; @@ -3928,13 +4912,16 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) inode->i_op = &ext4_dir_inode_operations; inode->i_fop = &ext4_dir_operations; } else if (S_ISLNK(inode->i_mode)) { - if (ext4_inode_is_fast_symlink(inode)) + if (ext4_inode_is_fast_symlink(inode)) { inode->i_op = &ext4_fast_symlink_inode_operations; - else { + nd_terminate_link(ei->i_data, inode->i_size, + sizeof(ei->i_data) - 1); + } else { inode->i_op = &ext4_symlink_inode_operations; ext4_set_aops(inode); } - } else { + } else if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode) || + S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) { inode->i_op = &ext4_special_inode_operations; if (raw_inode->i_block[0]) init_special_inode(inode, inode->i_mode, @@ -3942,8 +4929,15 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) else init_special_inode(inode, inode->i_mode, new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); + } else { + brelse(bh); + ret = -EIO; + ext4_error(inode->i_sb, __func__, + "bogus i_mode (%o) for inode=%lu", + inode->i_mode, inode->i_ino); + goto bad_inode; } - brelse (iloc.bh); + brelse(iloc.bh); ext4_set_inode_flags(inode); unlock_new_inode(inode); return inode; @@ -3960,7 +4954,6 @@ static int ext4_inode_blocks_set(handle_t *handle, struct inode *inode = &(ei->vfs_inode); u64 i_blocks = inode->i_blocks; struct super_block *sb = inode->i_sb; - int err = 0; if (i_blocks <= ~0U) { /* @@ -3970,36 +4963,27 @@ static int ext4_inode_blocks_set(handle_t *handle, raw_inode->i_blocks_lo = cpu_to_le32(i_blocks); raw_inode->i_blocks_high = 0; ei->i_flags &= ~EXT4_HUGE_FILE_FL; - } else if (i_blocks <= 0xffffffffffffULL) { + return 0; + } + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) + return -EFBIG; + + if (i_blocks <= 0xffffffffffffULL) { /* * i_blocks can be represented in a 48 bit variable * as multiple of 512 bytes */ - err = ext4_update_rocompat_feature(handle, sb, - EXT4_FEATURE_RO_COMPAT_HUGE_FILE); - if (err) - goto err_out; - /* i_block is stored in the split 48 bit fields */ raw_inode->i_blocks_lo = cpu_to_le32(i_blocks); raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32); ei->i_flags &= ~EXT4_HUGE_FILE_FL; } else { - /* - * i_blocks should be represented in a 48 bit variable - * as multiple of file system block size - */ - err = ext4_update_rocompat_feature(handle, sb, - EXT4_FEATURE_RO_COMPAT_HUGE_FILE); - if (err) - goto err_out; ei->i_flags |= EXT4_HUGE_FILE_FL; /* i_block is stored in file system block size */ i_blocks = i_blocks >> (inode->i_blkbits - 9); raw_inode->i_blocks_lo = cpu_to_le32(i_blocks); raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32); } -err_out: - return err; + return 0; } /* @@ -4025,14 +5009,14 @@ static int ext4_do_update_inode(handle_t *handle, ext4_get_inode_flags(ei); raw_inode->i_mode = cpu_to_le16(inode->i_mode); - if(!(test_opt(inode->i_sb, NO_UID32))) { + if (!(test_opt(inode->i_sb, NO_UID32))) { raw_inode->i_uid_low = cpu_to_le16(low_16_bits(inode->i_uid)); raw_inode->i_gid_low = cpu_to_le16(low_16_bits(inode->i_gid)); /* * Fix up interoperability with old kernels. Otherwise, old inodes get * re-used with the upper 16 bits of the uid/gid intact */ - if(!ei->i_dtime) { + if (!ei->i_dtime) { raw_inode->i_uid_high = cpu_to_le16(high_16_bits(inode->i_uid)); raw_inode->i_gid_high = @@ -4059,8 +5043,7 @@ static int ext4_do_update_inode(handle_t *handle, if (ext4_inode_blocks_set(handle, raw_inode, ei)) goto out_brelse; raw_inode->i_dtime = cpu_to_le32(ei->i_dtime); - /* clear the migrate flag in the raw_inode */ - raw_inode->i_flags = cpu_to_le32(ei->i_flags & ~EXT4_EXT_MIGRATE); + raw_inode->i_flags = cpu_to_le32(ei->i_flags); if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != cpu_to_le32(EXT4_OS_HURD)) raw_inode->i_file_acl_high = @@ -4084,8 +5067,8 @@ static int ext4_do_update_inode(handle_t *handle, EXT4_SET_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_LARGE_FILE); sb->s_dirt = 1; - handle->h_sync = 1; - err = ext4_journal_dirty_metadata(handle, + ext4_handle_sync(handle); + err = ext4_handle_dirty_metadata(handle, inode, EXT4_SB(sb)->s_sbh); } } @@ -4101,8 +5084,9 @@ static int ext4_do_update_inode(handle_t *handle, cpu_to_le32(new_encode_dev(inode->i_rdev)); raw_inode->i_block[2] = 0; } - } else for (block = 0; block < EXT4_N_BLOCKS; block++) - raw_inode->i_block[block] = ei->i_data[block]; + } else + for (block = 0; block < EXT4_N_BLOCKS; block++) + raw_inode->i_block[block] = ei->i_data[block]; raw_inode->i_disk_version = cpu_to_le32(inode->i_version); if (ei->i_extra_isize) { @@ -4112,15 +5096,14 @@ static int ext4_do_update_inode(handle_t *handle, raw_inode->i_extra_isize = cpu_to_le16(ei->i_extra_isize); } - - BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); - rc = ext4_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); + rc = ext4_handle_dirty_metadata(handle, inode, bh); if (!err) err = rc; ei->i_state &= ~EXT4_STATE_NEW; out_brelse: - brelse (bh); + brelse(bh); ext4_std_error(inode->i_sb, err); return err; } @@ -4162,19 +5145,40 @@ out_brelse: */ int ext4_write_inode(struct inode *inode, int wait) { + int err; + if (current->flags & PF_MEMALLOC) return 0; - if (ext4_journal_current_handle()) { - jbd_debug(1, "called recursively, non-PF_MEMALLOC!\n"); - dump_stack(); - return -EIO; - } + if (EXT4_SB(inode->i_sb)->s_journal) { + if (ext4_journal_current_handle()) { + jbd_debug(1, "called recursively, non-PF_MEMALLOC!\n"); + dump_stack(); + return -EIO; + } - if (!wait) - return 0; + if (!wait) + return 0; + + err = ext4_force_commit(inode->i_sb); + } else { + struct ext4_iloc iloc; - return ext4_force_commit(inode->i_sb); + err = ext4_get_inode_loc(inode, &iloc); + if (err) + return err; + if (wait) + sync_dirty_buffer(iloc.bh); + if (buffer_req(iloc.bh) && !buffer_uptodate(iloc.bh)) { + ext4_error(inode->i_sb, __func__, + "IO error syncing inode, " + "inode=%lu, block=%llu", + inode->i_ino, + (unsigned long long)iloc.bh->b_blocknr); + err = -EIO; + } + } + return err; } /* @@ -4223,7 +5227,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) error = PTR_ERR(handle); goto err_out; } - error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0; + error = vfs_dq_transfer(inode, attr) ? -EDQUOT : 0; if (error) { ext4_journal_stop(handle); return error; @@ -4328,62 +5332,136 @@ int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, return 0; } +static int ext4_indirect_trans_blocks(struct inode *inode, int nrblocks, + int chunk) +{ + int indirects; + + /* if nrblocks are contiguous */ + if (chunk) { + /* + * With N contiguous data blocks, it need at most + * N/EXT4_ADDR_PER_BLOCK(inode->i_sb) indirect blocks + * 2 dindirect blocks + * 1 tindirect block + */ + indirects = nrblocks / EXT4_ADDR_PER_BLOCK(inode->i_sb); + return indirects + 3; + } + /* + * 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; +} + +static int ext4_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) +{ + if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) + return ext4_indirect_trans_blocks(inode, nrblocks, chunk); + return ext4_ext_index_trans_blocks(inode, nrblocks, chunk); +} + /* - * How many blocks doth make a writepage()? - * - * With N blocks per page, it may be: - * N data blocks - * 2 indirect block - * 2 dindirect - * 1 tindirect - * N+5 bitmap blocks (from the above) - * N+5 group descriptor summary blocks - * 1 inode block - * 1 superblock. - * 2 * EXT4_SINGLEDATA_TRANS_BLOCKS for the quote files + * Account for index blocks, block groups bitmaps and block group + * descriptor blocks if modify datablocks and index blocks + * worse case, the indexs blocks spread over different block groups * - * 3 * (N + 5) + 2 + 2 * EXT4_SINGLEDATA_TRANS_BLOCKS + * If datablocks are discontiguous, they are possible to spread over + * different block groups too. If they are contiuguous, with flexbg, + * they could still across block group boundary. * - * With ordered or writeback data it's the same, less the N data blocks. + * Also account for superblock, inode, quota and xattr blocks + */ +int ext4_meta_trans_blocks(struct inode *inode, int nrblocks, int chunk) +{ + ext4_group_t groups, ngroups = ext4_get_groups_count(inode->i_sb); + int gdpblocks; + int idxblocks; + int ret = 0; + + /* + * How many index blocks need to touch to modify nrblocks? + * The "Chunk" flag indicating whether the nrblocks is + * physically contiguous on disk + * + * For Direct IO and fallocate, they calls get_block to allocate + * one single extent at a time, so they could set the "Chunk" flag + */ + idxblocks = ext4_index_trans_blocks(inode, nrblocks, chunk); + + ret = idxblocks; + + /* + * Now let's see how many group bitmaps and group descriptors need + * to account + */ + groups = idxblocks; + if (chunk) + groups += 1; + else + groups += nrblocks; + + gdpblocks = groups; + if (groups > ngroups) + groups = ngroups; + if (groups > EXT4_SB(inode->i_sb)->s_gdb_count) + gdpblocks = EXT4_SB(inode->i_sb)->s_gdb_count; + + /* bitmaps and block group descriptor blocks */ + ret += groups + gdpblocks; + + /* Blocks for super block, inode, quota and xattr blocks */ + ret += EXT4_META_TRANS_BLOCKS(inode->i_sb); + + return ret; +} + +/* + * Calulate the total number of credits to reserve to fit + * the modification of a single pages into a single transaction, + * which may include multiple chunks of block allocations. * - * If the inode's direct blocks can hold an integral number of pages then a - * page cannot straddle two indirect blocks, and we can only touch one indirect - * and dindirect block, and the "5" above becomes "3". + * This could be called via ext4_write_begin() * - * This still overestimates under most circumstances. If we were to pass the - * start and end offsets in here as well we could do block_to_path() on each - * block and work out the exact number of indirects which are touched. Pah. + * We need to consider the worse case, when + * one new block per extent. */ - int ext4_writepage_trans_blocks(struct inode *inode) { int bpp = ext4_journal_blocks_per_page(inode); - int indirects = (EXT4_NDIR_BLOCKS % bpp) ? 5 : 3; int ret; - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) - return ext4_ext_writepage_trans_blocks(inode, bpp); + ret = ext4_meta_trans_blocks(inode, bpp, 0); + /* Account for data blocks for journalled mode */ if (ext4_should_journal_data(inode)) - ret = 3 * (bpp + indirects) + 2; - else - ret = 2 * (bpp + indirects) + 2; - -#ifdef CONFIG_QUOTA - /* We know that structure was already allocated during DQUOT_INIT so - * we will be updating only the data blocks + inodes */ - ret += 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); -#endif - + ret += bpp; return ret; } /* + * Calculate the journal credits for a chunk of data modification. + * + * This is called from DIO, fallocate or whoever calling + * ext4_get_blocks() to map/allocate a chunk of contiguous disk blocks. + * + * journal buffers for data blocks are not included here, as DIO + * and fallocate do no need to journal data buffers. + */ +int ext4_chunk_trans_blocks(struct inode *inode, int nrblocks) +{ + return ext4_meta_trans_blocks(inode, nrblocks, 1); +} + +/* * The caller must have previously called ext4_reserve_inode_write(). * Give this, we know that the caller already has write access to iloc->bh. */ int ext4_mark_iloc_dirty(handle_t *handle, - struct inode *inode, struct ext4_iloc *iloc) + struct inode *inode, struct ext4_iloc *iloc) { int err = 0; @@ -4408,16 +5486,15 @@ int ext4_reserve_inode_write(handle_t *handle, struct inode *inode, struct ext4_iloc *iloc) { - int err = 0; - if (handle) { - err = ext4_get_inode_loc(inode, iloc); - if (!err) { - BUFFER_TRACE(iloc->bh, "get_write_access"); - err = ext4_journal_get_write_access(handle, iloc->bh); - if (err) { - brelse(iloc->bh); - iloc->bh = NULL; - } + int err; + + err = ext4_get_inode_loc(inode, iloc); + if (!err) { + BUFFER_TRACE(iloc->bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, iloc->bh); + if (err) { + brelse(iloc->bh); + iloc->bh = NULL; } } ext4_std_error(inode->i_sb, err); @@ -4489,7 +5566,8 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) might_sleep(); err = ext4_reserve_inode_write(handle, inode, &iloc); - if (EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize && + if (ext4_handle_valid(handle) && + EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize && !(EXT4_I(inode)->i_state & EXT4_STATE_NO_EXPAND)) { /* * We need extra buffer credits since we may write into EA block @@ -4529,7 +5607,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) * i_size has been changed by generic_commit_write() and we thus need * to include the updated inode in the current transaction. * - * Also, DQUOT_ALLOC_SPACE() will always dirty the inode when blocks + * Also, vfs_dq_alloc_block() will always dirty the inode when blocks * are allocated to the file. * * If the inode is marked synchronous, we don't honour that here - doing @@ -4538,22 +5616,14 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) */ void ext4_dirty_inode(struct inode *inode) { - handle_t *current_handle = ext4_journal_current_handle(); handle_t *handle; handle = ext4_journal_start(inode, 2); if (IS_ERR(handle)) goto out; - if (current_handle && - current_handle->h_transaction != handle->h_transaction) { - /* This task has a transaction open against a different fs */ - printk(KERN_EMERG "%s: transactions do not match!\n", - __func__); - } else { - jbd_debug(5, "marking dirty. outer handle=%p\n", - current_handle); - ext4_mark_inode_dirty(handle, inode); - } + + ext4_mark_inode_dirty(handle, inode); + ext4_journal_stop(handle); out: return; @@ -4578,8 +5648,9 @@ static int ext4_pin_inode(handle_t *handle, struct inode *inode) BUFFER_TRACE(iloc.bh, "get_write_access"); err = jbd2_journal_get_write_access(handle, iloc.bh); if (!err) - err = ext4_journal_dirty_metadata(handle, - iloc.bh); + err = ext4_handle_dirty_metadata(handle, + inode, + iloc.bh); brelse(iloc.bh); } } @@ -4605,6 +5676,8 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) */ journal = EXT4_JOURNAL(inode); + if (!journal) + return 0; if (is_journal_aborted(journal)) return -EROFS; @@ -4634,7 +5707,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) return PTR_ERR(handle); err = ext4_mark_inode_dirty(handle, inode); - handle->h_sync = 1; + ext4_handle_sync(handle); ext4_journal_stop(handle); ext4_std_error(inode->i_sb, err); @@ -4646,11 +5719,13 @@ static int ext4_bh_unmapped(handle_t *handle, struct buffer_head *bh) return !buffer_mapped(bh); } -int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) +int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) { + struct page *page = vmf->page; loff_t size; unsigned long len; int ret = -EINVAL; + void *fsdata; struct file *file = vma->vm_file; struct inode *inode = file->f_path.dentry->d_inode; struct address_space *mapping = inode->i_mapping; @@ -4675,12 +5750,21 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) else len = PAGE_CACHE_SIZE; + lock_page(page); + /* + * return if we have all the buffers mapped. This avoid + * the need to call write_begin/write_end which does a + * journal_start/journal_stop which can block and take + * long time + */ if (page_has_buffers(page)) { - /* return if we have all the buffers mapped */ if (!walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, - ext4_bh_unmapped)) + ext4_bh_unmapped)) { + unlock_page(page); goto out_unlock; + } } + unlock_page(page); /* * OK, we need to fill the hole... Do write_begin write_end * to do block allocation/reservation.We are not holding @@ -4689,15 +5773,17 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) * on the same page though */ ret = mapping->a_ops->write_begin(file, mapping, page_offset(page), - len, AOP_FLAG_UNINTERRUPTIBLE, &page, NULL); + len, AOP_FLAG_UNINTERRUPTIBLE, &page, &fsdata); if (ret < 0) goto out_unlock; ret = mapping->a_ops->write_end(file, mapping, page_offset(page), - len, len, page, NULL); + len, len, page, fsdata); if (ret < 0) goto out_unlock; ret = 0; out_unlock: + if (ret) + ret = VM_FAULT_SIGBUS; up_read(&inode->i_alloc_sem); return ret; }