Add e2fsck -D option which optimizes/compresses directories.

General cleanup of rehash code.  (Fixed a few bugs, reformatted
code to make it more maintainable, etc.)

1 files changed, 284 insertions, 169 deletions

diff --git a/e2fsck/rehash.c b/e2fsck/rehash.c
index 95bb6785..2c64ec4a 100644
--- a/e2fsck/rehash.c
+++ b/e2fsck/rehash.c
@@ -57,6 +57,7 @@ struct fill_dir_struct {
 	struct hash_entry *harray;
 	int max_array, num_array;
 	int dir_size;
+	int compress;
 	ino_t parent;
 };
 
@@ -115,14 +116,16 @@ static int fill_dir_block(ext2_filsys fs,
 			fd->err = EXT2_ET_DIR_CORRUPTED;
 			return BLOCK_ABORT;
 		}
+		dir_offset += dirent->rec_len;
 		if (dirent->inode == 0)
-			goto next;
-		if (((dirent->name_len&0xFF) == 1) && (dirent->name[0] == '.'))
-			goto next;
-		if (((dirent->name_len&0xFF) == 2) &&
+			continue;
+		if (!fd->compress && ((dirent->name_len&0xFF) == 1) &&
+		    (dirent->name[0] == '.'))
+			continue;
+		if (!fd->compress && ((dirent->name_len&0xFF) == 2) &&
 		    (dirent->name[0] == '.') && (dirent->name[1] == '.')) {
 			fd->parent = dirent->inode;
-			goto next;
+			continue;
 		}
 		if (fd->num_array >= fd->max_array) {
 			new_array = realloc(fd->harray,
@@ -134,19 +137,20 @@ static int fill_dir_block(ext2_filsys fs,
 			fd->harray = new_array;
 			fd->max_array += 500;
 		}
-		ent = fd->harray + fd->num_array;
+		ent = fd->harray + fd->num_array++;
 		ent->dir = dirent;
-		fd->err = ext2fs_dirhash(fs->super->s_def_hash_version,
-					 dirent->name,
-					 dirent->name_len & 0xFF,
-					 fs->super->s_hash_seed,
-					 &ent->hash, &ent->minor_hash);
-		if (fd->err)
-			return BLOCK_ABORT;
-		fd->num_array++;
 		fd->dir_size += EXT2_DIR_REC_LEN(dirent->name_len & 0xFF);
-	next:
-		dir_offset += dirent->rec_len;
+		if (fd->compress)
+			ent->hash = ent->minor_hash = 0;
+		else {
+			fd->err = ext2fs_dirhash(fs->super->s_def_hash_version,
+						 dirent->name,
+						 dirent->name_len & 0xFF,
+						 fs->super->s_hash_seed,
+						 &ent->hash, &ent->minor_hash);
+			if (fd->err)
+				return BLOCK_ABORT;
+		}
 	}
 	
 	return 0;
@@ -174,6 +178,30 @@ static EXT2_QSORT_TYPE hash_cmp(const void *a, const void *b)
 	return ret;
 }
 
+/* Used for sorting the hash entry */
+static EXT2_QSORT_TYPE name_cmp(const void *a, const void *b)
+{
+	const struct hash_entry *he_a = (const struct hash_entry *) a;
+	const struct hash_entry *he_b = (const struct hash_entry *) b;
+	int	ret;
+	int	min_len;
+
+	min_len = he_a->dir->name_len;
+	if (min_len > he_b->dir->name_len)
+		min_len = he_b->dir->name_len;
+
+	ret = strncmp(he_a->dir->name, he_b->dir->name, min_len);
+	if (ret == 0) {
+		if (he_a->dir->name_len > he_b->dir->name_len)
+			ret = 1;
+		else if (he_a->dir->name_len < he_b->dir->name_len)
+			ret = -1;
+		else
+			ret = 0;
+	}
+	return ret;
+}
+
 static errcode_t alloc_size_dir(ext2_filsys fs, struct out_dir *outdir, 
 				int blocks)
 {
@@ -206,7 +234,7 @@ static void free_out_dir(struct out_dir *outdir)
 	outdir->num =0;
 }
 
-errcode_t get_next_block(ext2_filsys fs, struct out_dir *outdir,
+static errcode_t get_next_block(ext2_filsys fs, struct out_dir *outdir,
 			 char ** ret)
 {
 	errcode_t	retval;
@@ -217,17 +245,87 @@ errcode_t get_next_block(ext2_filsys fs, struct out_dir *outdir,
 			return retval;
 	}
 	*ret = outdir->buf + (outdir->num++ * fs->blocksize);
+	memset(*ret, 0, fs->blocksize);
 	return 0;
 }
 
 
-struct ext2_dx_root_info *set_root_node(ext2_filsys fs, char *buf,
+static errcode_t copy_dir_entries(ext2_filsys fs,
+				  struct fill_dir_struct *fd,
+				  struct out_dir *outdir)
+{
+	errcode_t		retval;
+	char			*block_start;
+	struct hash_entry 	*ent;
+	struct ext2_dir_entry	*dirent;
+	int			i, rec_len, left;
+	ext2_dirhash_t		prev_hash;
+	int			offset;
+	
+	outdir->max = 0;
+	retval = alloc_size_dir(fs, outdir,
+				(fd->dir_size / fs->blocksize) + 2);
+	if (retval)
+		return retval;
+	outdir->num = fd->compress ? 0 : 1;
+	offset = 0;
+	outdir->hashes[0] = 0;
+	prev_hash = 1;
+	if ((retval = get_next_block(fs, outdir, &block_start)))
+		return retval;
+	dirent = (struct ext2_dir_entry *) block_start;
+	left = fs->blocksize;
+	for (i=0; i < fd->num_array; i++) {
+		ent = fd->harray + i;
+		rec_len = EXT2_DIR_REC_LEN(ent->dir->name_len & 0xFF);
+		dirent = (struct ext2_dir_entry *) (block_start + offset);
+		left = fs->blocksize - offset;
+		if (rec_len > left) {
+			if (left) {
+				dirent->rec_len = left;
+				dirent->inode = 0;
+				dirent->name_len = 0;
+				offset += left;
+				left = 0;
+			}
+			if ((retval = get_next_block(fs, outdir,
+						      &block_start)))
+				return retval;
+			offset = 0; left = fs->blocksize;
+			dirent = (struct ext2_dir_entry *) block_start;
+		}
+		if (offset == 0) {
+			if (ent->hash == prev_hash)
+				outdir->hashes[outdir->num-1] = ent->hash | 1;
+			else
+				outdir->hashes[outdir->num-1] = ent->hash;
+		}
+		dirent->inode = ent->dir->inode;
+		dirent->name_len = ent->dir->name_len;
+		dirent->rec_len = rec_len;
+		memcpy(dirent->name, ent->dir->name, dirent->name_len & 0xFF);
+		offset += rec_len;
+		left -= rec_len;
+		if (left < 12) {
+			dirent->rec_len += left;
+			offset += left;
+		}
+		prev_hash = ent->hash;
+	}
+	if (left)
+		dirent->rec_len += left;
+
+	return 0;
+}
+
+
+static struct ext2_dx_root_info *set_root_node(ext2_filsys fs, char *buf,
 				    ext2_ino_t ino, ext2_ino_t parent)
 {
 	struct ext2_dir_entry 		*dir;
 	struct ext2_dx_root_info  	*root;
 	struct ext2_dx_countlimit	*limits;
-	int				filetype;
+	int				filetype = 0;
 
 	if (fs->super->s_feature_incompat & EXT2_FEATURE_INCOMPAT_FILETYPE)
 		filetype = EXT2_FT_DIR << 8;
@@ -260,7 +358,7 @@ struct ext2_dx_root_info *set_root_node(ext2_filsys fs, char *buf,
 }
 
 
-struct ext2_dx_entry *set_int_node(ext2_filsys fs, char *buf)
+static struct ext2_dx_entry *set_int_node(ext2_filsys fs, char *buf)
 {
 	struct ext2_dir_entry 		*dir;
 	struct ext2_dx_countlimit	*limits;
@@ -277,6 +375,84 @@ struct ext2_dx_entry *set_int_node(ext2_filsys fs, char *buf)
 	return (struct ext2_dx_entry *) limits;
 }
 
+/*
+ * This function takes the leaf nodes which have been written in
+ * outdir, and populates the root node and any necessary interior nodes.
+ */
+static errcode_t calculate_tree(ext2_filsys fs,
+				struct out_dir *outdir,
+				ext2_ino_t ino,
+				ext2_ino_t parent)
+{
+	struct ext2_dx_root_info  	*root_info;
+	struct ext2_dx_entry 		*root, *dx_ent = 0;
+	struct ext2_dx_countlimit	*root_limit, *limit;
+	errcode_t			retval;
+	char				* block_start;
+	int				i, c1, c2, nblks;
+	int				limit_offset, root_offset;
+	
+	root_info = set_root_node(fs, outdir->buf, ino, parent);
+	root_offset = limit_offset = ((char *) root_info - outdir->buf) +
+		root_info->info_length;
+	root_limit = (struct ext2_dx_countlimit *) (outdir->buf + limit_offset);
+	c1 = root_limit->limit;
+	nblks = outdir->num;
+
+	/* Write out the pointer blocks */
+	if (nblks-1 <= c1) {
+		/* Just write out the root block, and we're done */
+		root = (struct ext2_dx_entry *) (outdir->buf + root_offset);
+		for (i=1; i < nblks; i++) {
+			root->block = ext2fs_cpu_to_le32(i);
+			if (i != 1)
+				root->hash =
+					ext2fs_cpu_to_le32(outdir->hashes[i]);
+			root++;
+			c1--;
+		}
+	} else {
+		c2 = 0;
+		limit = 0;
+		root_info->indirect_levels = 1;
+		for (i=1; i < nblks; i++) {
+			if (c1 == 0)
+				return ENOSPC;
+			if (c2 == 0) {
+				if (limit)
+					limit->limit = limit->count = 
+		ext2fs_cpu_to_le16(limit->limit);
+				root = (struct ext2_dx_entry *)
+					(outdir->buf + root_offset);
+				root->block = ext2fs_cpu_to_le32(outdir->num);
+				if (i != 1)
+					root->hash =
+			ext2fs_cpu_to_le32(outdir->hashes[i]);
+				if ((retval =  get_next_block(fs, outdir,
+							      &block_start)))
+					return retval;
+				dx_ent = set_int_node(fs, block_start);
+				limit = (struct ext2_dx_countlimit *) dx_ent;
+				c2 = limit->limit;
+				root_offset += sizeof(struct ext2_dx_entry);
+				c1--;
+			}
+			dx_ent->block = ext2fs_cpu_to_le32(i);
+			if (c2 != limit->limit)
+				dx_ent->hash =
+					ext2fs_cpu_to_le32(outdir->hashes[i]);
+			dx_ent++;
+			c2--;
+		}
+		limit->count = ext2fs_cpu_to_le16(limit->limit - c2);
+		limit->limit = ext2fs_cpu_to_le16(limit->limit);
+	}
+	root_limit = (struct ext2_dx_countlimit *) (outdir->buf + limit_offset);
+	root_limit->count = ext2fs_cpu_to_le16(root_limit->limit - c1);
+	root_limit->limit = ext2fs_cpu_to_le16(root_limit->limit);
+
+	return 0;
+}
 
 struct write_dir_struct {
 	struct out_dir *outdir;
@@ -286,8 +462,7 @@ struct write_dir_struct {
 };
 
 /*
- * This makes sure we have enough space to write out the modified
- * directory.
+ * Helper function which writes out a directory block.
  */
 static int write_dir_block(ext2_filsys fs,
 			   blk_t	*block_nr,
@@ -297,9 +472,8 @@ static int write_dir_block(ext2_filsys fs,
 			   void *priv_data)
 {
 	struct write_dir_struct	*wd = (struct write_dir_struct *) priv_data;
-	blk_t	new_blk, blk;
+	blk_t	blk;
 	char	*dir;
-	errcode_t	retval;
 
 	if (*block_nr == 0)
 		return 0;
@@ -322,9 +496,9 @@ static int write_dir_block(ext2_filsys fs,
 	return 0;
 }
 
-
 static errcode_t write_directory(e2fsck_t ctx, ext2_filsys fs,
-				 struct out_dir *outdir, ext2_ino_t ino)
+				 struct out_dir *outdir,
+				 ext2_ino_t ino, int compress)
 {
 	struct write_dir_struct wd;
 	errcode_t	retval;
@@ -347,7 +521,8 @@ static errcode_t write_directory(e2fsck_t ctx, ext2_filsys fs,
 		return wd.err;
 
 	e2fsck_read_inode(ctx, ino, &inode, "rehash_dir");
-	inode.i_flags |= EXT2_INDEX_FL;
+	if (!compress)
+		inode.i_flags |= EXT2_INDEX_FL;
 	inode.i_size = outdir->num * fs->blocksize;
 	inode.i_blocks -= (fs->blocksize / 512) * wd.cleared;
 	e2fsck_write_inode(ctx, ino, &inode, "rehash_dir");
@@ -360,22 +535,11 @@ errcode_t e2fsck_rehash_dir(e2fsck_t ctx, ext2_ino_t ino)
 	ext2_filsys 		fs = ctx->fs;
 	errcode_t		retval;
 	struct ext2_inode 	inode;
-	char			*dir_buf = 0, *block_start;
-	struct hash_entry 	*ent;
-	struct ext2_dir_entry	*dirent;
+	char			*dir_buf = 0;
 	struct fill_dir_struct	fd;
-	int			i, rec_len, left, c1, c2, nblks;
-	ext2_dirhash_t		prev_hash;
-	int			offset;
-	void			*new_mem;
 	struct out_dir		outdir;
-	struct ext2_dx_root_info  	*root_info;
-	struct ext2_dx_entry 	*root, *dx_ent;
-	struct ext2_dx_countlimit	*root_limit, *limit;
 	
 	e2fsck_read_inode(ctx, ino, &inode, "rehash_dir");
-	if ((inode.i_size / fs->blocksize) < 3)
-		return 0;
 
 	retval = ENOMEM;
 	fd.harray = 0;
@@ -394,6 +558,10 @@ errcode_t e2fsck_rehash_dir(e2fsck_t ctx, ext2_ino_t ino)
 	fd.inode = &inode;
 	fd.err = 0;
 	fd.dir_size = 0;
+	fd.compress = 0;
+	if (!(fs->super->s_feature_compat & EXT2_FEATURE_COMPAT_DIR_INDEX) ||
+	    (inode.i_size / fs->blocksize) < 3)
+		fd.compress = 1;
 	fd.parent = 0;
 
 	/* Read in the entire directory into memory */
@@ -409,170 +577,117 @@ errcode_t e2fsck_rehash_dir(e2fsck_t ctx, ext2_ino_t ino)
 	       fd.num_array, fd.dir_size, ino);
 #endif
 
-	/* Sort the list into hash order */
-	qsort(fd.harray, fd.num_array, sizeof(struct hash_entry), hash_cmp);
-
-	outdir.max = 0;
-	retval = alloc_size_dir(fs, &outdir,
-				(fd.dir_size / fs->blocksize) + 2);
+	/* Sort the list */
+	if (fd.compress)
+		qsort(fd.harray+2, fd.num_array-2,
+		      sizeof(struct hash_entry), name_cmp);
+	else
+		qsort(fd.harray, fd.num_array,
+		      sizeof(struct hash_entry), hash_cmp);
+
+	/*
+	 * Copy the directory entries.  In a htree directory these
+	 * will become the leaf nodes.
+	 */
+	retval = copy_dir_entries(fs, &fd, &outdir);
 	if (retval)
 		goto errout;
-	outdir.num = 1; offset = 0;
-	outdir.hashes[0] = 0;
-	prev_hash = 1;
-	if ((retval = get_next_block(fs, &outdir, &block_start)))
-		goto errout;
-	for (i=0; i < fd.num_array; i++) {
-		ent = fd.harray + i;
-		rec_len = EXT2_DIR_REC_LEN(ent->dir->name_len & 0xFF);
-		dirent = (struct ext2_dir_entry *) (block_start + offset);
-		left = fs->blocksize - offset;
-		if (rec_len > left) {
-			if (left) {
-				dirent->rec_len = left;
-				dirent->inode = 0;
-				dirent->name_len = 0;
-				offset += left;
-				left = 0;
-			}
-			if ((retval =  get_next_block(fs, &outdir,
-						      &block_start)))
-				goto errout;
-			offset = 0; left = fs->blocksize;
-			dirent = (struct ext2_dir_entry *) block_start;
-		}
-		if (offset == 0) {
-			if (ent->hash == prev_hash)
-				outdir.hashes[outdir.num-1] = ent->hash | 1;
-			else
-				outdir.hashes[outdir.num-1] = ent->hash;
-		}
-		dirent->inode = ent->dir->inode;
-		dirent->name_len = ent->dir->name_len;
-		dirent->rec_len = rec_len;
-		memcpy(dirent->name, ent->dir->name, dirent->name_len & 0xFF);
-		offset += rec_len;
-		left -= rec_len;
-		if (left < 12) {
-			dirent->rec_len += left;
-			offset += left;
-		}
-		prev_hash = ent->hash;
-	}
-	if (left)
-		dirent->rec_len += left;
+	
 	free(dir_buf); dir_buf = 0;
 
-	root_info = set_root_node(fs, outdir.buf, ino, fd.parent);
-	root_limit = (struct ext2_dx_countlimit *)
-		((char *)root_info + root_info->info_length);
-	root = (struct ext2_dx_entry *) root_limit;
-	c1 = root_limit->limit;
-	nblks = outdir.num;
-
-	/* Write out the pointer blocks */
-	if (nblks-1 <= c1) {
-		/* Just write out the root block, and we're done */
-		for (i=1; i < nblks; i++) {
-			root->block = ext2fs_cpu_to_le32(i);
-			if (i != 1)
-				root->hash =
-					ext2fs_cpu_to_le32(outdir.hashes[i]);
-			root++;
-			c1--;
-		}
-	} else {
-		c2 = 0;
-		limit = 0;
-		root_info->indirect_levels = 1;
-		for (i=1; i < nblks; i++) {
-			if (c1 == 0) {
-				retval = ENOSPC;
-				goto errout;
-			}
-			if (c2 == 0) {
-				if (limit)
-					limit->limit = limit->count = 
-		ext2fs_cpu_to_le16(limit->limit);
-				root->block = ext2fs_cpu_to_le32(outdir.num);
-				if (i != 1)
-					root->hash =
-			ext2fs_cpu_to_le32(outdir.hashes[i]);
-				if ((retval =  get_next_block(fs, &outdir,
-							      &block_start)))
-					goto errout;
-				dx_ent = set_int_node(fs, block_start);
-				limit = (struct ext2_dx_countlimit *) dx_ent;
-				c2 = limit->limit;
-				root++;
-				c1--;
-			}
-			dx_ent->block = ext2fs_cpu_to_le32(i);
-			if (c2 != limit->limit)
-				dx_ent->hash =
-					ext2fs_cpu_to_le32(outdir.hashes[i]);
-			dx_ent++;
-			c2--;
-		}
-		limit->count = ext2fs_cpu_to_le16(limit->limit - c2);
-		limit->limit = ext2fs_cpu_to_le16(limit->limit);
+	if (!fd.compress) {
+		/* Calculate the interior nodes */
+		retval = calculate_tree(fs, &outdir, ino, fd.parent);
+		if (retval)
+			goto errout;
 	}
-	root_limit->count = ext2fs_cpu_to_le16(root_limit->limit - c1);
-	root_limit->limit = ext2fs_cpu_to_le16(root_limit->limit);
-
-	retval = write_directory(ctx, fs, &outdir, ino);
+	
+	retval = write_directory(ctx, fs, &outdir, ino, fd.compress);
 	if (retval)
 		goto errout;
 
 errout:
 	if (dir_buf)
 		free(dir_buf);
+	if (fd.harray)
+		free(fd.harray);
+
 	free_out_dir(&outdir);
 	return retval;
 }
 
 void e2fsck_rehash_directories(e2fsck_t ctx)
 {
-	errcode_t	retval;
 	struct problem_context	pctx;
-	ext2_u32_iterate iter;
-	ext2_ino_t	ino,lpf;
-	static const char name[] = "lost+found";
-	int	first = 1;
+#ifdef RESOURCE_TRACK
+	struct resource_track	rtrack;
+#endif
+	struct dir_info		*dir;
+	ext2_u32_iterate 	iter;
+	ext2_ino_t		ino;
+	errcode_t		retval;
+	int			i, all_dirs, dir_index, first = 1;
+
+#ifdef RESOURCE_TRACK
+	init_resource_track(&rtrack);
+#endif
 
-	if (!ctx->dirs_to_hash)
+	all_dirs = ctx->options & E2F_OPT_COMPRESS_DIRS;
+
+	if (!ctx->dirs_to_hash && !all_dirs)
 		return;
 
-	retval = ext2fs_lookup(ctx->fs, EXT2_ROOT_INO, name,
-			       sizeof(name)-1, 0, &lpf);
-	if (retval)
-		lpf = 0;
+	e2fsck_get_lost_and_found(ctx, 0);
 		
 	clear_problem_context(&pctx);
-	retval = ext2fs_u32_list_iterate_begin(ctx->dirs_to_hash, &iter);
-	if (retval) {
-		pctx.errcode = retval;
-		fix_problem(ctx, PR_3A_REHASH_ITER, &pctx);
-		return;
+
+	dir_index = ctx->fs->super->s_feature_compat & EXT2_FEATURE_COMPAT_DIR_INDEX;
+	if (all_dirs)
+		i = 0;
+	else {
+		retval = ext2fs_u32_list_iterate_begin(ctx->dirs_to_hash, 
+						       &iter);
+		if (retval) {
+			pctx.errcode = retval;
+			fix_problem(ctx, PR_3A_OPTIMIZE_ITER, &pctx);
+			return;
+		}
 	}
-	while (ext2fs_u32_list_iterate(iter, &ino)) {
-		if (ino == lpf)
+	while (1) {
+		if (all_dirs) {
+			if ((dir = e2fsck_dir_info_iter(ctx, &i)) == 0)
+				break;
+			ino = dir->ino;
+		} else {
+			if (!ext2fs_u32_list_iterate(iter, &ino))
+				break;
+		}
+		if (ino == ctx->lost_and_found)
 			continue;
 		pctx.dir = ino;
 		if (first) {
 			fix_problem(ctx, PR_3A_PASS_HEADER, &pctx);
 			first = 0;
 		}
-		fix_problem(ctx, PR_3A_REHASH_DIR, &pctx);
+		fix_problem(ctx, PR_3A_OPTIMIZE_DIR, &pctx);
 		pctx.errcode = e2fsck_rehash_dir(ctx, ino);
 		if (pctx.errcode) {
-			end_problem_latch(ctx, PR_LATCH_REHASH_DIR);
-			fix_problem(ctx, PR_3A_REHASH_DIR_ERR, &pctx);
+			end_problem_latch(ctx, PR_LATCH_OPTIMIZE_DIR);
+			fix_problem(ctx, PR_3A_OPTIMIZE_DIR_ERR, &pctx);
 		}
 	}
-	end_problem_latch(ctx, PR_LATCH_REHASH_DIR);
-	ext2fs_u32_list_iterate_end(iter);
+	end_problem_latch(ctx, PR_LATCH_OPTIMIZE_DIR);
+	if (!all_dirs)
+		ext2fs_u32_list_iterate_end(iter);
 	
-	ext2fs_u32_list_free(ctx->dirs_to_hash);
+	if (ctx->dirs_to_hash)
+		ext2fs_u32_list_free(ctx->dirs_to_hash);
 	ctx->dirs_to_hash = 0;
+
+#ifdef RESOURCE_TRACK
+	if (ctx->options & E2F_OPT_TIME2) {
+		e2fsck_clear_progbar(ctx);
+		print_resource_track("Pass 3A", &rtrack);
+	}
+#endif
 }