Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 fixes from Ted Ts'o:
 "Ext4 bug fixes.

  We also reserved code points for encryption and read-only images (for
  which the implementation is mostly just the reserved code point for a
  read-only feature :-)"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: fix indirect punch hole corruption
  ext4: ignore journal checksum on remount; don't fail
  ext4: remove duplicate remount check for JOURNAL_CHECKSUM change
  ext4: fix mmap data corruption in nodelalloc mode when blocksize < pagesize
  ext4: support read-only images
  ext4: change to use setup_timer() instead of init_timer()
  ext4: reserve codepoints used by the ext4 encryption feature
  jbd2: complain about descriptor block checksum errors

Author: torvalds

fs/ext4/ext4.h

@@ -364,7 +364,8 @@ struct flex_groups {
 #define EXT4_DIRTY_FL			0x00000100
 #define EXT4_COMPRBLK_FL		0x00000200 /* One or more compressed clusters */
 #define EXT4_NOCOMPR_FL			0x00000400 /* Don't compress */
-#define EXT4_ECOMPR_FL			0x00000800 /* Compression error */
+	/* nb: was previously EXT2_ECOMPR_FL */
+#define EXT4_ENCRYPT_FL			0x00000800 /* encrypted file */
 /* End compression flags --- maybe not all used */
 #define EXT4_INDEX_FL			0x00001000 /* hash-indexed directory */
 #define EXT4_IMAGIC_FL			0x00002000 /* AFS directory */
@@ -421,7 +422,7 @@ enum {
 	EXT4_INODE_DIRTY	= 8,
 	EXT4_INODE_COMPRBLK	= 9,	/* One or more compressed clusters */
 	EXT4_INODE_NOCOMPR	= 10,	/* Don't compress */
-	EXT4_INODE_ECOMPR	= 11,	/* Compression error */
+	EXT4_INODE_ENCRYPT	= 11,	/* Compression error */
 /* End compression flags --- maybe not all used */
 	EXT4_INODE_INDEX	= 12,	/* hash-indexed directory */
 	EXT4_INODE_IMAGIC	= 13,	/* AFS directory */
@@ -466,7 +467,7 @@ static inline void ext4_check_flag_values(void)
 	CHECK_FLAG_VALUE(DIRTY);
 	CHECK_FLAG_VALUE(COMPRBLK);
 	CHECK_FLAG_VALUE(NOCOMPR);
-	CHECK_FLAG_VALUE(ECOMPR);
+	CHECK_FLAG_VALUE(ENCRYPT);
 	CHECK_FLAG_VALUE(INDEX);
 	CHECK_FLAG_VALUE(IMAGIC);
 	CHECK_FLAG_VALUE(JOURNAL_DATA);
@@ -1048,6 +1049,12 @@ extern void ext4_set_bits(void *bm, int cur, int len);
 /* Metadata checksum algorithm codes */
 #define EXT4_CRC32C_CHKSUM		1
 
+/* Encryption algorithms */
+#define EXT4_ENCRYPTION_MODE_INVALID		0
+#define EXT4_ENCRYPTION_MODE_AES_256_XTS	1
+#define EXT4_ENCRYPTION_MODE_AES_256_GCM	2
+#define EXT4_ENCRYPTION_MODE_AES_256_CBC	3
+
 /*
  * Structure of the super block
  */
@@ -1161,7 +1168,8 @@ struct ext4_super_block {
 	__le32	s_grp_quota_inum;	/* inode for tracking group quota */
 	__le32	s_overhead_clusters;	/* overhead blocks/clusters in fs */
 	__le32	s_backup_bgs[2];	/* groups with sparse_super2 SBs */
-	__le32	s_reserved[106];	/* Padding to the end of the block */
+	__u8	s_encrypt_algos[4];	/* Encryption algorithms in use  */
+	__le32	s_reserved[105];	/* Padding to the end of the block */
 	__le32	s_checksum;		/* crc32c(superblock) */
 };
 
@@ -1527,6 +1535,7 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei)
  * GDT_CSUM bits are mutually exclusive.
  */
 #define EXT4_FEATURE_RO_COMPAT_METADATA_CSUM	0x0400
+#define EXT4_FEATURE_RO_COMPAT_READONLY		0x1000
 
 #define EXT4_FEATURE_INCOMPAT_COMPRESSION	0x0001
 #define EXT4_FEATURE_INCOMPAT_FILETYPE		0x0002
@@ -1542,6 +1551,7 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei)
 #define EXT4_FEATURE_INCOMPAT_BG_USE_META_CSUM	0x2000 /* use crc32c for bg */
 #define EXT4_FEATURE_INCOMPAT_LARGEDIR		0x4000 /* >2GB or 3-lvl htree */
 #define EXT4_FEATURE_INCOMPAT_INLINE_DATA	0x8000 /* data in inode */
+#define EXT4_FEATURE_INCOMPAT_ENCRYPT		0x10000
 
 #define EXT2_FEATURE_COMPAT_SUPP	EXT4_FEATURE_COMPAT_EXT_ATTR
 #define EXT2_FEATURE_INCOMPAT_SUPP	(EXT4_FEATURE_INCOMPAT_FILETYPE| \

fs/ext4/indirect.c

@@ -1401,10 +1401,7 @@ int ext4_ind_remove_space(handle_t *handle, struct inode *inode,
 				 * to free. Everything was covered by the start
 				 * of the range.
 				 */
-				return 0;
-			} else {
-				/* Shared branch grows from an indirect block */
-				partial2--;
+				goto do_indirects;
 			}
 		} else {
 			/*
@@ -1435,56 +1432,96 @@ int ext4_ind_remove_space(handle_t *handle, struct inode *inode,
 	/* Punch happened within the same level (n == n2) */
 	partial = ext4_find_shared(inode, n, offsets, chain, &nr);
 	partial2 = ext4_find_shared(inode, n2, offsets2, chain2, &nr2);
-	/*
-	 * ext4_find_shared returns Indirect structure which
-	 * points to the last element which should not be
-	 * removed by truncate. But this is end of the range
-	 * in punch_hole so we need to point to the next element
-	 */
-	partial2->p++;
-	while ((partial > chain) || (partial2 > chain2)) {
-		/* We're at the same block, so we're almost finished */
-		if ((partial->bh && partial2->bh) &&
-		    (partial->bh->b_blocknr == partial2->bh->b_blocknr)) {
-			if ((partial > chain) && (partial2 > chain2)) {
+
+	/* Free top, but only if partial2 isn't its subtree. */
+	if (nr) {
+		int level = min(partial - chain, partial2 - chain2);
+		int i;
+		int subtree = 1;
+
+		for (i = 0; i <= level; i++) {
+			if (offsets[i] != offsets2[i]) {
+				subtree = 0;
+				break;
+			}
+		}
+
+		if (!subtree) {
+			if (partial == chain) {
+				/* Shared branch grows from the inode */
+				ext4_free_branches(handle, inode, NULL,
+						   &nr, &nr+1,
+						   (chain+n-1) - partial);
+				*partial->p = 0;
+			} else {
+				/* Shared branch grows from an indirect block */
+				BUFFER_TRACE(partial->bh, "get_write_access");
 				ext4_free_branches(handle, inode, partial->bh,
-						   partial->p + 1,
-						   partial2->p,
+						   partial->p,
+						   partial->p+1,
 						   (chain+n-1) - partial);
-				BUFFER_TRACE(partial->bh, "call brelse");
-				brelse(partial->bh);
-				BUFFER_TRACE(partial2->bh, "call brelse");
-				brelse(partial2->bh);
 			}
-			return 0;
 		}
+	}
+
+	if (!nr2) {
 		/*
-		 * Clear the ends of indirect blocks on the shared branch
-		 * at the start of the range
+		 * ext4_find_shared returns Indirect structure which
+		 * points to the last element which should not be
+		 * removed by truncate. But this is end of the range
+		 * in punch_hole so we need to point to the next element
 		 */
-		if (partial > chain) {
+		partial2->p++;
+	}
+
+	while (partial > chain || partial2 > chain2) {
+		int depth = (chain+n-1) - partial;
+		int depth2 = (chain2+n2-1) - partial2;
+
+		if (partial > chain && partial2 > chain2 &&
+		    partial->bh->b_blocknr == partial2->bh->b_blocknr) {
+			/*
+			 * We've converged on the same block. Clear the range,
+			 * then we're done.
+			 */
 			ext4_free_branches(handle, inode, partial->bh,
-				   partial->p + 1,
-				   (__le32 *)partial->bh->b_data+addr_per_block,
-				   (chain+n-1) - partial);
+					   partial->p + 1,
+					   partial2->p,
+					   (chain+n-1) - partial);
 			BUFFER_TRACE(partial->bh, "call brelse");
 			brelse(partial->bh);
-			partial--;
+			BUFFER_TRACE(partial2->bh, "call brelse");
+			brelse(partial2->bh);
+			return 0;
 		}
+
 		/*
-		 * Clear the ends of indirect blocks on the shared branch
-		 * at the end of the range
+		 * The start and end partial branches may not be at the same
+		 * level even though the punch happened within one level. So, we
+		 * give them a chance to arrive at the same level, then walk
+		 * them in step with each other until we converge on the same
+		 * block.
 		 */
-		if (partial2 > chain2) {
+		if (partial > chain && depth <= depth2) {
+			ext4_free_branches(handle, inode, partial->bh,
+					   partial->p + 1,
+					   (__le32 *)partial->bh->b_data+addr_per_block,
+					   (chain+n-1) - partial);
+			BUFFER_TRACE(partial->bh, "call brelse");
+			brelse(partial->bh);
+			partial--;
+		}
+		if (partial2 > chain2 && depth2 <= depth) {
 			ext4_free_branches(handle, inode, partial2->bh,
 					   (__le32 *)partial2->bh->b_data,
 					   partial2->p,
-					   (chain2+n-1) - partial2);
+					   (chain2+n2-1) - partial2);
 			BUFFER_TRACE(partial2->bh, "call brelse");
 			brelse(partial2->bh);
 			partial2--;
 		}
 	}
+	return 0;
 
 do_indirects:
 	/* Kill the remaining (whole) subtrees */

fs/ext4/inode.c

@@ -1024,6 +1024,7 @@ static int ext4_write_end(struct file *file,
 {
 	handle_t *handle = ext4_journal_current_handle();
 	struct inode *inode = mapping->host;
+	loff_t old_size = inode->i_size;
 	int ret = 0, ret2;
 	int i_size_changed = 0;
 
@@ -1054,6 +1055,8 @@ static int ext4_write_end(struct file *file,
 	unlock_page(page);
 	page_cache_release(page);
 
+	if (old_size < pos)
+		pagecache_isize_extended(inode, old_size, pos);
 	/*
 	 * Don't mark the inode dirty under page lock. First, it unnecessarily
 	 * makes the holding time of page lock longer. Second, it forces lock
@@ -1095,6 +1098,7 @@ static int ext4_journalled_write_end(struct file *file,
 {
 	handle_t *handle = ext4_journal_current_handle();
 	struct inode *inode = mapping->host;
+	loff_t old_size = inode->i_size;
 	int ret = 0, ret2;
 	int partial = 0;
 	unsigned from, to;
@@ -1127,6 +1131,9 @@ static int ext4_journalled_write_end(struct file *file,
 	unlock_page(page);
 	page_cache_release(page);
 
+	if (old_size < pos)
+		pagecache_isize_extended(inode, old_size, pos);
+
 	if (size_changed) {
 		ret2 = ext4_mark_inode_dirty(handle, inode);
 		if (!ret)

fs/ext4/super.c

@@ -2779,6 +2779,12 @@ static int ext4_feature_set_ok(struct super_block *sb, int readonly)
 	if (readonly)
 		return 1;
 
+	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_READONLY)) {
+		ext4_msg(sb, KERN_INFO, "filesystem is read-only");
+		sb->s_flags |= MS_RDONLY;
+		return 1;
+	}
+
 	/* Check that feature set is OK for a read-write mount */
 	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT4_FEATURE_RO_COMPAT_SUPP)) {
 		ext4_msg(sb, KERN_ERR, "couldn't mount RDWR because of "
@@ -3936,9 +3942,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
 	spin_lock_init(&sbi->s_next_gen_lock);
 
-	init_timer(&sbi->s_err_report);
-	sbi->s_err_report.function = print_daily_error_info;
-	sbi->s_err_report.data = (unsigned long) sb;
+	setup_timer(&sbi->s_err_report, print_daily_error_info,
+		(unsigned long) sb);
 
 	/* Register extent status tree shrinker */
 	if (ext4_es_register_shrinker(sbi))
@@ -4866,9 +4871,6 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 	if (sbi->s_journal && sbi->s_journal->j_task->io_context)
 		journal_ioprio = sbi->s_journal->j_task->io_context->ioprio;
 
-	/*
-	 * Allow the "check" option to be passed as a remount option.
-	 */
 	if (!parse_options(data, sb, NULL, &journal_ioprio, 1)) {
 		err = -EINVAL;
 		goto restore_opts;
@@ -4877,17 +4879,8 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
 	    test_opt(sb, JOURNAL_CHECKSUM)) {
 		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
-			 "during remount not supported");
-		err = -EINVAL;
-		goto restore_opts;
-	}
-
-	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
-	    test_opt(sb, JOURNAL_CHECKSUM)) {
-		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
-			 "during remount not supported");
-		err = -EINVAL;
-		goto restore_opts;
+			 "during remount not supported; ignoring");
+		sbi->s_mount_opt ^= EXT4_MOUNT_JOURNAL_CHECKSUM;
 	}
 
 	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) {
@@ -4963,7 +4956,9 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 				ext4_mark_recovery_complete(sb, es);
 		} else {
 			/* Make sure we can mount this feature set readwrite */
-			if (!ext4_feature_set_ok(sb, 0)) {
+			if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
+					EXT4_FEATURE_RO_COMPAT_READONLY) ||
+			    !ext4_feature_set_ok(sb, 0)) {
 				err = -EROFS;
 				goto restore_opts;
 			}

fs/jbd2/recovery.c

@@ -524,6 +524,9 @@ static int do_one_pass(journal_t *journal,
 			if (descr_csum_size > 0 &&
 			    !jbd2_descr_block_csum_verify(journal,
 							  bh->b_data)) {
+				printk(KERN_ERR "JBD2: Invalid checksum "
+				       "recovering block %lu in log\n",
+				       next_log_block);
 				err = -EIO;
 				brelse(bh);
 				goto failed;