[mirror_ubuntu-jammy-kernel.git] / fs / ext4 / verity.c

// SPDX-License-Identifier: GPL-2.0
/*
 * fs/ext4/verity.c: fs-verity support for ext4
 *
 * Copyright 2019 Google LLC
 */

/*
 * Implementation of fsverity_operations for ext4.
 *
 * ext4 stores the verity metadata (Merkle tree and fsverity_descriptor) past
 * the end of the file, starting at the first 64K boundary beyond i_size.  This
 * approach works because (a) verity files are readonly, and (b) pages fully
 * beyond i_size aren't visible to userspace but can be read/written internally
 * by ext4 with only some relatively small changes to ext4.  This approach
 * avoids having to depend on the EA_INODE feature and on rearchitecturing
 * ext4's xattr support to support paging multi-gigabyte xattrs into memory, and
 * to support encrypting xattrs.  Note that the verity metadata *must* be
 * encrypted when the file is, since it contains hashes of the plaintext data.
 *
 * Using a 64K boundary rather than a 4K one keeps things ready for
 * architectures with 64K pages, and it doesn't necessarily waste space on-disk
 * since there can be a hole between i_size and the start of the Merkle tree.
 */

#include <linux/quotaops.h>

#include "ext4.h"
#include "ext4_extents.h"
#include "ext4_jbd2.h"

static inline loff_t ext4_verity_metadata_pos(const struct inode *inode)
{
	return round_up(inode->i_size, 65536);
}

/*
 * Read some verity metadata from the inode.  __vfs_read() can't be used because
 * we need to read beyond i_size.
 */
static int pagecache_read(struct inode *inode, void *buf, size_t count,
			  loff_t pos)
{
	while (count) {
		size_t n = min_t(size_t, count,
				 PAGE_SIZE - offset_in_page(pos));
		struct page *page;

		page = read_mapping_page(inode->i_mapping, pos >> PAGE_SHIFT,
					 NULL);
		if (IS_ERR(page))
			return PTR_ERR(page);

		memcpy_from_page(buf, page, offset_in_page(pos), n);

		put_page(page);

		buf += n;
		pos += n;
		count -= n;
	}
	return 0;
}

/*
 * Write some verity metadata to the inode for FS_IOC_ENABLE_VERITY.
 * kernel_write() can't be used because the file descriptor is readonly.
 */
static int pagecache_write(struct inode *inode, const void *buf, size_t count,
			   loff_t pos)
{
	if (pos + count > inode->i_sb->s_maxbytes)
		return -EFBIG;

	while (count) {
		size_t n = min_t(size_t, count,
				 PAGE_SIZE - offset_in_page(pos));
		struct page *page;
		void *fsdata;
		int res;

		res = pagecache_write_begin(NULL, inode->i_mapping, pos, n, 0,
					    &page, &fsdata);
		if (res)
			return res;

		memcpy_to_page(page, offset_in_page(pos), buf, n);

		res = pagecache_write_end(NULL, inode->i_mapping, pos, n, n,
					  page, fsdata);
		if (res < 0)
			return res;
		if (res != n)
			return -EIO;

		buf += n;
		pos += n;
		count -= n;
	}
	return 0;
}

static int ext4_begin_enable_verity(struct file *filp)
{
	struct inode *inode = file_inode(filp);
	const int credits = 2; /* superblock and inode for ext4_orphan_add() */
	handle_t *handle;
	int err;

	if (IS_DAX(inode) || ext4_test_inode_flag(inode, EXT4_INODE_DAX))
		return -EINVAL;

	if (ext4_verity_in_progress(inode))
		return -EBUSY;

	/*
	 * Since the file was opened readonly, we have to initialize the jbd
	 * inode and quotas here and not rely on ->open() doing it.  This must
	 * be done before evicting the inline data.
	 */

	err = ext4_inode_attach_jinode(inode);
	if (err)
		return err;

	err = dquot_initialize(inode);
	if (err)
		return err;

	err = ext4_convert_inline_data(inode);
	if (err)
		return err;

	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
		ext4_warning_inode(inode,
				   "verity is only allowed on extent-based files");
		return -EOPNOTSUPP;
	}

	/*
	 * ext4 uses the last allocated block to find the verity descriptor, so
	 * we must remove any other blocks past EOF which might confuse things.
	 */
	err = ext4_truncate(inode);
	if (err)
		return err;

	handle = ext4_journal_start(inode, EXT4_HT_INODE, credits);
	if (IS_ERR(handle))
		return PTR_ERR(handle);

	err = ext4_orphan_add(handle, inode);
	if (err == 0)
		ext4_set_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS);

	ext4_journal_stop(handle);
	return err;
}

/*
 * ext4 stores the verity descriptor beginning on the next filesystem block
 * boundary after the Merkle tree.  Then, the descriptor size is stored in the
 * last 4 bytes of the last allocated filesystem block --- which is either the
 * block in which the descriptor ends, or the next block after that if there
 * weren't at least 4 bytes remaining.
 *
 * We can't simply store the descriptor in an xattr because it *must* be
 * encrypted when ext4 encryption is used, but ext4 encryption doesn't encrypt
 * xattrs.  Also, if the descriptor includes a large signature blob it may be
 * too large to store in an xattr without the EA_INODE feature.
 */
static int ext4_write_verity_descriptor(struct inode *inode, const void *desc,
					size_t desc_size, u64 merkle_tree_size)
{
	const u64 desc_pos = round_up(ext4_verity_metadata_pos(inode) +
				      merkle_tree_size, i_blocksize(inode));
	const u64 desc_end = desc_pos + desc_size;
	const __le32 desc_size_disk = cpu_to_le32(desc_size);
	const u64 desc_size_pos = round_up(desc_end + sizeof(desc_size_disk),
					   i_blocksize(inode)) -
				  sizeof(desc_size_disk);
	int err;

	err = pagecache_write(inode, desc, desc_size, desc_pos);
	if (err)
		return err;

	return pagecache_write(inode, &desc_size_disk, sizeof(desc_size_disk),
			       desc_size_pos);
}

static int ext4_end_enable_verity(struct file *filp, const void *desc,
				  size_t desc_size, u64 merkle_tree_size)
{
	struct inode *inode = file_inode(filp);
	const int credits = 2; /* superblock and inode for ext4_orphan_del() */
	handle_t *handle;
	struct ext4_iloc iloc;
	int err = 0;

	/*
	 * If an error already occurred (which fs/verity/ signals by passing
	 * desc == NULL), then only clean-up is needed.
	 */
	if (desc == NULL)
		goto cleanup;

	/* Append the verity descriptor. */
	err = ext4_write_verity_descriptor(inode, desc, desc_size,
					   merkle_tree_size);
	if (err)
		goto cleanup;

	/*
	 * Write all pages (both data and verity metadata).  Note that this must
	 * happen before clearing EXT4_STATE_VERITY_IN_PROGRESS; otherwise pages
	 * beyond i_size won't be written properly.  For crash consistency, this
	 * also must happen before the verity inode flag gets persisted.
	 */
	err = filemap_write_and_wait(inode->i_mapping);
	if (err)
		goto cleanup;

	/*
	 * Finally, set the verity inode flag and remove the inode from the
	 * orphan list (in a single transaction).
	 */

	handle = ext4_journal_start(inode, EXT4_HT_INODE, credits);
	if (IS_ERR(handle)) {
		err = PTR_ERR(handle);
		goto cleanup;
	}

	err = ext4_orphan_del(handle, inode);
	if (err)
		goto stop_and_cleanup;

	err = ext4_reserve_inode_write(handle, inode, &iloc);
	if (err)
		goto stop_and_cleanup;

	ext4_set_inode_flag(inode, EXT4_INODE_VERITY);
	ext4_set_inode_flags(inode, false);
	err = ext4_mark_iloc_dirty(handle, inode, &iloc);
	if (err)
		goto stop_and_cleanup;

	ext4_journal_stop(handle);

	ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS);
	return 0;

stop_and_cleanup:
	ext4_journal_stop(handle);
cleanup:
	/*
	 * Verity failed to be enabled, so clean up by truncating any verity
	 * metadata that was written beyond i_size (both from cache and from
	 * disk), removing the inode from the orphan list (if it wasn't done
	 * already), and clearing EXT4_STATE_VERITY_IN_PROGRESS.
	 */
	truncate_inode_pages(inode->i_mapping, inode->i_size);
	ext4_truncate(inode);
	ext4_orphan_del(NULL, inode);
	ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS);
	return err;
}

static int ext4_get_verity_descriptor_location(struct inode *inode,
					       size_t *desc_size_ret,
					       u64 *desc_pos_ret)
{
	struct ext4_ext_path *path;
	struct ext4_extent *last_extent;
	u32 end_lblk;
	u64 desc_size_pos;
	__le32 desc_size_disk;
	u32 desc_size;
	u64 desc_pos;
	int err;

	/*
	 * Descriptor size is in last 4 bytes of last allocated block.
	 * See ext4_write_verity_descriptor().
	 */

	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
		EXT4_ERROR_INODE(inode, "verity file doesn't use extents");
		return -EFSCORRUPTED;
	}

	path = ext4_find_extent(inode, EXT_MAX_BLOCKS - 1, NULL, 0);
	if (IS_ERR(path))
		return PTR_ERR(path);

	last_extent = path[path->p_depth].p_ext;
	if (!last_extent) {
		EXT4_ERROR_INODE(inode, "verity file has no extents");
		ext4_ext_drop_refs(path);
		kfree(path);
		return -EFSCORRUPTED;
	}

	end_lblk = le32_to_cpu(last_extent->ee_block) +
		   ext4_ext_get_actual_len(last_extent);
	desc_size_pos = (u64)end_lblk << inode->i_blkbits;
	ext4_ext_drop_refs(path);
	kfree(path);

	if (desc_size_pos < sizeof(desc_size_disk))
		goto bad;
	desc_size_pos -= sizeof(desc_size_disk);

	err = pagecache_read(inode, &desc_size_disk, sizeof(desc_size_disk),
			     desc_size_pos);
	if (err)
		return err;
	desc_size = le32_to_cpu(desc_size_disk);

	/*
	 * The descriptor is stored just before the desc_size_disk, but starting
	 * on a filesystem block boundary.
	 */

	if (desc_size > INT_MAX || desc_size > desc_size_pos)
		goto bad;

	desc_pos = round_down(desc_size_pos - desc_size, i_blocksize(inode));
	if (desc_pos < ext4_verity_metadata_pos(inode))
		goto bad;

	*desc_size_ret = desc_size;
	*desc_pos_ret = desc_pos;
	return 0;

bad:
	EXT4_ERROR_INODE(inode, "verity file corrupted; can't find descriptor");
	return -EFSCORRUPTED;
}

static int ext4_get_verity_descriptor(struct inode *inode, void *buf,
				      size_t buf_size)
{
	size_t desc_size = 0;
	u64 desc_pos = 0;
	int err;

	err = ext4_get_verity_descriptor_location(inode, &desc_size, &desc_pos);
	if (err)
		return err;

	if (buf_size) {
		if (desc_size > buf_size)
			return -ERANGE;
		err = pagecache_read(inode, buf, desc_size, desc_pos);
		if (err)
			return err;
	}
	return desc_size;
}

static struct page *ext4_read_merkle_tree_page(struct inode *inode,
					       pgoff_t index,
					       unsigned long num_ra_pages)
{
	DEFINE_READAHEAD(ractl, NULL, NULL, inode->i_mapping, index);
	struct page *page;

	index += ext4_verity_metadata_pos(inode) >> PAGE_SHIFT;

	page = find_get_page_flags(inode->i_mapping, index, FGP_ACCESSED);
	if (!page || !PageUptodate(page)) {
		if (page)
			put_page(page);
		else if (num_ra_pages > 1)
			page_cache_ra_unbounded(&ractl, num_ra_pages, 0);
		page = read_mapping_page(inode->i_mapping, index, NULL);
	}
	return page;
}

static int ext4_write_merkle_tree_block(struct inode *inode, const void *buf,
					u64 index, int log_blocksize)
{
	loff_t pos = ext4_verity_metadata_pos(inode) + (index << log_blocksize);

	return pagecache_write(inode, buf, 1 << log_blocksize, pos);
}

const struct fsverity_operations ext4_verityops = {
	.begin_enable_verity	= ext4_begin_enable_verity,
	.end_enable_verity	= ext4_end_enable_verity,
	.get_verity_descriptor	= ext4_get_verity_descriptor,
	.read_merkle_tree_page	= ext4_read_merkle_tree_page,
	.write_merkle_tree_block = ext4_write_merkle_tree_block,
};
Commit	Line	Data
c93d8f88 EB	1	// SPDX-License-Identifier: GPL-2.0
	2	/*
	3	* fs/ext4/verity.c: fs-verity support for ext4
	4	*
	5	* Copyright 2019 Google LLC
	6	*/
	7
	8	/*
	9	* Implementation of fsverity_operations for ext4.
	10	*
	11	* ext4 stores the verity metadata (Merkle tree and fsverity_descriptor) past
	12	* the end of the file, starting at the first 64K boundary beyond i_size. This
	13	* approach works because (a) verity files are readonly, and (b) pages fully
	14	* beyond i_size aren't visible to userspace but can be read/written internally
	15	* by ext4 with only some relatively small changes to ext4. This approach
	16	* avoids having to depend on the EA_INODE feature and on rearchitecturing
	17	* ext4's xattr support to support paging multi-gigabyte xattrs into memory, and
	18	* to support encrypting xattrs. Note that the verity metadata must be
	19	* encrypted when the file is, since it contains hashes of the plaintext data.
	20	*
	21	* Using a 64K boundary rather than a 4K one keeps things ready for
	22	* architectures with 64K pages, and it doesn't necessarily waste space on-disk
	23	* since there can be a hole between i_size and the start of the Merkle tree.
	24	*/
	25
	26	#include <linux/quotaops.h>
	27
	28	#include "ext4.h"
	29	#include "ext4_extents.h"
	30	#include "ext4_jbd2.h"
	31
	32	static inline loff_t ext4_verity_metadata_pos(const struct inode *inode)
	33	{
	34	return round_up(inode->i_size, 65536);
	35	}
	36
	37	/*
	38	* Read some verity metadata from the inode. __vfs_read() can't be used because
	39	* we need to read beyond i_size.
	40	*/
	41	static int pagecache_read(struct inode inode, void buf, size_t count,
	42	loff_t pos)
	43	{
	44	while (count) {
	45	size_t n = min_t(size_t, count,
	46	PAGE_SIZE - offset_in_page(pos));
	47	struct page *page;
c93d8f88 EB	48
	49	page = read_mapping_page(inode->i_mapping, pos >> PAGE_SHIFT,
	50	NULL);
	51	if (IS_ERR(page))
	52	return PTR_ERR(page);
	53
4d93874b	54	memcpy_from_page(buf, page, offset_in_page(pos), n);
c93d8f88 EB	55
	56	put_page(page);
	57
	58	buf += n;
	59	pos += n;
	60	count -= n;
	61	}
	62	return 0;
	63	}
	64
	65	/*
	66	* Write some verity metadata to the inode for FS_IOC_ENABLE_VERITY.
	67	* kernel_write() can't be used because the file descriptor is readonly.
	68	*/
	69	static int pagecache_write(struct inode inode, const void buf, size_t count,
	70	loff_t pos)
	71	{
	72	if (pos + count > inode->i_sb->s_maxbytes)
	73	return -EFBIG;
	74
	75	while (count) {
	76	size_t n = min_t(size_t, count,
	77	PAGE_SIZE - offset_in_page(pos));
	78	struct page *page;
	79	void *fsdata;
c93d8f88 EB	80	int res;
	81
	82	res = pagecache_write_begin(NULL, inode->i_mapping, pos, n, 0,
	83	&page, &fsdata);
	84	if (res)
	85	return res;
	86
bd256fda	87	memcpy_to_page(page, offset_in_page(pos), buf, n);
c93d8f88 EB	88
	89	res = pagecache_write_end(NULL, inode->i_mapping, pos, n, n,
	90	page, fsdata);
	91	if (res < 0)
	92	return res;
	93	if (res != n)
	94	return -EIO;
	95
	96	buf += n;
	97	pos += n;
	98	count -= n;
	99	}
	100	return 0;
	101	}
	102
	103	static int ext4_begin_enable_verity(struct file *filp)
	104	{
	105	struct inode *inode = file_inode(filp);
	106	const int credits = 2; /* superblock and inode for ext4_orphan_add() */
	107	handle_t *handle;
	108	int err;
	109
b383a73f	110	if (IS_DAX(inode) \|\| ext4_test_inode_flag(inode, EXT4_INODE_DAX))
6c0d077f IW	111	return -EINVAL;
6c0d077f IW	112
c93d8f88 EB	113	if (ext4_verity_in_progress(inode))
	114	return -EBUSY;
	115
	116	/*
	117	* Since the file was opened readonly, we have to initialize the jbd
	118	* inode and quotas here and not rely on ->open() doing it. This must
	119	* be done before evicting the inline data.
	120	*/
	121
	122	err = ext4_inode_attach_jinode(inode);
	123	if (err)
	124	return err;
	125
	126	err = dquot_initialize(inode);
	127	if (err)
	128	return err;
	129
	130	err = ext4_convert_inline_data(inode);
	131	if (err)
	132	return err;
	133
	134	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
	135	ext4_warning_inode(inode,
	136	"verity is only allowed on extent-based files");
	137	return -EOPNOTSUPP;
	138	}
	139
	140	/*
	141	* ext4 uses the last allocated block to find the verity descriptor, so
	142	* we must remove any other blocks past EOF which might confuse things.
	143	*/
	144	err = ext4_truncate(inode);
	145	if (err)
	146	return err;
	147
	148	handle = ext4_journal_start(inode, EXT4_HT_INODE, credits);
	149	if (IS_ERR(handle))
	150	return PTR_ERR(handle);
	151
	152	err = ext4_orphan_add(handle, inode);
	153	if (err == 0)
	154	ext4_set_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS);
	155
	156	ext4_journal_stop(handle);
	157	return err;
	158	}
	159
	160	/*
	161	* ext4 stores the verity descriptor beginning on the next filesystem block
	162	* boundary after the Merkle tree. Then, the descriptor size is stored in the
	163	* last 4 bytes of the last allocated filesystem block --- which is either the
	164	* block in which the descriptor ends, or the next block after that if there
	165	* weren't at least 4 bytes remaining.
	166	*
	167	* We can't simply store the descriptor in an xattr because it must be
	168	* encrypted when ext4 encryption is used, but ext4 encryption doesn't encrypt
	169	* xattrs. Also, if the descriptor includes a large signature blob it may be
	170	* too large to store in an xattr without the EA_INODE feature.
	171	*/
	172	static int ext4_write_verity_descriptor(struct inode inode, const void desc,
	173	size_t desc_size, u64 merkle_tree_size)
	174	{
	175	const u64 desc_pos = round_up(ext4_verity_metadata_pos(inode) +
	176	merkle_tree_size, i_blocksize(inode));
177	const u64 desc_end = desc_pos + desc_size;
178	const __le32 desc_size_disk = cpu_to_le32(desc_size);
179	const u64 desc_size_pos = round_up(desc_end + sizeof(desc_size_disk),
180	i_blocksize(inode)) -
181	sizeof(desc_size_disk);
182	int err;
183
184	err = pagecache_write(inode, desc, desc_size, desc_pos);
185	if (err)
186	return err;
187
188	return pagecache_write(inode, &desc_size_disk, sizeof(desc_size_disk),
189	desc_size_pos);
190	}
191
192	static int ext4_end_enable_verity(struct file filp, const void desc,
193	size_t desc_size, u64 merkle_tree_size)
194	{
195	struct inode *inode = file_inode(filp);
196	const int credits = 2; /* superblock and inode for ext4_orphan_del() */
197	handle_t *handle;
f053cf7a	198	struct ext4_iloc iloc;
c93d8f88	199	int err = 0;
c93d8f88	200
f053cf7a EB	201	/*
	202	* If an error already occurred (which fs/verity/ signals by passing
	203	* desc == NULL), then only clean-up is needed.
	204	*/
	205	if (desc == NULL)
	206	goto cleanup;
c93d8f88	207
f053cf7a EB	208	/* Append the verity descriptor. */
	209	err = ext4_write_verity_descriptor(inode, desc, desc_size,
	210	merkle_tree_size);
	211	if (err)
	212	goto cleanup;
c93d8f88 EB	213
c93d8f88 EB	214	/*
f053cf7a EB	215	* Write all pages (both data and verity metadata). Note that this must
	216	* happen before clearing EXT4_STATE_VERITY_IN_PROGRESS; otherwise pages
	217	* beyond i_size won't be written properly. For crash consistency, this
	218	* also must happen before the verity inode flag gets persisted.
c93d8f88	219	*/
f053cf7a EB	220	err = filemap_write_and_wait(inode->i_mapping);
	221	if (err)
	222	goto cleanup;
c93d8f88	223
f053cf7a EB	224	/*
	225	* Finally, set the verity inode flag and remove the inode from the
	226	* orphan list (in a single transaction).
	227	*/
c93d8f88 EB	228
	229	handle = ext4_journal_start(inode, EXT4_HT_INODE, credits);
	230	if (IS_ERR(handle)) {
f053cf7a EB	231	err = PTR_ERR(handle);
f053cf7a EB	232	goto cleanup;
c93d8f88 EB	233	}
c93d8f88 EB	234
f053cf7a EB	235	err = ext4_orphan_del(handle, inode);
	236	if (err)
	237	goto stop_and_cleanup;
	238
	239	err = ext4_reserve_inode_write(handle, inode, &iloc);
	240	if (err)
	241	goto stop_and_cleanup;
c93d8f88	242
f053cf7a EB	243	ext4_set_inode_flag(inode, EXT4_INODE_VERITY);
	244	ext4_set_inode_flags(inode, false);
	245	err = ext4_mark_iloc_dirty(handle, inode, &iloc);
	246	if (err)
	247	goto stop_and_cleanup;
c93d8f88	248
c93d8f88	249	ext4_journal_stop(handle);
f053cf7a EB	250
	251	ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS);
	252	return 0;
	253
	254	stop_and_cleanup:
	255	ext4_journal_stop(handle);
	256	cleanup:
	257	/*
	258	* Verity failed to be enabled, so clean up by truncating any verity
	259	* metadata that was written beyond i_size (both from cache and from
	260	* disk), removing the inode from the orphan list (if it wasn't done
	261	* already), and clearing EXT4_STATE_VERITY_IN_PROGRESS.
	262	*/
	263	truncate_inode_pages(inode->i_mapping, inode->i_size);
	264	ext4_truncate(inode);
	265	ext4_orphan_del(NULL, inode);
	266	ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS);
	267	return err;
c93d8f88 EB	268	}
	269
	270	static int ext4_get_verity_descriptor_location(struct inode *inode,
	271	size_t *desc_size_ret,
	272	u64 *desc_pos_ret)
	273	{
	274	struct ext4_ext_path *path;
	275	struct ext4_extent *last_extent;
	276	u32 end_lblk;
	277	u64 desc_size_pos;
	278	__le32 desc_size_disk;
	279	u32 desc_size;
	280	u64 desc_pos;
	281	int err;
	282
	283	/*
	284	* Descriptor size is in last 4 bytes of last allocated block.
	285	* See ext4_write_verity_descriptor().
	286	*/
	287
	288	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
	289	EXT4_ERROR_INODE(inode, "verity file doesn't use extents");
	290	return -EFSCORRUPTED;
	291	}
	292
	293	path = ext4_find_extent(inode, EXT_MAX_BLOCKS - 1, NULL, 0);
	294	if (IS_ERR(path))
	295	return PTR_ERR(path);
	296
	297	last_extent = path[path->p_depth].p_ext;
	298	if (!last_extent) {
	299	EXT4_ERROR_INODE(inode, "verity file has no extents");
	300	ext4_ext_drop_refs(path);
	301	kfree(path);
	302	return -EFSCORRUPTED;
	303	}
	304
	305	end_lblk = le32_to_cpu(last_extent->ee_block) +
	306	ext4_ext_get_actual_len(last_extent);
	307	desc_size_pos = (u64)end_lblk << inode->i_blkbits;
	308	ext4_ext_drop_refs(path);
	309	kfree(path);
	310
	311	if (desc_size_pos < sizeof(desc_size_disk))
	312	goto bad;
	313	desc_size_pos -= sizeof(desc_size_disk);
	314
	315	err = pagecache_read(inode, &desc_size_disk, sizeof(desc_size_disk),
	316	desc_size_pos);
	317	if (err)
	318	return err;
	319	desc_size = le32_to_cpu(desc_size_disk);
	320
	321	/*
	322	* The descriptor is stored just before the desc_size_disk, but starting
	323	* on a filesystem block boundary.
	324	*/
	325
	326	if (desc_size > INT_MAX \|\| desc_size > desc_size_pos)
	327	goto bad;
	328
	329	desc_pos = round_down(desc_size_pos - desc_size, i_blocksize(inode));
	330	if (desc_pos < ext4_verity_metadata_pos(inode))
	331	goto bad;
332
333	*desc_size_ret = desc_size;
334	*desc_pos_ret = desc_pos;
335	return 0;
336
337	bad:
338	EXT4_ERROR_INODE(inode, "verity file corrupted; can't find descriptor");
339	return -EFSCORRUPTED;
340	}
341
342	static int ext4_get_verity_descriptor(struct inode inode, void buf,
343	size_t buf_size)
344	{
345	size_t desc_size = 0;
346	u64 desc_pos = 0;
347	int err;
348
349	err = ext4_get_verity_descriptor_location(inode, &desc_size, &desc_pos);
350	if (err)
351	return err;
352
353	if (buf_size) {
354	if (desc_size > buf_size)
355	return -ERANGE;
356	err = pagecache_read(inode, buf, desc_size, desc_pos);
357	if (err)
358	return err;
359	}
360	return desc_size;
361	}
362
363	static struct page ext4_read_merkle_tree_page(struct inode inode,
fd39073d EB	364	pgoff_t index,
fd39073d EB	365	unsigned long num_ra_pages)
c93d8f88	366	{
fcd9ae4f	367	DEFINE_READAHEAD(ractl, NULL, NULL, inode->i_mapping, index);
fd39073d EB	368	struct page *page;
fd39073d EB	369
c93d8f88 EB	370	index += ext4_verity_metadata_pos(inode) >> PAGE_SHIFT;
c93d8f88 EB	371
fd39073d EB	372	page = find_get_page_flags(inode->i_mapping, index, FGP_ACCESSED);
	373	if (!page \|\| !PageUptodate(page)) {
	374	if (page)
	375	put_page(page);
	376	else if (num_ra_pages > 1)
73bb49da	377	page_cache_ra_unbounded(&ractl, num_ra_pages, 0);
fd39073d EB	378	page = read_mapping_page(inode->i_mapping, index, NULL);
	379	}
	380	return page;
c93d8f88 EB	381	}
	382
	383	static int ext4_write_merkle_tree_block(struct inode inode, const void buf,
	384	u64 index, int log_blocksize)
	385	{
	386	loff_t pos = ext4_verity_metadata_pos(inode) + (index << log_blocksize);
	387
	388	return pagecache_write(inode, buf, 1 << log_blocksize, pos);
	389	}
	390
	391	const struct fsverity_operations ext4_verityops = {
	392	.begin_enable_verity = ext4_begin_enable_verity,
	393	.end_enable_verity = ext4_end_enable_verity,
	394	.get_verity_descriptor = ext4_get_verity_descriptor,
	395	.read_merkle_tree_page = ext4_read_merkle_tree_page,
	396	.write_merkle_tree_block = ext4_write_merkle_tree_block,
	397	};