[PATCH 1/4] f2fs: fix to avoid reading out encrypted data in page cache

[PATCH 1/4] f2fs: fix to avoid reading out encrypted data in page cache

[Chao Yu]

From: Chao Yu <yuchao0@huawei.com>

For encrypted inode, if user overwrites data of the inode, f2fs will read
encrypted data into page cache, and then do the decryption.

However reader can race with overwriter, and it will see encrypted data
which has not been decrypted by overwriter yet. Fix it by moving decrypting
work to background and keep page non-uptodated until data is decrypted.

Thread A				Thread B
- f2fs_file_write_iter
 - __generic_file_write_iter
  - generic_perform_write
   - f2fs_write_begin
    - f2fs_submit_page_bio
					- generic_file_read_iter
					 - do_generic_file_read
					  - lock_page_killable
					  - unlock_page
					  - copy_page_to_iter
					  hit the encrypted data in updated page
    - lock_page
    - fscrypt_decrypt_page

Signed-off-by: Chao Yu <yuchao0@huawei.com>
---
 fs/f2fs/data.c | 90 ++++++++++++++++++++++++++++++----------------------------
 1 file changed, 47 insertions(+), 43 deletions(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 3a03285..2bfb4f1 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -955,6 +955,37 @@ out:
 	return ret;
 }
 
+struct bio *f2fs_grab_bio(struct inode *inode, block_t blkaddr,
+							unsigned nr_pages)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct fscrypt_ctx *ctx = NULL;
+	struct block_device *bdev = sbi->sb->s_bdev;
+	struct bio *bio;
+
+	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode)) {
+		ctx = fscrypt_get_ctx(inode, GFP_NOFS);
+		if (IS_ERR(ctx))
+			return ERR_CAST(ctx);
+
+		/* wait the page to be moved by cleaning */
+		f2fs_wait_on_encrypted_page_writeback(sbi, blkaddr);
+	}
+
+	bio = bio_alloc(GFP_KERNEL, min_t(int, nr_pages, BIO_MAX_PAGES));
+	if (!bio) {
+		if (ctx)
+			fscrypt_release_ctx(ctx);
+		return ERR_PTR(-ENOMEM);
+	}
+	bio->bi_bdev = bdev;
+	bio->bi_iter.bi_sector = SECTOR_FROM_BLOCK(blkaddr);
+	bio->bi_end_io = f2fs_read_end_io;
+	bio->bi_private = ctx;
+
+	return bio;
+}
+
 /*
  * This function was originally taken from fs/mpage.c, and customized for f2fs.
  * Major change was from block_size == page_size in f2fs by default.
@@ -973,7 +1004,6 @@ static int f2fs_mpage_readpages(struct address_space *mapping,
 	sector_t last_block;
 	sector_t last_block_in_file;
 	sector_t block_nr;
-	struct block_device *bdev = inode->i_sb->s_bdev;
 	struct f2fs_map_blocks map;
 
 	map.m_pblk = 0;
@@ -1048,31 +1078,9 @@ submit_and_realloc:
 			bio = NULL;
 		}
 		if (bio == NULL) {
-			struct fscrypt_ctx *ctx = NULL;
-
-			if (f2fs_encrypted_inode(inode) &&
-					S_ISREG(inode->i_mode)) {
-
-				ctx = fscrypt_get_ctx(inode, GFP_NOFS);
-				if (IS_ERR(ctx))
-					goto set_error_page;
-
-				/* wait the page to be moved by cleaning */
-				f2fs_wait_on_encrypted_page_writeback(
-						F2FS_I_SB(inode), block_nr);
-			}
-
-			bio = bio_alloc(GFP_KERNEL,
-				min_t(int, nr_pages, BIO_MAX_PAGES));
-			if (!bio) {
-				if (ctx)
-					fscrypt_release_ctx(ctx);
+			bio = f2fs_grab_bio(inode, block_nr, nr_pages);
+			if (IS_ERR(bio))
 				goto set_error_page;
-			}
-			bio->bi_bdev = bdev;
-			bio->bi_iter.bi_sector = SECTOR_FROM_BLOCK(block_nr);
-			bio->bi_end_io = f2fs_read_end_io;
-			bio->bi_private = ctx;
 		}
 
 		if (bio_add_page(bio, page, blocksize, 0) < blocksize)
@@ -1622,18 +1630,21 @@ repeat:
 	if (blkaddr == NEW_ADDR) {
 		zero_user_segment(page, 0, PAGE_SIZE);
 	} else {
-		struct f2fs_io_info fio = {
-			.sbi = sbi,
-			.type = DATA,
-			.rw = READ_SYNC,
-			.old_blkaddr = blkaddr,
-			.new_blkaddr = blkaddr,
-			.page = page,
-			.encrypted_page = NULL,
-		};
-		err = f2fs_submit_page_bio(&fio);
-		if (err)
+		struct bio *bio;
+
+		bio = f2fs_grab_bio(inode, blkaddr, 1);
+		if (IS_ERR(bio)) {
+			err = PTR_ERR(bio);
 			goto fail;
+		}
+
+		if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
+			bio_put(bio);
+			err = -EFAULT;
+			goto fail;
+		}
+
+		__submit_bio(sbi, READ_SYNC, bio, DATA);
 
 		lock_page(page);
 		if (unlikely(!PageUptodate(page))) {
@@ -1644,13 +1655,6 @@ repeat:
 			f2fs_put_page(page, 1);
 			goto repeat;
 		}
-
-		/* avoid symlink page */
-		if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode)) {
-			err = fscrypt_decrypt_page(page);
-			if (err)
-				goto fail;
-		}
 	}
 out_update:
 	SetPageUptodate(page);
-- 
2.7.2

[PATCH 2/4] f2fs: fix to detect truncation prior rather than EIO during read

[Chao Yu]

From: Chao Yu <yuchao0@huawei.com>

In procedure of synchonized read, after sending out the read request, reader
will try to lock the page for waiting device to finish the read jobs and
unlock the page, but meanwhile, truncater will race with reader, so after
reader get lock of the page, it should check page's mapping to detect
whether someone has truncated the page in advance, then reader has the
chance to do the retry if truncation was done, otherwise read can be failed
due to previous condition check.

Signed-off-by: Chao Yu <yuchao0@huawei.com>
---
 fs/f2fs/data.c | 16 ++++++++--------
 fs/f2fs/gc.c   |  4 ++--
 fs/f2fs/node.c |  6 +++---
 3 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 2bfb4f1..1a4700c 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -500,14 +500,14 @@ repeat:
 
 	/* wait for read completion */
 	lock_page(page);
-	if (unlikely(!PageUptodate(page))) {
-		f2fs_put_page(page, 1);
-		return ERR_PTR(-EIO);
-	}
 	if (unlikely(page->mapping != mapping)) {
 		f2fs_put_page(page, 1);
 		goto repeat;
 	}
+	if (unlikely(!PageUptodate(page))) {
+		f2fs_put_page(page, 1);
+		return ERR_PTR(-EIO);
+	}
 	return page;
 }
 
@@ -1647,14 +1647,14 @@ repeat:
 		__submit_bio(sbi, READ_SYNC, bio, DATA);
 
 		lock_page(page);
-		if (unlikely(!PageUptodate(page))) {
-			err = -EIO;
-			goto fail;
-		}
 		if (unlikely(page->mapping != mapping)) {
 			f2fs_put_page(page, 1);
 			goto repeat;
 		}
+		if (unlikely(!PageUptodate(page))) {
+			err = -EIO;
+			goto fail;
+		}
 	}
 out_update:
 	SetPageUptodate(page);
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index a15ae8a..d888490 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -593,11 +593,11 @@ static void move_encrypted_block(struct inode *inode, block_t bidx)
 	/* write page */
 	lock_page(fio.encrypted_page);
 
-	if (unlikely(!PageUptodate(fio.encrypted_page))) {
+	if (unlikely(fio.encrypted_page->mapping != META_MAPPING(fio.sbi))) {
 		err = -EIO;
 		goto put_page_out;
 	}
-	if (unlikely(fio.encrypted_page->mapping != META_MAPPING(fio.sbi))) {
+	if (unlikely(!PageUptodate(fio.encrypted_page))) {
 		err = -EIO;
 		goto put_page_out;
 	}
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index 729fb1e..69171ce 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -1146,13 +1146,13 @@ repeat:
 
 	lock_page(page);
 
-	if (unlikely(!PageUptodate(page)))
-		goto out_err;
-
 	if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
 		f2fs_put_page(page, 1);
 		goto repeat;
 	}
+
+	if (unlikely(!PageUptodate(page)))
+		goto out_err;
 page_hit:
 	if(unlikely(nid != nid_of_node(page))) {
 		f2fs_bug_on(sbi, 1);
-- 
2.7.2

[PATCH 3/4] f2fs: fix to redirty page if fail to gc data page

[Chao Yu]

From: Chao Yu <yuchao0@huawei.com>

If we fail to move data page during foreground GC, we should give another
chance to writeback that page which was set dirty previously by writer.

Signed-off-by: Chao Yu <yuchao0@huawei.com>
---
 fs/f2fs/gc.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index d888490..b744465 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -653,12 +653,23 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type)
 			.page = page,
 			.encrypted_page = NULL,
 		};
+		bool is_dirty = PageDirty(page);
+		int err;
+
+retry:
 		set_page_dirty(page);
 		f2fs_wait_on_page_writeback(page, DATA, true);
 		if (clear_page_dirty_for_io(page))
 			inode_dec_dirty_pages(inode);
+
 		set_cold_data(page);
-		do_write_data_page(&fio);
+
+		err = do_write_data_page(&fio);
+		if (err == -ENOMEM && is_dirty) {
+			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			goto retry;
+		}
+
 		clear_cold_data(page);
 	}
 out:
-- 
2.7.2

[PATCH 4/4] f2fs: add nodiscard mount option

[Chao Yu]

From: Chao Yu <yuchao0@huawei.com>

This patch adds 'nodiscard' mount option.

Signed-off-by: Chao Yu <yuchao0@huawei.com>
---
 Documentation/filesystems/f2fs.txt | 4 +++-
 fs/f2fs/super.c                    | 4 ++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/Documentation/filesystems/f2fs.txt b/Documentation/filesystems/f2fs.txt
index 3a5ce24..ecd8080 100644
--- a/Documentation/filesystems/f2fs.txt
+++ b/Documentation/filesystems/f2fs.txt
@@ -109,7 +109,9 @@ background_gc=%s       Turn on/off cleaning operations, namely garbage
 disable_roll_forward   Disable the roll-forward recovery routine
 norecovery             Disable the roll-forward recovery routine, mounted read-
                        only (i.e., -o ro,disable_roll_forward)
-discard                Issue discard/TRIM commands when a segment is cleaned.
+discard/nodiscard      Enable/disable real-time discard in f2fs, if discard is
+                       enabled, f2fs will issue discard/TRIM commands when a
+		       segment is cleaned.
 no_heap                Disable heap-style segment allocation which finds free
                        segments for data from the beginning of main area, while
 		       for node from the end of main area.
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 512bbb4..7b9b0c9 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -76,6 +76,7 @@ enum {
 	Opt_disable_roll_forward,
 	Opt_norecovery,
 	Opt_discard,
+	Opt_nodiscard,
 	Opt_noheap,
 	Opt_user_xattr,
 	Opt_nouser_xattr,
@@ -106,6 +107,7 @@ static match_table_t f2fs_tokens = {
 	{Opt_disable_roll_forward, "disable_roll_forward"},
 	{Opt_norecovery, "norecovery"},
 	{Opt_discard, "discard"},
+	{Opt_nodiscard, "nodiscard"},
 	{Opt_noheap, "no_heap"},
 	{Opt_user_xattr, "user_xattr"},
 	{Opt_nouser_xattr, "nouser_xattr"},
@@ -426,6 +428,8 @@ static int parse_options(struct super_block *sb, char *options)
 					"the device does not support discard");
 			}
 			break;
+		case Opt_nodiscard:
+			clear_opt(sbi, DISCARD);
 		case Opt_noheap:
 			set_opt(sbi, NOHEAP);
 			break;
-- 
2.7.2