X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=fs%2Fext3%2Finode.c;h=47839de34f7fdb52f489d79d24c298d50279b1d5;hb=refs%2Fheads%2Fvserver;hp=23d956ff0870ecc23f6679675256300fbff63f98;hpb=16c70f8c1b54b61c3b951b6fb220df250fe09b32;p=linux-2.6.git diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 23d956ff0..47839de34 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -13,11 +13,11 @@ * Copyright (C) 1991, 1992 Linus Torvalds * * Goal-directed block allocation by Stephen Tweedie - * (sct@redhat.com), 1993, 1998 + * (sct@redhat.com), 1993, 1998 * Big-endian to little-endian byte-swapping/bitmaps by * David S. Miller (davem@caip.rutgers.edu), 1995 * 64-bit file support on 64-bit platforms by Jakub Jelinek - * (jj@sunsite.ms.mff.cuni.cz) + * (jj@sunsite.ms.mff.cuni.cz) * * Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000 */ @@ -36,7 +36,8 @@ #include #include #include -#include +#include +#include #include "xattr.h" #include "acl.h" @@ -56,7 +57,7 @@ static int ext3_inode_is_fast_symlink(struct inode *inode) /* * The ext3 forget function must perform a revoke if we are freeing data * which has been journaled. Metadata (eg. indirect blocks) must be - * revoked in all cases. + * revoked in all cases. * * "bh" may be NULL: a metadata block may have been freed from memory * but there may still be a record of it in the journal, and that record @@ -106,7 +107,7 @@ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, * Work out how many blocks we need to proceed with the next chunk of a * truncate transaction. */ -static unsigned long blocks_for_truncate(struct inode *inode) +static unsigned long blocks_for_truncate(struct inode *inode) { unsigned long needed; @@ -123,13 +124,13 @@ static unsigned long blocks_for_truncate(struct inode *inode) /* But we need to bound the transaction so we don't overflow the * journal. */ - if (needed > EXT3_MAX_TRANS_DATA) + if (needed > EXT3_MAX_TRANS_DATA) needed = EXT3_MAX_TRANS_DATA; return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed; } -/* +/* * Truncate transactions can be complex and absolutely huge. So we need to * be able to restart the transaction at a conventient checkpoint to make * sure we don't overflow the journal. @@ -137,9 +138,9 @@ static unsigned long blocks_for_truncate(struct inode *inode) * start_transaction gets us a new handle for a truncate transaction, * and extend_transaction tries to extend the existing one a bit. If * extend fails, we need to propagate the failure up and restart the - * transaction in the top-level truncate loop. --sct + * transaction in the top-level truncate loop. --sct */ -static handle_t *start_transaction(struct inode *inode) +static handle_t *start_transaction(struct inode *inode) { handle_t *result; @@ -216,12 +217,12 @@ void ext3_delete_inode (struct inode * inode) ext3_orphan_del(handle, inode); EXT3_I(inode)->i_dtime = get_seconds(); - /* + /* * One subtle ordering requirement: if anything has gone wrong * (transaction abort, IO errors, whatever), then we can still * do these next steps (the fs will already have been marked as * having errors), but we can't free the inode if the mark_dirty - * fails. + * fails. */ if (ext3_mark_inode_dirty(handle, inode)) /* If that failed, just do the required in-core inode clear. */ @@ -399,7 +400,7 @@ no_block: * + if there is a block to the left of our position - allocate near it. * + if pointer will live in indirect block - allocate near that block. * + if pointer will live in inode - allocate in the same - * cylinder group. + * cylinder group. * * In the latter case we colour the starting block by the callers PID to * prevent it from clashing with concurrent allocations for a different inode @@ -471,7 +472,7 @@ static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block, * ext3_blks_to_allocate: Look up the block map and count the number * of direct blocks need to be allocated for the given branch. * - * @branch: chain of indirect blocks + * @branch: chain of indirect blocks * @k: number of blocks need for indirect blocks * @blks: number of data blocks to be mapped. * @blocks_to_boundary: the offset in the indirect block @@ -745,7 +746,7 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, jbd_debug(5, "splicing indirect only\n"); BUFFER_TRACE(where->bh, "call ext3_journal_dirty_metadata"); err = ext3_journal_dirty_metadata(handle, where->bh); - if (err) + if (err) goto err_out; } else { /* @@ -1074,7 +1075,7 @@ struct buffer_head *ext3_bread(handle_t *handle, struct inode *inode, return bh; if (buffer_uptodate(bh)) return bh; - ll_rw_block(READ, 1, &bh); + ll_rw_block(READ_META, 1, &bh); wait_on_buffer(bh); if (buffer_uptodate(bh)) return bh; @@ -1099,7 +1100,7 @@ static int walk_page_buffers( handle_t *handle, for ( bh = head, block_start = 0; ret == 0 && (bh != head || !block_start); - block_start = block_end, bh = next) + block_start = block_end, bh = next) { next = bh->b_this_page; block_end = block_start + blocksize; @@ -1138,7 +1139,7 @@ static int walk_page_buffers( handle_t *handle, * So what we do is to rely on the fact that journal_stop/journal_start * will _not_ run commit under these circumstances because handle->h_ref * is elevated. We'll still have enough credits for the tiny quotafile - * write. + * write. */ static int do_journal_get_write_access(handle_t *handle, struct buffer_head *bh) @@ -1283,7 +1284,7 @@ static int ext3_journalled_commit_write(struct file *file, if (inode->i_size > EXT3_I(inode)->i_disksize) { EXT3_I(inode)->i_disksize = inode->i_size; ret2 = ext3_mark_inode_dirty(handle, inode); - if (!ret) + if (!ret) ret = ret2; } ret2 = ext3_journal_stop(handle); @@ -1292,7 +1293,7 @@ static int ext3_journalled_commit_write(struct file *file, return ret; } -/* +/* * bmap() is special. It gets used by applications such as lilo and by * the swapper to find the on-disk block of a specific piece of data. * @@ -1301,10 +1302,10 @@ static int ext3_journalled_commit_write(struct file *file, * filesystem and enables swap, then they may get a nasty shock when the * data getting swapped to that swapfile suddenly gets overwritten by * the original zero's written out previously to the journal and - * awaiting writeback in the kernel's buffer cache. + * awaiting writeback in the kernel's buffer cache. * * So, if we see any bmap calls here on a modified, data-journaled file, - * take extra steps to flush any blocks which might be in the cache. + * take extra steps to flush any blocks which might be in the cache. */ static sector_t ext3_bmap(struct address_space *mapping, sector_t block) { @@ -1313,16 +1314,16 @@ static sector_t ext3_bmap(struct address_space *mapping, sector_t block) int err; if (EXT3_I(inode)->i_state & EXT3_STATE_JDATA) { - /* + /* * This is a REALLY heavyweight approach, but the use of * bmap on dirty files is expected to be extremely rare: * only if we run lilo or swapon on a freshly made file - * do we expect this to happen. + * do we expect this to happen. * * (bmap requires CAP_SYS_RAWIO so this does not * represent an unprivileged user DOS attack --- we'd be * in trouble if mortal users could trigger this path at - * will.) + * will.) * * NB. EXT3_STATE_JDATA is not set on files other than * regular files. If somebody wants to bmap a directory @@ -1458,7 +1459,7 @@ static int ext3_ordered_writepage(struct page *page, */ /* - * And attach them to the current transaction. But only if + * And attach them to the current transaction. But only if * block_write_full_page() succeeded. Otherwise they are unmapped, * and generally junk. */ @@ -1531,12 +1532,12 @@ static int ext3_journalled_writepage(struct page *page, goto no_write; } - if (!page_has_buffers(page) || PageFsMisc(page)) { + if (!page_has_buffers(page) || PageChecked(page)) { /* * It's mmapped pagecache. Add buffers and journal it. There * doesn't seem much point in redirtying the page here. */ - ClearPageFsMisc(page); + ClearPageChecked(page); ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, ext3_get_block); if (ret != 0) { @@ -1593,7 +1594,7 @@ static void ext3_invalidatepage(struct page *page, unsigned long offset) * If it's a full truncate we just forget about the pending dirtying */ if (offset == 0) - ClearPageFsMisc(page); + ClearPageChecked(page); journal_invalidatepage(journal, page, offset); } @@ -1602,7 +1603,7 @@ static int ext3_releasepage(struct page *page, gfp_t wait) { journal_t *journal = EXT3_JOURNAL(page->mapping->host); - WARN_ON(PageFsMisc(page)); + WARN_ON(PageChecked(page)); if (!page_has_buffers(page)) return 0; return journal_try_to_free_buffers(journal, page, wait); @@ -1645,7 +1646,7 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, } } - ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, + ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, offset, nr_segs, ext3_get_block, NULL); @@ -1698,7 +1699,7 @@ out: */ static int ext3_journalled_set_page_dirty(struct page *page) { - SetPageFsMisc(page); + SetPageChecked(page); return __set_page_dirty_nobuffers(page); } @@ -2026,7 +2027,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, __le32 *first, __le32 *last) { ext3_fsblk_t block_to_free = 0; /* Starting block # of a run */ - unsigned long count = 0; /* Number of blocks in the run */ + unsigned long count = 0; /* Number of blocks in the run */ __le32 *block_to_free_p = NULL; /* Pointer into inode/ind corresponding to block_to_free */ @@ -2055,7 +2056,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, } else if (nr == block_to_free + count) { count++; } else { - ext3_clear_blocks(handle, inode, this_bh, + ext3_clear_blocks(handle, inode, this_bh, block_to_free, count, block_to_free_p, p); block_to_free = nr; @@ -2185,7 +2186,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, *p = 0; BUFFER_TRACE(parent_bh, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, + ext3_journal_dirty_metadata(handle, parent_bh); } } @@ -2541,7 +2542,7 @@ make_io: */ get_bh(bh); bh->b_end_io = end_buffer_read_sync; - submit_bh(READ, bh); + submit_bh(READ_META, bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) { ext3_error(inode->i_sb, "ext3_get_inode_loc", @@ -2594,13 +2595,9 @@ int ext3_sync_flags(struct inode *inode) int err = 0; oldflags = EXT3_I(inode)->i_flags; - newflags = oldflags & ~(EXT3_APPEND_FL | - EXT3_IMMUTABLE_FL | EXT3_IUNLINK_FL | - EXT3_BARRIER_FL | EXT3_NOATIME_FL | - EXT3_SYNC_FL | EXT3_DIRSYNC_FL); + newflags = oldflags & ~(EXT3_IMMUTABLE_FL | + EXT3_IUNLINK_FL | EXT3_BARRIER_FL); - if (IS_APPEND(inode)) - newflags |= EXT3_APPEND_FL; if (IS_IMMUTABLE(inode)) newflags |= EXT3_IMMUTABLE_FL; if (IS_IUNLINK(inode)) @@ -2608,14 +2605,6 @@ int ext3_sync_flags(struct inode *inode) if (IS_BARRIER(inode)) newflags |= EXT3_BARRIER_FL; - /* we do not want to copy superblock flags */ - if (inode->i_flags & S_NOATIME) - newflags |= EXT3_NOATIME_FL; - if (inode->i_flags & S_SYNC) - newflags |= EXT3_SYNC_FL; - if (inode->i_flags & S_DIRSYNC) - newflags |= EXT3_DIRSYNC_FL; - if (oldflags ^ newflags) { handle_t *handle; struct ext3_iloc iloc; @@ -2666,10 +2655,10 @@ void ext3_read_inode(struct inode * inode) uid |= le16_to_cpu(raw_inode->i_uid_high) << 16; gid |= le16_to_cpu(raw_inode->i_gid_high) << 16; } - inode->i_uid = INOXID_UID(XID_TAG(inode), uid, gid); - inode->i_gid = INOXID_GID(XID_TAG(inode), uid, gid); - inode->i_xid = INOXID_XID(XID_TAG(inode), uid, gid, - le16_to_cpu(raw_inode->i_raw_xid)); + inode->i_uid = INOTAG_UID(DX_TAG(inode), uid, gid); + inode->i_gid = INOTAG_GID(DX_TAG(inode), uid, gid); + inode->i_tag = INOTAG_TAG(DX_TAG(inode), uid, gid, + le16_to_cpu(raw_inode->i_raw_tag)); inode->i_nlink = le16_to_cpu(raw_inode->i_links_count); inode->i_size = le32_to_cpu(raw_inode->i_size); @@ -2767,7 +2756,7 @@ void ext3_read_inode(struct inode * inode) if (raw_inode->i_block[0]) init_special_inode(inode, inode->i_mode, old_decode_dev(le32_to_cpu(raw_inode->i_block[0]))); - else + else init_special_inode(inode, inode->i_mode, new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); } @@ -2787,15 +2776,15 @@ bad_inode: * * The caller must have write access to iloc->bh. */ -static int ext3_do_update_inode(handle_t *handle, - struct inode *inode, +static int ext3_do_update_inode(handle_t *handle, + struct inode *inode, struct ext3_iloc *iloc) { struct ext3_inode *raw_inode = ext3_raw_inode(iloc); struct ext3_inode_info *ei = EXT3_I(inode); struct buffer_head *bh = iloc->bh; - uid_t uid = XIDINO_UID(XID_TAG(inode), inode->i_uid, inode->i_xid); - gid_t gid = XIDINO_GID(XID_TAG(inode), inode->i_gid, inode->i_xid); + uid_t uid = TAGINO_UID(DX_TAG(inode), inode->i_uid, inode->i_tag); + gid_t gid = TAGINO_GID(DX_TAG(inode), inode->i_gid, inode->i_tag); int err = 0, rc, block; /* For fields not not tracking in the in-memory inode, @@ -2828,8 +2817,8 @@ static int ext3_do_update_inode(handle_t *handle, raw_inode->i_uid_high = 0; raw_inode->i_gid_high = 0; } -#ifdef CONFIG_INOXID_INTERN - raw_inode->i_raw_xid = cpu_to_le16(inode->i_xid); +#ifdef CONFIG_TAGGING_INTERN + raw_inode->i_raw_tag = cpu_to_le16(inode->i_tag); #endif raw_inode->i_links_count = cpu_to_le16(inode->i_nlink); raw_inode->i_size = cpu_to_le32(ei->i_disksize); @@ -2968,7 +2957,7 @@ int ext3_write_inode(struct inode *inode, int wait) * commit will leave the blocks being flushed in an unused state on * disk. (On recovery, the inode will get truncated and the blocks will * be freed, so we have a strong guarantee that no future commit will - * leave these blocks visible to the user.) + * leave these blocks visible to the user.) * * Called with inode->sem down. */ @@ -2984,7 +2973,7 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr) if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) || (ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid) || - (ia_valid & ATTR_XID && attr->ia_xid != inode->i_xid)) { + (ia_valid & ATTR_TAG && attr->ia_tag != inode->i_tag)) { handle_t *handle; /* (user+group)*(old+new) structure, inode write (sb, @@ -3006,8 +2995,8 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr) inode->i_uid = attr->ia_uid; if (attr->ia_valid & ATTR_GID) inode->i_gid = attr->ia_gid; - if ((attr->ia_valid & ATTR_XID) && IS_TAGXID(inode)) - inode->i_xid = attr->ia_xid; + if ((attr->ia_valid & ATTR_TAG) && IS_TAGGED(inode)) + inode->i_tag = attr->ia_tag; error = ext3_mark_inode_dirty(handle, inode); ext3_journal_stop(handle); } @@ -3114,13 +3103,13 @@ int ext3_mark_iloc_dirty(handle_t *handle, return err; } -/* +/* * On success, We end up with an outstanding reference count against - * iloc->bh. This _must_ be cleaned up later. + * iloc->bh. This _must_ be cleaned up later. */ int -ext3_reserve_inode_write(handle_t *handle, struct inode *inode, +ext3_reserve_inode_write(handle_t *handle, struct inode *inode, struct ext3_iloc *iloc) { int err = 0; @@ -3210,7 +3199,7 @@ out: } #if 0 -/* +/* * Bind an inode's backing buffer_head into this transaction, to prevent * it from being flushed to disk early. Unlike * ext3_reserve_inode_write, this leaves behind no bh reference and @@ -3228,7 +3217,7 @@ static int ext3_pin_inode(handle_t *handle, struct inode *inode) BUFFER_TRACE(iloc.bh, "get_write_access"); err = journal_get_write_access(handle, iloc.bh); if (!err) - err = ext3_journal_dirty_metadata(handle, + err = ext3_journal_dirty_metadata(handle, iloc.bh); brelse(iloc.bh); }