1 files changed, 119 insertions, 116 deletions

diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 113837e7ba98..fa2208bae2e1 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -93,31 +93,29 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(iocb->ki_filp);
-	struct mutex *aio_mutex = NULL;
 	struct blk_plug plug;
 	int o_direct = iocb->ki_flags & IOCB_DIRECT;
+	int unaligned_aio = 0;
 	int overwrite = 0;
 	ssize_t ret;
 
+	inode_lock(inode);
+	ret = generic_write_checks(iocb, from);
+	if (ret <= 0)
+		goto out;
+
 	/*
-	 * Unaligned direct AIO must be serialized; see comment above
-	 * In the case of O_APPEND, assume that we must always serialize
+	 * Unaligned direct AIO must be serialized among each other as zeroing
+	 * of partial blocks of two competing unaligned AIOs can result in data
+	 * corruption.
 	 */
-	if (o_direct &&
-	    ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) &&
+	if (o_direct && ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) &&
 	    !is_sync_kiocb(iocb) &&
-	    (iocb->ki_flags & IOCB_APPEND ||
-	     ext4_unaligned_aio(inode, from, iocb->ki_pos))) {
-		aio_mutex = ext4_aio_mutex(inode);
-		mutex_lock(aio_mutex);
+	    ext4_unaligned_aio(inode, from, iocb->ki_pos)) {
+		unaligned_aio = 1;
 		ext4_unwritten_wait(inode);
 	}
 
-	mutex_lock(&inode->i_mutex);
-	ret = generic_write_checks(iocb, from);
-	if (ret <= 0)
-		goto out;
-
 	/*
 	 * If we have encountered a bitmap-format file, the size limit
 	 * is smaller than s_maxbytes, which is for extent-mapped files.
@@ -139,7 +137,7 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		blk_start_plug(&plug);
 
 		/* check whether we do a DIO overwrite or not */
-		if (ext4_should_dioread_nolock(inode) && !aio_mutex &&
+		if (ext4_should_dioread_nolock(inode) && !unaligned_aio &&
 		    !file->f_mapping->nrpages && pos + length <= i_size_read(inode)) {
 			struct ext4_map_blocks map;
 			unsigned int blkbits = inode->i_blkbits;
@@ -169,7 +167,7 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	}
 
 	ret = __generic_file_write_iter(iocb, from);
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 
 	if (ret > 0) {
 		ssize_t err;
@@ -181,55 +179,43 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	if (o_direct)
 		blk_finish_plug(&plug);
 
-	if (aio_mutex)
-		mutex_unlock(aio_mutex);
 	return ret;
 
 out:
-	mutex_unlock(&inode->i_mutex);
-	if (aio_mutex)
-		mutex_unlock(aio_mutex);
+	inode_unlock(inode);
 	return ret;
 }
 
 #ifdef CONFIG_FS_DAX
-static void ext4_end_io_unwritten(struct buffer_head *bh, int uptodate)
-{
-	struct inode *inode = bh->b_assoc_map->host;
-	/* XXX: breaks on 32-bit > 16TB. Is that even supported? */
-	loff_t offset = (loff_t)(uintptr_t)bh->b_private << inode->i_blkbits;
-	int err;
-	if (!uptodate)
-		return;
-	WARN_ON(!buffer_unwritten(bh));
-	err = ext4_convert_unwritten_extents(NULL, inode, offset, bh->b_size);
-}
-
 static int ext4_dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	int result;
 	handle_t *handle = NULL;
-	struct super_block *sb = file_inode(vma->vm_file)->i_sb;
+	struct inode *inode = file_inode(vma->vm_file);
+	struct super_block *sb = inode->i_sb;
 	bool write = vmf->flags & FAULT_FLAG_WRITE;
 
 	if (write) {
 		sb_start_pagefault(sb);
 		file_update_time(vma->vm_file);
+		down_read(&EXT4_I(inode)->i_mmap_sem);
 		handle = ext4_journal_start_sb(sb, EXT4_HT_WRITE_PAGE,
 						EXT4_DATA_TRANS_BLOCKS(sb));
-	}
+	} else
+		down_read(&EXT4_I(inode)->i_mmap_sem);
 
 	if (IS_ERR(handle))
 		result = VM_FAULT_SIGBUS;
 	else
-		result = __dax_fault(vma, vmf, ext4_get_block_dax,
-						ext4_end_io_unwritten);
+		result = __dax_fault(vma, vmf, ext4_dax_mmap_get_block, NULL);
 
 	if (write) {
 		if (!IS_ERR(handle))
 			ext4_journal_stop(handle);
+		up_read(&EXT4_I(inode)->i_mmap_sem);
 		sb_end_pagefault(sb);
-	}
+	} else
+		up_read(&EXT4_I(inode)->i_mmap_sem);
 
 	return result;
 }
@@ -246,44 +232,73 @@ static int ext4_dax_pmd_fault(struct vm_area_struct *vma, unsigned long addr,
 	if (write) {
 		sb_start_pagefault(sb);
 		file_update_time(vma->vm_file);
+		down_read(&EXT4_I(inode)->i_mmap_sem);
 		handle = ext4_journal_start_sb(sb, EXT4_HT_WRITE_PAGE,
 				ext4_chunk_trans_blocks(inode,
 							PMD_SIZE / PAGE_SIZE));
-	}
+	} else
+		down_read(&EXT4_I(inode)->i_mmap_sem);
 
 	if (IS_ERR(handle))
 		result = VM_FAULT_SIGBUS;
 	else
 		result = __dax_pmd_fault(vma, addr, pmd, flags,
-				ext4_get_block_dax, ext4_end_io_unwritten);
+				ext4_dax_mmap_get_block, NULL);
 
 	if (write) {
 		if (!IS_ERR(handle))
 			ext4_journal_stop(handle);
+		up_read(&EXT4_I(inode)->i_mmap_sem);
 		sb_end_pagefault(sb);
-	}
+	} else
+		up_read(&EXT4_I(inode)->i_mmap_sem);
 
 	return result;
 }
 
-static int ext4_dax_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
+/*
+ * Handle write fault for VM_MIXEDMAP mappings. Similarly to ext4_dax_fault()
+ * handler we check for races agaist truncate. Note that since we cycle through
+ * i_mmap_sem, we are sure that also any hole punching that began before we
+ * were called is finished by now and so if it included part of the file we
+ * are working on, our pte will get unmapped and the check for pte_same() in
+ * wp_pfn_shared() fails. Thus fault gets retried and things work out as
+ * desired.
+ */
+static int ext4_dax_pfn_mkwrite(struct vm_area_struct *vma,
+				struct vm_fault *vmf)
 {
-	return dax_mkwrite(vma, vmf, ext4_get_block_dax,
-				ext4_end_io_unwritten);
+	struct inode *inode = file_inode(vma->vm_file);
+	struct super_block *sb = inode->i_sb;
+	loff_t size;
+	int ret;
+
+	sb_start_pagefault(sb);
+	file_update_time(vma->vm_file);
+	down_read(&EXT4_I(inode)->i_mmap_sem);
+	size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	if (vmf->pgoff >= size)
+		ret = VM_FAULT_SIGBUS;
+	else
+		ret = dax_pfn_mkwrite(vma, vmf);
+	up_read(&EXT4_I(inode)->i_mmap_sem);
+	sb_end_pagefault(sb);
+
+	return ret;
 }
 
 static const struct vm_operations_struct ext4_dax_vm_ops = {
 	.fault		= ext4_dax_fault,
 	.pmd_fault	= ext4_dax_pmd_fault,
-	.page_mkwrite	= ext4_dax_mkwrite,
-	.pfn_mkwrite	= dax_pfn_mkwrite,
+	.page_mkwrite	= ext4_dax_fault,
+	.pfn_mkwrite	= ext4_dax_pfn_mkwrite,
 };
 #else
 #define ext4_dax_vm_ops	ext4_file_vm_ops
 #endif
 
 static const struct vm_operations_struct ext4_file_vm_ops = {
-	.fault		= filemap_fault,
+	.fault		= ext4_filemap_fault,
 	.map_pages	= filemap_map_pages,
 	.page_mkwrite   = ext4_page_mkwrite,
 };
@@ -314,6 +329,7 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 	struct super_block *sb = inode->i_sb;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	struct vfsmount *mnt = filp->f_path.mnt;
+	struct dentry *dir;
 	struct path path;
 	char buf[64], *cp;
 	int ret;
@@ -357,6 +373,18 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 		if (ext4_encryption_info(inode) == NULL)
 			return -ENOKEY;
 	}
+
+	dir = dget_parent(file_dentry(filp));
+	if (ext4_encrypted_inode(d_inode(dir)) &&
+	    !ext4_is_child_context_consistent_with_parent(d_inode(dir), inode)) {
+		ext4_warning(inode->i_sb,
+			     "Inconsistent encryption contexts: %lu/%lu\n",
+			     (unsigned long) d_inode(dir)->i_ino,
+			     (unsigned long) inode->i_ino);
+		dput(dir);
+		return -EPERM;
+	}
+	dput(dir);
 	/*
 	 * Set up the jbd2_inode if we are opening the inode for
 	 * writing and the journal is present
@@ -387,7 +415,7 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
  */
 static int ext4_find_unwritten_pgoff(struct inode *inode,
 				     int whence,
-				     struct ext4_map_blocks *map,
+				     ext4_lblk_t end_blk,
 				     loff_t *offset)
 {
 	struct pagevec pvec;
@@ -402,10 +430,10 @@ static int ext4_find_unwritten_pgoff(struct inode *inode,
 	blkbits = inode->i_sb->s_blocksize_bits;
 	startoff = *offset;
 	lastoff = startoff;
-	endoff = (loff_t)(map->m_lblk + map->m_len) << blkbits;
+	endoff = (loff_t)end_blk << blkbits;
 
-	index = startoff >> PAGE_CACHE_SHIFT;
-	end = endoff >> PAGE_CACHE_SHIFT;
+	index = startoff >> PAGE_SHIFT;
+	end = endoff >> PAGE_SHIFT;
 
 	pagevec_init(&pvec, 0);
 	do {
@@ -520,18 +548,17 @@ out:
 static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize)
 {
 	struct inode *inode = file->f_mapping->host;
-	struct ext4_map_blocks map;
 	struct extent_status es;
 	ext4_lblk_t start, last, end;
 	loff_t dataoff, isize;
 	int blkbits;
-	int ret = 0;
+	int ret;
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 
 	isize = i_size_read(inode);
 	if (offset >= isize) {
-		mutex_unlock(&inode->i_mutex);
+		inode_unlock(inode);
 		return -ENXIO;
 	}
 
@@ -542,44 +569,35 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize)
 	dataoff = offset;
 
 	do {
-		map.m_lblk = last;
-		map.m_len = end - last + 1;
-		ret = ext4_map_blocks(NULL, inode, &map, 0);
-		if (ret > 0 && !(map.m_flags & EXT4_MAP_UNWRITTEN)) {
-			if (last != start)
-				dataoff = (loff_t)last << blkbits;
-			break;
+		ret = ext4_get_next_extent(inode, last, end - last + 1, &es);
+		if (ret <= 0) {
+			/* No extent found -> no data */
+			if (ret == 0)
+				ret = -ENXIO;
+			inode_unlock(inode);
+			return ret;
 		}
 
-		/*
-		 * If there is a delay extent at this offset,
-		 * it will be as a data.
-		 */
-		ext4_es_find_delayed_extent_range(inode, last, last, &es);
-		if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) {
-			if (last != start)
-				dataoff = (loff_t)last << blkbits;
+		last = es.es_lblk;
+		if (last != start)
+			dataoff = (loff_t)last << blkbits;
+		if (!ext4_es_is_unwritten(&es))
 			break;
-		}
 
 		/*
 		 * If there is a unwritten extent at this offset,
 		 * it will be as a data or a hole according to page
 		 * cache that has data or not.
 		 */
-		if (map.m_flags & EXT4_MAP_UNWRITTEN) {
-			int unwritten;
-			unwritten = ext4_find_unwritten_pgoff(inode, SEEK_DATA,
-							      &map, &dataoff);
-			if (unwritten)
-				break;
-		}
-
-		last++;
+		if (ext4_find_unwritten_pgoff(inode, SEEK_DATA,
+					      es.es_lblk + es.es_len, &dataoff))
+			break;
+		last += es.es_len;
 		dataoff = (loff_t)last << blkbits;
+		cond_resched();
 	} while (last <= end);
 
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 
 	if (dataoff > isize)
 		return -ENXIO;
@@ -593,18 +611,17 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize)
 static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize)
 {
 	struct inode *inode = file->f_mapping->host;
-	struct ext4_map_blocks map;
 	struct extent_status es;
 	ext4_lblk_t start, last, end;
 	loff_t holeoff, isize;
 	int blkbits;
-	int ret = 0;
+	int ret;
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 
 	isize = i_size_read(inode);
 	if (offset >= isize) {
-		mutex_unlock(&inode->i_mutex);
+		inode_unlock(inode);
 		return -ENXIO;
 	}
 
@@ -615,47 +632,33 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize)
 	holeoff = offset;
 
 	do {
-		map.m_lblk = last;
-		map.m_len = end - last + 1;
-		ret = ext4_map_blocks(NULL, inode, &map, 0);
-		if (ret > 0 && !(map.m_flags & EXT4_MAP_UNWRITTEN)) {
-			last += ret;
-			holeoff = (loff_t)last << blkbits;
-			continue;
+		ret = ext4_get_next_extent(inode, last, end - last + 1, &es);
+		if (ret < 0) {
+			inode_unlock(inode);
+			return ret;
 		}
-
-		/*
-		 * If there is a delay extent at this offset,
-		 * we will skip this extent.
-		 */
-		ext4_es_find_delayed_extent_range(inode, last, last, &es);
-		if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) {
-			last = es.es_lblk + es.es_len;
-			holeoff = (loff_t)last << blkbits;
-			continue;
+		/* Found a hole? */
+		if (ret == 0 || es.es_lblk > last) {
+			if (last != start)
+				holeoff = (loff_t)last << blkbits;
+			break;
 		}
-
 		/*
 		 * If there is a unwritten extent at this offset,
 		 * it will be as a data or a hole according to page
 		 * cache that has data or not.
 		 */
-		if (map.m_flags & EXT4_MAP_UNWRITTEN) {
-			int unwritten;
-			unwritten = ext4_find_unwritten_pgoff(inode, SEEK_HOLE,
-							      &map, &holeoff);
-			if (!unwritten) {
-				last += ret;
-				holeoff = (loff_t)last << blkbits;
-				continue;
-			}
-		}
+		if (ext4_es_is_unwritten(&es) &&
+		    ext4_find_unwritten_pgoff(inode, SEEK_HOLE,
+					      last + es.es_len, &holeoff))
+			break;
 
-		/* find a hole */
-		break;
+		last += es.es_len;
+		holeoff = (loff_t)last << blkbits;
+		cond_resched();
 	} while (last <= end);
 
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 
 	if (holeoff > isize)
 		holeoff = isize;