Skip to contentSkip to conversation

Consolidate FS read I/O callbacks code

PATCH Superseded Chandan Rajendra 2 participants 19 messages Jun 16, 2019 – Jun 25, 2019

Subsystems (5)

v3 of 5 Superseded Latest is v5 →

lore
Share & tools
lore
#0 lore raw

v3 cover Consolidate FS read I/O callbacks code

This patchset moves the "FS read I/O callbacks" code into a file of its own (i.e. fs/read_callbacks.c) and modifies the generic do_mpage_readpge() to make use of the functionality provided.

"FS read I/O callbacks" code implements the state machine that needs to be executed after reading data from files that are encrypted and/or have verity metadata associated with them.

With these changes in place, the patchset changes Ext4 to use mpage_readpage[s] instead of its own custom ext4_readpage[s]() functions. This is done to reduce duplication of code across filesystems. Also, "FS read I/O callbacks" source files will be built only if CONFIG_FS_ENCRYPTION is enabled.

The patchset also modifies fs/buffer.c to get file encryption/decryption to work with subpage-sized blocks.

The patches can also be obtained from
https://github.com/chandanr/linux.git at branch subpage-encryption-v3.

Changelog:
V2 -> V3:
1. Split the V2 patch "Consolidate 'read callbacks' into a new file" into three patches,
- Introduce the read_callbacks functionality.
- Convert encryption to use read_callbacks.
- Remove union from struct fscrypt_context.
2. fs/Kconfig
Do not explicitly set the default value of 'n' for FS_READ_CALLBACKS.
3. fs/crypto/Kconfig
Select CONFIG_FS_READ_CALLBACKS only if CONFIG_BLOCK is selected.
4. Remove verity associated code in read_callbacks code.
5. Introduce a callback argument to read_callbacks_setup() function which gets invoked for each page for bio. F2FS uses this to perform custom operations like decrementing the value of f2fs_sb_info->nr_pages[].
6. Encapsulate the details of "read callbacks" (e.g. Usage of "struct read_callbacks *ctx") within its own functions. When CONFIG_FS_READ_CALLBACKS is set to 'n', the corresponding stub functions return approriate error values.
7. Split fscrypt_decrypt() function into fscrypt_decrypt_bio() and fscrypt_decrypt_bh().
8. Split end_read_callbacks() function into end_read_callbacks_bio() and end_read_callbacks_bh().

V1 -> V2:
1. Removed the phrase "post_read_process" from file names and functions. Instead we now use the phrase "read_callbacks" in its place.
2. When performing changes associated with (1), the changes made by the patch "Remove the term 'bio' from post read processing" are made in the earlier patch "Consolidate 'read callbacks' into a new file". Hence the patch "Remove the term 'bio' from post read processing" is removed from the patchset.

RFC V2 -> V1:
1. Test and verify FS_CFLG_OWN_PAGES subset of fscrypt_encrypt_page() code by executing fstests on UBIFS.
2. Implement F2fs function call back to check if the contents of a page holding a verity file's data needs to be verified.

RFC V1 -> RFC V2:
1. Describe the purpose of "Post processing code" in the cover letter.
2. Fix build errors when CONFIG_FS_VERITY is enabled.

Chandan Rajendra (7):
FS: Introduce read callbacks
Integrate read callbacks into Ext4 and F2FS
fscrypt: remove struct fscrypt_ctx
fs/mpage.c: Integrate read callbacks
ext4: Wire up ext4_readpage[s] to use mpage_readpage[s] Add decryption support for sub-pagesized blocks
ext4: Enable encryption for subpage-sized blocks

 Documentation/filesystems/fscrypt.rst |   4 +-
 fs/Kconfig                            |   3 +
 fs/Makefile                           |   2 +
 fs/buffer.c                           |  55 +++--
 fs/crypto/Kconfig                     |   1 +
 fs/crypto/bio.c                       |  44 ++--
 fs/crypto/crypto.c                    |  90 +-------
 fs/crypto/fscrypt_private.h           |   3 +
 fs/ext4/Makefile                      |   2 +-
 fs/ext4/inode.c                       |   5 +-
 fs/ext4/readpage.c                    | 295 --------------------------
 fs/ext4/super.c                       |   7 -
 fs/f2fs/data.c                        | 124 ++---------
 fs/f2fs/super.c                       |   9 +-
 fs/mpage.c                            |  11 +-
 fs/read_callbacks.c                   | 233 ++++++++++++++++++++
 include/linux/buffer_head.h           |   1 +
 include/linux/fscrypt.h               |  38 ----
 include/linux/read_callbacks.h        |  45 ++++
 19 files changed, 390 insertions(+), 582 deletions(-)
 delete mode 100644 fs/ext4/readpage.c
 create mode 100644 fs/read_callbacks.c
 create mode 100644 include/linux/read_callbacks.h
Signature · 1 line
2.19.1
#1 lore raw

PATCH v3 1/7 FS: Introduce read callbacks

Read callbacks implements a state machine to be executed after a buffered read I/O is completed. They help in further processing the file data read from the backing store. Currently, decryption is the only post processing step to be supported.

The execution of the state machine is to be initiated by the endio function associated with the read operation.

---
 fs/Kconfig                     |   3 +
 fs/Makefile                    |   2 +
 fs/crypto/Kconfig              |   1 +
 fs/crypto/bio.c                |  11 +++
 fs/read_callbacks.c            | 174 +++++++++++++++++++++++++++++++++
 include/linux/fscrypt.h        |   5 +
 include/linux/read_callbacks.h |  38 +++++++
 7 files changed, 234 insertions(+)
 create mode 100644 fs/read_callbacks.c
 create mode 100644 include/linux/read_callbacks.h
fs/Kconfig +3 -0
index f1046cf6ad85..d869859c88da 100644
@@ -21,6 +21,9 @@ if BLOCK
 config FS_IOMAP
 	bool
 
+config FS_READ_CALLBACKS
+       bool
+
 source "fs/ext2/Kconfig"
 source "fs/ext4/Kconfig"
 source "fs/jbd2/Kconfig"
fs/Makefile +2 -0
index c9aea23aba56..a1a06f0db5c1 100644
@@ -21,6 +21,8 @@ else
 obj-y +=	no-block.o
 endif
 
+obj-$(CONFIG_FS_READ_CALLBACKS) += read_callbacks.o
+
 obj-$(CONFIG_PROC_FS) += proc_namespace.o
 
 obj-y				+= notify/
fs/crypto/Kconfig +1 -0
index 24ed99e2eca0..7752f9964280 100644
@@ -9,6 +9,7 @@ config FS_ENCRYPTION
 	select CRYPTO_CTS
 	select CRYPTO_SHA256
 	select KEYS
+	select FS_READ_CALLBACKS if BLOCK
 	help
 	  Enable encryption of files and directories.  This
 	  feature is similar to ecryptfs, but it is more memory
fs/crypto/bio.c +11 -0
index 82da2510721f..f677ff93d464 100644
@@ -24,6 +24,7 @@
 #include <linux/module.h>
 #include <linux/bio.h>
 #include <linux/namei.h>
+#include <linux/read_callbacks.h>
 #include "fscrypt_private.h"
 
 static void __fscrypt_decrypt_bio(struct bio *bio, bool done)
@@ -68,6 +69,16 @@ void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx, struct bio *bio)
 }
 EXPORT_SYMBOL(fscrypt_enqueue_decrypt_bio);
 
+void fscrypt_decrypt_work(struct work_struct *work)
+{
+	struct read_callbacks_ctx *ctx =
+		container_of(work, struct read_callbacks_ctx, work);
+
+	fscrypt_decrypt_bio(ctx->bio);
+
+	read_callbacks(ctx);
+}
+
 int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
 				sector_t pblk, unsigned int len)
 {
fs/read_callbacks.c new file +174 -0

174 lines not on this page · raw message

include/linux/fscrypt.h +5 -0
index bd8f207a2fb6..159b8ddcd670 100644
@@ -235,6 +235,7 @@ static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
 extern void fscrypt_decrypt_bio(struct bio *);
 extern void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx,
 					struct bio *bio);
+extern void fscrypt_decrypt_work(struct work_struct *work);
 extern int fscrypt_zeroout_range(const struct inode *, pgoff_t, sector_t,
 				 unsigned int);
 
@@ -440,6 +441,10 @@ static inline void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx,
 {
 }
 
+static inline void fscrypt_decrypt_work(struct work_struct *work)
+{
+}
+
 static inline int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
 					sector_t pblk, unsigned int len)
 {
include/linux/read_callbacks.h new file +38 -0
new file mode 100644
index 000000000000..aa1ec8ed7a6a
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _READ_CALLBACKS_H
+#define _READ_CALLBACKS_H
+
+typedef void (*end_page_op_t)(struct bio *bio, struct page *page);
+
+struct read_callbacks_ctx {
+	struct bio *bio;
+	struct inode *inode;
+	struct work_struct work;
+	unsigned int cur_step;
+	unsigned int enabled_steps;
+	end_page_op_t page_op;
+};
+
+#ifdef CONFIG_FS_READ_CALLBACKS
+void read_callbacks(struct read_callbacks_ctx *ctx);
+int read_callbacks_end_bio(struct bio *bio);
+int read_callbacks_setup(struct inode *inode, struct bio *bio,
+			end_page_op_t page_op);
+#else
+static inline void read_callbacks(struct read_callbacks_ctx *ctx)
+{
+}
+
+static inline int read_callbacks_end_bio(struct bio *bio)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int read_callbacks_setup(struct inode *inode,
+				struct bio *bio, end_page_op_t page_op)
+{
+	return -EOPNOTSUPP;
+}
+#endif
+
+#endif	/* _READ_CALLBACKS_H */
Signature · 1 line
2.19.1
#4 lore raw

PATCH v3 2/7 Integrate read callbacks into Ext4 and F2FS

In reply to Chandan Rajendra

This commit gets Ext4 and F2FS to make use of read callbacks API to perform decryption of file data read from the disk.

---
 fs/crypto/bio.c             |  30 +--------
 fs/crypto/crypto.c          |   1 +
 fs/crypto/fscrypt_private.h |   3 +
 fs/ext4/readpage.c          |  29 +++------
 fs/f2fs/data.c              | 124 +++++++-----------------------------
 fs/f2fs/super.c             |   9 +--
 fs/read_callbacks.c         |   1 -
 include/linux/fscrypt.h     |  18 ------
 8 files changed, 40 insertions(+), 175 deletions(-)
fs/crypto/bio.c +1 -29
index f677ff93d464..4076d704e2e4 100644
@@ -27,7 +27,7 @@
 #include <linux/read_callbacks.h>
 #include "fscrypt_private.h"
 
-static void __fscrypt_decrypt_bio(struct bio *bio, bool done)
+static void fscrypt_decrypt_bio(struct bio *bio)
 {
 	struct bio_vec *bv;
 	struct bvec_iter_all iter_all;
@@ -38,37 +38,9 @@ static void __fscrypt_decrypt_bio(struct bio *bio, bool done)
 							   bv->bv_offset);
 		if (ret)
 			SetPageError(page);
-		else if (done)
-			SetPageUptodate(page);
-		if (done)
-			unlock_page(page);
 	}
 }
 
-void fscrypt_decrypt_bio(struct bio *bio)
-{
-	__fscrypt_decrypt_bio(bio, false);
-}
-EXPORT_SYMBOL(fscrypt_decrypt_bio);
-
-static void completion_pages(struct work_struct *work)
-{
-	struct fscrypt_ctx *ctx = container_of(work, struct fscrypt_ctx, work);
-	struct bio *bio = ctx->bio;
-
-	__fscrypt_decrypt_bio(bio, true);
-	fscrypt_release_ctx(ctx);
-	bio_put(bio);
-}
-
-void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx, struct bio *bio)
-{
-	INIT_WORK(&ctx->work, completion_pages);
-	ctx->bio = bio;
-	fscrypt_enqueue_decrypt_work(&ctx->work);
-}
-EXPORT_SYMBOL(fscrypt_enqueue_decrypt_bio);
-
 void fscrypt_decrypt_work(struct work_struct *work)
 {
 	struct read_callbacks_ctx *ctx =
fs/crypto/crypto.c +1 -0
index 45c3d0427fb2..e0aa20cfcaa7 100644
@@ -54,6 +54,7 @@ struct kmem_cache *fscrypt_info_cachep;
 
 void fscrypt_enqueue_decrypt_work(struct work_struct *work)
 {
+	INIT_WORK(work, fscrypt_decrypt_work);
 	queue_work(fscrypt_read_workqueue, work);
 }
 EXPORT_SYMBOL(fscrypt_enqueue_decrypt_work);
fs/crypto/fscrypt_private.h +3 -0
index 8978eec9d766..0bd65921efbf 100644
@@ -113,6 +113,9 @@ static inline bool fscrypt_valid_enc_modes(u32 contents_mode,
 	return false;
 }
 
+/* bio.c */
+void fscrypt_decrypt_work(struct work_struct *work);
+
 /* crypto.c */
 extern struct kmem_cache *fscrypt_info_cachep;
 extern int fscrypt_initialize(unsigned int cop_flags);
fs/ext4/readpage.c +10 -19
index c916017db334..652796b95545 100644
@@ -44,6 +44,7 @@
 #include <linux/backing-dev.h>
 #include <linux/pagevec.h>
 #include <linux/cleancache.h>
+#include <linux/read_callbacks.h>
 
 #include "ext4.h"
 
@@ -73,14 +74,9 @@ static void mpage_end_io(struct bio *bio)
 	struct bio_vec *bv;
 	struct bvec_iter_all iter_all;
 
-	if (ext4_bio_encrypted(bio)) {
-		if (bio->bi_status) {
-			fscrypt_release_ctx(bio->bi_private);
-		} else {
-			fscrypt_enqueue_decrypt_bio(bio->bi_private, bio);
-			return;
-		}
-	}
+	if (read_callbacks_end_bio(bio))
+		return;
+
 	bio_for_each_segment_all(bv, bio, iter_all) {
 		struct page *page = bv->bv_page;
 
@@ -241,24 +237,19 @@ int ext4_mpage_readpages(struct address_space *mapping,
 			bio = NULL;
 		}
 		if (bio == NULL) {
-			struct fscrypt_ctx *ctx = NULL;
-
-			if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode)) {
-				ctx = fscrypt_get_ctx(GFP_NOFS);
-				if (IS_ERR(ctx))
-					goto set_error_page;
-			}
 			bio = bio_alloc(GFP_KERNEL,
 				min_t(int, nr_pages, BIO_MAX_PAGES));
-			if (!bio) {
-				if (ctx)
-					fscrypt_release_ctx(ctx);
+			if (!bio)
+				goto set_error_page;
+
+			if (read_callbacks_setup(inode, bio, NULL)) {
+				bio_put(bio);
 				goto set_error_page;
 			}
+
 			bio_set_dev(bio, bdev);
 			bio->bi_iter.bi_sector = blocks[0] << (blkbits - 9);
 			bio->bi_end_io = mpage_end_io;
-			bio->bi_private = ctx;
 			bio_set_op_attrs(bio, REQ_OP_READ,
 						is_readahead ? REQ_RAHEAD : 0);
 		}
fs/f2fs/data.c +23 -101

176 lines not on this page · raw message

fs/f2fs/super.c +2 -7
index 6b959bbb336a..e2e6a8c7236c 100644
@@ -3627,15 +3627,11 @@ static int __init init_f2fs_fs(void)
 	err = register_filesystem(&f2fs_fs_type);
 	if (err)
 		goto free_shrinker;
+
 	f2fs_create_root_stats();
-	err = f2fs_init_post_read_processing();
-	if (err)
-		goto free_root_stats;
+
 	return 0;
 
-free_root_stats:
-	f2fs_destroy_root_stats();
-	unregister_filesystem(&f2fs_fs_type);
 free_shrinker:
 	unregister_shrinker(&f2fs_shrinker_info);
 free_sysfs:
@@ -3656,7 +3652,6 @@ static int __init init_f2fs_fs(void)
 
 static void __exit exit_f2fs_fs(void)
 {
-	f2fs_destroy_post_read_processing();
 	f2fs_destroy_root_stats();
 	unregister_filesystem(&f2fs_fs_type);
 	unregister_shrinker(&f2fs_shrinker_info);
fs/read_callbacks.c +0 -1
index a4196e3de05f..4b7fc2a349cd 100644
@@ -76,7 +76,6 @@ void read_callbacks(struct read_callbacks_ctx *ctx)
 	switch (++ctx->cur_step) {
 	case STEP_DECRYPT:
 		if (ctx->enabled_steps & (1 << STEP_DECRYPT)) {
-			INIT_WORK(&ctx->work, fscrypt_decrypt_work);
 			fscrypt_enqueue_decrypt_work(&ctx->work);
 			return;
 		}
include/linux/fscrypt.h +0 -18
index 159b8ddcd670..4ec0bd7cc0ba 100644
@@ -232,10 +232,6 @@ static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
 }
 
 /* bio.c */
-extern void fscrypt_decrypt_bio(struct bio *);
-extern void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx,
-					struct bio *bio);
-extern void fscrypt_decrypt_work(struct work_struct *work);
 extern int fscrypt_zeroout_range(const struct inode *, pgoff_t, sector_t,
 				 unsigned int);
 
@@ -431,20 +427,6 @@ static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
 	return !memcmp(de_name, fname->disk_name.name, fname->disk_name.len);
 }
 
-/* bio.c */
-static inline void fscrypt_decrypt_bio(struct bio *bio)
-{
-}
-
-static inline void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx,
-					       struct bio *bio)
-{
-}
-
-static inline void fscrypt_decrypt_work(struct work_struct *work)
-{
-}
-
 static inline int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
 					sector_t pblk, unsigned int len)
 {
Signature · 1 line
2.19.1
#7 lore raw

PATCH v3 3/7 fscrypt: remove struct fscrypt_ctx

In reply to Chandan Rajendra

Commit "fscrypt: remove the 'write' part of struct fscrypt_ctx" reduced "struct fscrypt_ctx" to be used only for decryption. With "read callbacks" being integrated into Ext4 and F2FS, we don't use "struct fscrypt_ctx" anymore. Hence this commit removes the structure and the associated code.

---
 fs/crypto/crypto.c      | 89 +----------------------------------------
 include/linux/fscrypt.h | 25 ------------
 2 files changed, 2 insertions(+), 112 deletions(-)
fs/crypto/crypto.c +2 -87
index e0aa20cfcaa7..18c6b664b042 100644
@@ -32,24 +32,16 @@
 #include "fscrypt_private.h"
 
 static unsigned int num_prealloc_crypto_pages = 32;
-static unsigned int num_prealloc_crypto_ctxs = 128;
 
 module_param(num_prealloc_crypto_pages, uint, 0444);
 MODULE_PARM_DESC(num_prealloc_crypto_pages,
 		"Number of crypto pages to preallocate");
-module_param(num_prealloc_crypto_ctxs, uint, 0444);
-MODULE_PARM_DESC(num_prealloc_crypto_ctxs,
-		"Number of crypto contexts to preallocate");
 
 static mempool_t *fscrypt_bounce_page_pool = NULL;
 
-static LIST_HEAD(fscrypt_free_ctxs);
-static DEFINE_SPINLOCK(fscrypt_ctx_lock);
-
 static struct workqueue_struct *fscrypt_read_workqueue;
 static DEFINE_MUTEX(fscrypt_init_mutex);
 
-static struct kmem_cache *fscrypt_ctx_cachep;
 struct kmem_cache *fscrypt_info_cachep;
 
 void fscrypt_enqueue_decrypt_work(struct work_struct *work)
@@ -59,62 +51,6 @@ void fscrypt_enqueue_decrypt_work(struct work_struct *work)
 }
 EXPORT_SYMBOL(fscrypt_enqueue_decrypt_work);
 
-/**
- * fscrypt_release_ctx() - Release a decryption context
- * @ctx: The decryption context to release.
- *
- * If the decryption context was allocated from the pre-allocated pool, return
- * it to that pool.  Else, free it.
- */
-void fscrypt_release_ctx(struct fscrypt_ctx *ctx)
-{
-	unsigned long flags;
-
-	if (ctx->flags & FS_CTX_REQUIRES_FREE_ENCRYPT_FL) {
-		kmem_cache_free(fscrypt_ctx_cachep, ctx);
-	} else {
-		spin_lock_irqsave(&fscrypt_ctx_lock, flags);
-		list_add(&ctx->free_list, &fscrypt_free_ctxs);
-		spin_unlock_irqrestore(&fscrypt_ctx_lock, flags);
-	}
-}
-EXPORT_SYMBOL(fscrypt_release_ctx);
-
-/**
- * fscrypt_get_ctx() - Get a decryption context
- * @gfp_flags:   The gfp flag for memory allocation
- *
- * Allocate and initialize a decryption context.
- *
- * Return: A new decryption context on success; an ERR_PTR() otherwise.
- */
-struct fscrypt_ctx *fscrypt_get_ctx(gfp_t gfp_flags)
-{
-	struct fscrypt_ctx *ctx;
-	unsigned long flags;
-
-	/*
-	 * First try getting a ctx from the free list so that we don't have to
-	 * call into the slab allocator.
-	 */
-	spin_lock_irqsave(&fscrypt_ctx_lock, flags);
-	ctx = list_first_entry_or_null(&fscrypt_free_ctxs,
-					struct fscrypt_ctx, free_list);
-	if (ctx)
-		list_del(&ctx->free_list);
-	spin_unlock_irqrestore(&fscrypt_ctx_lock, flags);
-	if (!ctx) {
-		ctx = kmem_cache_zalloc(fscrypt_ctx_cachep, gfp_flags);
-		if (!ctx)
-			return ERR_PTR(-ENOMEM);
-		ctx->flags |= FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	} else {
-		ctx->flags &= ~FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	}
-	return ctx;
-}
-EXPORT_SYMBOL(fscrypt_get_ctx);
-
 struct page *fscrypt_alloc_bounce_page(gfp_t gfp_flags)
 {
 	return mempool_alloc(fscrypt_bounce_page_pool, gfp_flags);
@@ -401,11 +337,6 @@ const struct dentry_operations fscrypt_d_ops = {
 
 static void fscrypt_destroy(void)
 {
-	struct fscrypt_ctx *pos, *n;
-
-	list_for_each_entry_safe(pos, n, &fscrypt_free_ctxs, free_list)
-		kmem_cache_free(fscrypt_ctx_cachep, pos);
-	INIT_LIST_HEAD(&fscrypt_free_ctxs);
 	mempool_destroy(fscrypt_bounce_page_pool);
 	fscrypt_bounce_page_pool = NULL;
 }
@@ -421,7 +352,7 @@ static void fscrypt_destroy(void)
  */
 int fscrypt_initialize(unsigned int cop_flags)
 {
-	int i, res = -ENOMEM;
+	int res = -ENOMEM;
 
 	/* No need to allocate a bounce page pool if this FS won't use it. */
 	if (cop_flags & FS_CFLG_OWN_PAGES)
@@ -431,15 +362,6 @@ int fscrypt_initialize(unsigned int cop_flags)
 	if (fscrypt_bounce_page_pool)
 		goto already_initialized;
 
-	for (i = 0; i < num_prealloc_crypto_ctxs; i++) {
-		struct fscrypt_ctx *ctx;
-
-		ctx = kmem_cache_zalloc(fscrypt_ctx_cachep, GFP_NOFS);
-		if (!ctx)
-			goto fail;
-		list_add(&ctx->free_list, &fscrypt_free_ctxs);
-	}
-
 	fscrypt_bounce_page_pool =
 		mempool_create_page_pool(num_prealloc_crypto_pages, 0);
 	if (!fscrypt_bounce_page_pool)
@@ -494,18 +416,12 @@ static int __init fscrypt_init(void)
 	if (!fscrypt_read_workqueue)
 		goto fail;
 
-	fscrypt_ctx_cachep = KMEM_CACHE(fscrypt_ctx, SLAB_RECLAIM_ACCOUNT);
-	if (!fscrypt_ctx_cachep)
-		goto fail_free_queue;
-
 	fscrypt_info_cachep = KMEM_CACHE(fscrypt_info, SLAB_RECLAIM_ACCOUNT);
 	if (!fscrypt_info_cachep)
-		goto fail_free_ctx;
+		goto fail_free_queue;
 
 	return 0;
 
-fail_free_ctx:
-	kmem_cache_destroy(fscrypt_ctx_cachep);
 fail_free_queue:
 	destroy_workqueue(fscrypt_read_workqueue);
 fail:
@@ -522,7 +438,6 @@ static void __exit fscrypt_exit(void)
 
 	if (fscrypt_read_workqueue)
 		destroy_workqueue(fscrypt_read_workqueue);
-	kmem_cache_destroy(fscrypt_ctx_cachep);
 	kmem_cache_destroy(fscrypt_info_cachep);
 
 	fscrypt_essiv_cleanup();
include/linux/fscrypt.h +0 -25
index 4ec0bd7cc0ba..13012c589aa3 100644
@@ -19,7 +19,6 @@
 
 #define FS_CRYPTO_BLOCK_SIZE		16
 
-struct fscrypt_ctx;
 struct fscrypt_info;
 
 struct fscrypt_str {
@@ -63,18 +62,6 @@ struct fscrypt_operations {
 	unsigned int max_namelen;
 };
 
-/* Decryption work */
-struct fscrypt_ctx {
-	union {
-		struct {
-			struct bio *bio;
-			struct work_struct work;
-		};
-		struct list_head free_list;	/* Free list */
-	};
-	u8 flags;				/* Flags */
-};
-
 static inline bool fscrypt_has_encryption_key(const struct inode *inode)
 {
 	/* pairs with cmpxchg_release() in fscrypt_get_encryption_info() */
@@ -101,8 +88,6 @@ static inline void fscrypt_handle_d_move(struct dentry *dentry)
 
 /* crypto.c */
 extern void fscrypt_enqueue_decrypt_work(struct work_struct *);
-extern struct fscrypt_ctx *fscrypt_get_ctx(gfp_t);
-extern void fscrypt_release_ctx(struct fscrypt_ctx *);
 
 extern struct page *fscrypt_encrypt_pagecache_blocks(struct page *page,
 						     unsigned int len,
@@ -281,16 +266,6 @@ static inline void fscrypt_enqueue_decrypt_work(struct work_struct *work)
 {
 }
 
-static inline struct fscrypt_ctx *fscrypt_get_ctx(gfp_t gfp_flags)
-{
-	return ERR_PTR(-EOPNOTSUPP);
-}
-
-static inline void fscrypt_release_ctx(struct fscrypt_ctx *ctx)
-{
-	return;
-}
-
 static inline struct page *fscrypt_encrypt_pagecache_blocks(struct page *page,
 							    unsigned int len,
 							    unsigned int offs,
Signature · 1 line
2.19.1
#9 lore raw

PATCH v3 4/7 fs/mpage.c: Integrate read callbacks

In reply to Chandan Rajendra

This commit adds code to make do_mpage_readpage() to be "read callbacks" aware i.e. for files requiring decryption, do_mpage_readpage() now sets up the read callbacks state machine when allocating a bio and later starts execution of the state machine after file data is read from the underlying disk.

---
 fs/mpage.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
fs/mpage.c +10 -1
index 436a85260394..611ad122fc92 100644
@@ -30,6 +30,7 @@
 #include <linux/backing-dev.h>
 #include <linux/pagevec.h>
 #include <linux/cleancache.h>
+#include <linux/read_callbacks.h>
 #include "internal.h"
 
 /*
@@ -49,6 +50,8 @@ static void mpage_end_io(struct bio *bio)
 	struct bio_vec *bv;
 	struct bvec_iter_all iter_all;
 
+	if (read_callbacks_end_bio(bio))
+		return;
 	bio_for_each_segment_all(bv, bio, iter_all) {
 		struct page *page = bv->bv_page;
 		page_endio(page, bio_op(bio),
@@ -309,6 +312,12 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 					gfp);
 		if (args->bio == NULL)
 			goto confused;
+
+		if (read_callbacks_setup(inode, args->bio, NULL)) {
+			bio_put(args->bio);
+			args->bio = NULL;
+			goto confused;
+		}
 	}
 
 	length = first_hole << blkbits;
@@ -330,7 +339,7 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 confused:
 	if (args->bio)
 		args->bio = mpage_bio_submit(REQ_OP_READ, op_flags, args->bio);
-	if (!PageUptodate(page))
+	if (!PageUptodate(page) && !PageError(page))
 		block_read_full_page(page, args->get_block);
 	else
 		unlock_page(page);
Signature · 1 line
2.19.1
#12 lore raw

PATCH v3 5/7 ext4: Wire up ext4_readpage[s] to use mpage_readpage[s]

In reply to Chandan Rajendra

Now that do_mpage_readpage() is "post read process" aware, this commit gets ext4_readpage[s] to use mpage_readpage[s] and deletes ext4's readpage.c since the associated functionality is not required anymore.

---
 fs/ext4/Makefile   |   2 +-
 fs/ext4/inode.c    |   5 +-
 fs/ext4/readpage.c | 286 ---------------------------------------------
 3 files changed, 3 insertions(+), 290 deletions(-)
 delete mode 100644 fs/ext4/readpage.c
fs/ext4/Makefile +1 -1
index 8fdfcd3c3e04..7c38803a808d 100644
@@ -8,7 +8,7 @@ obj-$(CONFIG_EXT4_FS) += ext4.o
 ext4-y	:= balloc.o bitmap.o block_validity.o dir.o ext4_jbd2.o extents.o \
 		extents_status.o file.o fsmap.o fsync.o hash.o ialloc.o \
 		indirect.o inline.o inode.o ioctl.o mballoc.o migrate.o \
-		mmp.o move_extent.o namei.o page-io.o readpage.o resize.o \
+		mmp.o move_extent.o namei.o page-io.o resize.o \
 		super.o symlink.o sysfs.o xattr.o xattr_trusted.o xattr_user.o
 
 ext4-$(CONFIG_EXT4_FS_POSIX_ACL)	+= acl.o
fs/ext4/inode.c +2 -3
index f65357735a1a..60010b76b566 100644
@@ -3363,8 +3363,7 @@ static int ext4_readpage(struct file *file, struct page *page)
 		ret = ext4_readpage_inline(inode, page);
 
 	if (ret == -EAGAIN)
-		return ext4_mpage_readpages(page->mapping, NULL, page, 1,
-						false);
+		return mpage_readpage(page, ext4_get_block);
 
 	return ret;
 }
@@ -3379,7 +3378,7 @@ ext4_readpages(struct file *file, struct address_space *mapping,
 	if (ext4_has_inline_data(inode))
 		return 0;
 
-	return ext4_mpage_readpages(mapping, pages, NULL, nr_pages, true);
+	return mpage_readpages(mapping, pages, nr_pages, ext4_get_block);
 }
 
 static void ext4_invalidatepage(struct page *page, unsigned int offset,
fs/ext4/readpage.c deleted +0 -286

286 lines not on this page · raw message

Signature · 1 line
2.19.1
#13 lore raw

PATCH v3 6/7 Add decryption support for sub-pagesized blocks

In reply to Chandan Rajendra

To support decryption of sub-pagesized blocks this commit adds code to,
1. Track buffer head in "struct read_callbacks_ctx".
2. Pass buffer head argument to all read callbacks.
3. Add new fscrypt helper to decrypt the file data referred to by a buffer head.

---
 fs/buffer.c                    |  55 +++++++++------
 fs/crypto/bio.c                |  21 +++++-
 fs/f2fs/data.c                 |   2 +-
 fs/mpage.c                     |   2 +-
 fs/read_callbacks.c            | 118 +++++++++++++++++++++++++--------
 include/linux/buffer_head.h    |   1 +
 include/linux/read_callbacks.h |  13 +++-
 7 files changed, 158 insertions(+), 54 deletions(-)
fs/buffer.c +36 -19
index e450c55f6434..dcb67525dac9 100644
@@ -46,6 +46,7 @@
 #include <linux/bit_spinlock.h>
 #include <linux/pagevec.h>
 #include <linux/sched/mm.h>
+#include <linux/read_callbacks.h>
 #include <trace/events/block.h>
 
 static int fsync_buffers_list(spinlock_t *lock, struct list_head *list);
@@ -246,11 +247,7 @@ __find_get_block_slow(struct block_device *bdev, sector_t block)
 	return ret;
 }
 
-/*
- * I/O completion handler for block_read_full_page() - pages
- * which come unlocked at the end of I/O.
- */
-static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
+void end_buffer_async_read(struct buffer_head *bh)
 {
 	unsigned long flags;
 	struct buffer_head *first;
@@ -258,17 +255,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	struct page *page;
 	int page_uptodate = 1;
 
-	BUG_ON(!buffer_async_read(bh));
-
 	page = bh->b_page;
-	if (uptodate) {
-		set_buffer_uptodate(bh);
-	} else {
-		clear_buffer_uptodate(bh);
-		buffer_io_error(bh, ", async page read");
-		SetPageError(page);
-	}
-
 	/*
 	 * Be _very_ careful from here on. Bad things can happen if
 	 * two buffer heads end IO at almost the same time and both
@@ -307,6 +294,31 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	return;
 }
 
+/*
+ * I/O completion handler for block_read_full_page().  Pages are unlocked
+ * after the I/O completes and the read callbacks (if any) have executed.
+ */
+static void __end_buffer_async_read(struct buffer_head *bh, int uptodate)
+{
+	struct page *page;
+
+	BUG_ON(!buffer_async_read(bh));
+
+	if (read_callbacks_end_bh(bh, uptodate))
+		return;
+
+	page = bh->b_page;
+	if (uptodate) {
+		set_buffer_uptodate(bh);
+	} else {
+		clear_buffer_uptodate(bh);
+		buffer_io_error(bh, ", async page read");
+		SetPageError(page);
+	}
+
+	end_buffer_async_read(bh);
+}
+
 /*
  * Completion handler for block_write_full_page() - pages which are unlocked
  * during I/O, and which have PageWriteback cleared upon I/O completion.
@@ -379,7 +391,7 @@ EXPORT_SYMBOL(end_buffer_async_write);
  */
 static void mark_buffer_async_read(struct buffer_head *bh)
 {
-	bh->b_end_io = end_buffer_async_read;
+	bh->b_end_io = __end_buffer_async_read;
 	set_buffer_async_read(bh);
 }
 
@@ -2294,10 +2306,15 @@ int block_read_full_page(struct page *page, get_block_t *get_block)
 	 */
 	for (i = 0; i < nr; i++) {
 		bh = arr[i];
-		if (buffer_uptodate(bh))
-			end_buffer_async_read(bh, 1);
-		else
+		if (buffer_uptodate(bh)) {
+			__end_buffer_async_read(bh, 1);
+		} else {
+			if (WARN_ON(read_callbacks_setup(inode, NULL, bh, NULL))) {
+				__end_buffer_async_read(bh, 0);
+				continue;
+			}
 			submit_bh(REQ_OP_READ, 0, bh);
+		}
 	}
 	return 0;
 }
fs/crypto/bio.c +20 -1
index 4076d704e2e4..b836d648fd27 100644
@@ -24,6 +24,7 @@
 #include <linux/module.h>
 #include <linux/bio.h>
 #include <linux/namei.h>
+#include <linux/buffer_head.h>
 #include <linux/read_callbacks.h>
 #include "fscrypt_private.h"
 
@@ -41,12 +42,30 @@ static void fscrypt_decrypt_bio(struct bio *bio)
 	}
 }
 
+static void fscrypt_decrypt_bh(struct buffer_head *bh)
+{
+	struct page *page;
+	int ret;
+
+	page = bh->b_page;
+
+	ret = fscrypt_decrypt_pagecache_blocks(page, bh->b_size,
+					bh_offset(bh));
+	if (ret)
+		SetPageError(page);
+}
+
 void fscrypt_decrypt_work(struct work_struct *work)
 {
 	struct read_callbacks_ctx *ctx =
 		container_of(work, struct read_callbacks_ctx, work);
 
-	fscrypt_decrypt_bio(ctx->bio);
+	WARN_ON(!ctx->bh && !ctx->bio);
+
+	if (ctx->bio)
+		fscrypt_decrypt_bio(ctx->bio);
+	else
+		fscrypt_decrypt_bh(ctx->bh);
 
 	read_callbacks(ctx);
 }
fs/f2fs/data.c +1 -1
index 23b34399d809..1e8b1eb68a90 100644
@@ -520,7 +520,7 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 	bio->bi_end_io = f2fs_read_end_io;
 	bio_set_op_attrs(bio, REQ_OP_READ, op_flag);
 
-	ret = read_callbacks_setup(inode, bio, f2fs_end_page_op);
+	ret = read_callbacks_setup(inode, bio, NULL, f2fs_end_page_op);
 	if (ret) {
 		bio_put(bio);
 		return ERR_PTR(ret);
fs/mpage.c +1 -1
index 611ad122fc92..387c23b529eb 100644
@@ -313,7 +313,7 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 		if (args->bio == NULL)
 			goto confused;
 
-		if (read_callbacks_setup(inode, args->bio, NULL)) {
+		if (read_callbacks_setup(inode, args->bio, NULL, NULL)) {
 			bio_put(args->bio);
 			args->bio = NULL;
 			goto confused;
fs/read_callbacks.c +89 -29

163 lines not on this page · raw message

include/linux/buffer_head.h +1 -0
index 7b73ef7f902d..42d0d63c7a3b 100644
@@ -165,6 +165,7 @@ void create_empty_buffers(struct page *, unsigned long,
 void end_buffer_read_sync(struct buffer_head *bh, int uptodate);
 void end_buffer_write_sync(struct buffer_head *bh, int uptodate);
 void end_buffer_async_write(struct buffer_head *bh, int uptodate);
+void end_buffer_async_read(struct buffer_head *bh);
 
 /* Things to do with buffers at mapping->private_list */
 void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode);
include/linux/read_callbacks.h +10 -3
index aa1ec8ed7a6a..0b52d7961fb2 100644
@@ -5,6 +5,7 @@
 typedef void (*end_page_op_t)(struct bio *bio, struct page *page);
 
 struct read_callbacks_ctx {
+	struct buffer_head *bh;
 	struct bio *bio;
 	struct inode *inode;
 	struct work_struct work;
@@ -16,8 +17,9 @@ struct read_callbacks_ctx {
 #ifdef CONFIG_FS_READ_CALLBACKS
 void read_callbacks(struct read_callbacks_ctx *ctx);
 int read_callbacks_end_bio(struct bio *bio);
+int read_callbacks_end_bh(struct buffer_head *bh, int uptodate);
 int read_callbacks_setup(struct inode *inode, struct bio *bio,
-			end_page_op_t page_op);
+			struct buffer_head *bh, end_page_op_t page_op);
 #else
 static inline void read_callbacks(struct read_callbacks_ctx *ctx)
 {
@@ -28,8 +30,13 @@ static inline int read_callbacks_end_bio(struct bio *bio)
 	return -EOPNOTSUPP;
 }
 
-static inline int read_callbacks_setup(struct inode *inode,
-				struct bio *bio, end_page_op_t page_op)
+static inline int read_callbacks_end_bh(struct buffer_head *bh, int uptodate)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int read_callbacks_setup(struct inode *inode, struct bio *bio,
+				struct buffer_head *bh, end_page_op_t page_op)
 {
 	return -EOPNOTSUPP;
 }
Signature · 1 line
2.19.1
#16 lore raw

PATCH v3 7/7 ext4: Enable encryption for subpage-sized blocks

In reply to Chandan Rajendra

Now that we have the code to support encryption for subpage-sized blocks, this commit removes the conditional check in filesystem mount code.

The commit also changes the support statement in Documentation/filesystems/fscrypt.rst to reflect the fact that encryption of filesystems with blocksize less than page size now works.

---
 Documentation/filesystems/fscrypt.rst | 4 ++--
 fs/ext4/super.c                       | 7 -------
 2 files changed, 2 insertions(+), 9 deletions(-)
Documentation/filesystems/fscrypt.rst +2 -2
index 08c23b60e016..c3efe86bf2b2 100644
@@ -213,8 +213,8 @@ Contents encryption
 -------------------
 
 For file contents, each filesystem block is encrypted independently.
-Currently, only the case where the filesystem block size is equal to
-the system's page size (usually 4096 bytes) is supported.
+Starting from Linux kernel 5.4, encryption of filesystems with block
+size less than system's page size is supported.
 
 Each block's IV is set to the logical block number within the file as
 a little endian number, except that:
fs/ext4/super.c +0 -7
index 4079605d437a..63661a86d148 100644
@@ -4412,13 +4412,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		}
 	}
 
-	if ((DUMMY_ENCRYPTION_ENABLED(sbi) || ext4_has_feature_encrypt(sb)) &&
-	    (blocksize != PAGE_SIZE)) {
-		ext4_msg(sb, KERN_ERR,
-			 "Unsupported blocksize for fs encryption");
-		goto failed_mount_wq;
-	}
-
 	if (DUMMY_ENCRYPTION_ENABLED(sbi) && !sb_rdonly(sb) &&
 	    !ext4_has_feature_encrypt(sb)) {
 		ext4_set_feature_encrypt(sb);
Signature · 1 line
2.19.1
maintainer
#2 lore raw

Re: [PATCH V3 1/7] FS: Introduce read callbacks

In reply to Chandan Rajendra

Hi Chandan,

Quoted 26 lines · Chandan Rajendra

Read callbacks implements a state machine to be executed after a buffered read I/O is completed. They help in further processing the file data read from the backing store. Currently, decryption is the only post processing step to be supported.

The execution of the state machine is to be initiated by the endio function associated with the read operation.

---
 fs/Kconfig                     |   3 +
 fs/Makefile                    |   2 +
 fs/crypto/Kconfig              |   1 +
 fs/crypto/bio.c                |  11 +++
 fs/read_callbacks.c            | 174 +++++++++++++++++++++++++++++++++
 include/linux/fscrypt.h        |   5 +
 include/linux/read_callbacks.h |  38 +++++++
 7 files changed, 234 insertions(+)
 create mode 100644 fs/read_callbacks.c
 create mode 100644 include/linux/read_callbacks.h
fs/Kconfig +0 -0
index f1046cf6ad85..d869859c88da 100644

fs/Kconfig if BLOCK · line 25

Quoted 5 lines
fs/Kconfig +2 -0
@@ -21,6 +21,9 @@ if BLOCK
 config FS_IOMAP
 	bool
 
+config FS_READ_CALLBACKS
+       bool

This should be intended with a tab, not spaces.

Quoted 9 lines
fs/Kconfig +1 -0
+
 source "fs/ext2/Kconfig"
 source "fs/ext4/Kconfig"
 source "fs/jbd2/Kconfig"
fs/Makefile +0 -0
index c9aea23aba56..a1a06f0db5c1 100644

fs/Makefile else · line 26

Quoted 6 lines
fs/Makefile +2 -0
@@ -21,6 +21,8 @@ else
 obj-y +=	no-block.o
 endif
 
+obj-$(CONFIG_FS_READ_CALLBACKS) += read_callbacks.o
+
 obj-$(CONFIG_PROC_FS) += proc_namespace.o

Nit: maybe move this to just below the line for iomap.o, to be consistent with where FS_READ_CALLBACKS is in the Kconfig file.

Quoted 30 lines
fs/Makefile +0 -0
 obj-y				+= notify/
fs/crypto/Kconfig +1 -0
index 24ed99e2eca0..7752f9964280 100644
@@ -9,6 +9,7 @@ config FS_ENCRYPTION
 	select CRYPTO_CTS
 	select CRYPTO_SHA256
 	select KEYS
+	select FS_READ_CALLBACKS if BLOCK
 	help
 	  Enable encryption of files and directories.  This
 	  feature is similar to ecryptfs, but it is more memory
fs/crypto/bio.c +1 -0
index 82da2510721f..f677ff93d464 100644
@@ -24,6 +24,7 @@
 #include <linux/module.h>
 #include <linux/bio.h>
 #include <linux/namei.h>
+#include <linux/read_callbacks.h>
 #include "fscrypt_private.h"
 
 static void __fscrypt_decrypt_bio(struct bio *bio, bool done)
@@ -68,6 +69,16 @@ void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx, struct bio *bio)
 }
 EXPORT_SYMBOL(fscrypt_enqueue_decrypt_bio);
 

fs/crypto/bio.c void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx, struct bio *bio) · line 81

Quoted 10 lines
fs/crypto/bio.c +10 -0
+void fscrypt_decrypt_work(struct work_struct *work)
+{
+	struct read_callbacks_ctx *ctx =
+		container_of(work, struct read_callbacks_ctx, work);
+
+	fscrypt_decrypt_bio(ctx->bio);
+
+	read_callbacks(ctx);
+}
+

This seems like a layering violation, since it means that read_callbacks.c calls fs/crypto/ *and* fs/crypto/ calls read_callbacks.c. I don't think fs/crypto/ should be aware of read_callbacks at all. Instead we should have a clear ordering between the components: the filesystem uses read_callbacks.c and fs/crypto/, and read_callbacks.c uses fs/crypto/. So how about:

- Move fscrypt_decrypt_work(), fscrypt_decrypt_bh(), and fscrypt_decrypt_bio() into fs/read_callbacks.c and remove the "fscrypt_" prefix from them.

- Instead of selecting FS_READ_CALLBACKS from FS_ENCRYPTION, select it from EXT4_FS and F2FS_FS (if FS_ENCRYPTION). I.e., it's really the *filesystems* themselves that use read_callbacks, not fs/crypto/.

- Move the definition of read_callbacks_ctx into fs/read_callbacks.c, and make read_callbacks() static, so these are private to the read_callbacks component.

Quoted 27 lines
fs/crypto/bio.c +0 -0
 int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
 				sector_t pblk, unsigned int len)
 {
fs/read_callbacks.c new file +18 -0
new file mode 100644
index 000000000000..a4196e3de05f
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * This file tracks the state machine that needs to be executed after reading
+ * data from files that are encrypted and/or have verity metadata associated
+ * with them.
+ */
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/bio.h>
+#include <linux/fscrypt.h>
+#include <linux/read_callbacks.h>
+
+#define NUM_PREALLOC_READ_CALLBACK_CTXS	128
+
+static struct kmem_cache *read_callbacks_ctx_cache;
+static mempool_t *read_callbacks_ctx_pool;
+

fs/read_callbacks.c line 28

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+/* Read callback state machine steps */
+enum read_callbacks_step {
+	STEP_INITIAL = 0,
+	STEP_DECRYPT,
+};
+
+static void put_read_callbacks_ctx(struct read_callbacks_ctx *ctx)
+{
+	mempool_free(ctx, read_callbacks_ctx_pool);
+}

Maybe call this free_read_callbacks_ctx(), so that it doesn't sound like it's doing reference counting.

Quoted 20 lines
fs/read_callbacks.c new file +20 -0
+
+static void end_read_callbacks_bio(struct bio *bio)
+{
+	struct read_callbacks_ctx *ctx;
+	struct page *page;
+	struct bio_vec *bv;
+	struct bvec_iter_all iter_all;
+
+	ctx = bio->bi_private;
+
+	bio_for_each_segment_all(bv, bio, iter_all) {
+		page = bv->bv_page;
+
+		if (bio->bi_status || PageError(page)) {
+			ClearPageUptodate(page);
+			SetPageError(page);
+		} else {
+			SetPageUptodate(page);
+		}
+

fs/read_callbacks.c line 58

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+		if (ctx->page_op)
+			ctx->page_op(bio, page);
+
+		unlock_page(page);
+	}
+
+	put_read_callbacks_ctx(ctx);
+
+	bio_put(bio);
+}

The filesystem itself (or fs/mpage.c) actually has to implement almost all this logic as well anyway, because CONFIG_FS_READ_CALLBACKS may be unset. And the ->page_op() callback, which exists only because f2fs needs to update some counter, is very ugly.

IMO, it would be simpler to just make this whole function filesystem-specific, as a 'typedef void (*end_bio_func_t)(struct bio *bio);' which gets passed to the read_callbacks endio hook.

Of course, each end_bio_func_t would have to check PageError() to check whether any read_callbacks step failed. But to make that a bit easier and to make it get compiled out when CONFIG_FS_READ_CALLBACKS is unset, there could be a helper function in read_callbacks.h:

	#ifdef CONFIG_FS_READ_CALLBACKS
	static inline bool read_callbacks_failed(struct page *page)
	{
		return PageError(page);
	}
	#else
	static inline bool read_callbacks_failed(struct page *page)
	{
		return false;
	}
	#endif
Quoted 21 lines
fs/read_callbacks.c new file +21 -0
+
+/**
+ * read_callbacks() - Execute the read callbacks state machine.
+ * @ctx: The context structure tracking the current state.
+ *
+ * For each state, this function enqueues a work into appropriate subsystem's
+ * work queue. After performing further processing of the data in the bio's
+ * pages, the subsystem should invoke read_calbacks() to continue with the next
+ * state in the state machine.
+ */
+void read_callbacks(struct read_callbacks_ctx *ctx)
+{
+	/*
+	 * We use different work queues for decryption and for verity because
+	 * verity may require reading metadata pages that need decryption, and
+	 * we shouldn't recurse to the same workqueue.
+	 */
+	switch (++ctx->cur_step) {
+	case STEP_DECRYPT:
+		if (ctx->enabled_steps & (1 << STEP_DECRYPT)) {
+			INIT_WORK(&ctx->work, fscrypt_decrypt_work);

fs/read_callbacks.c line 89

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+			fscrypt_enqueue_decrypt_work(&ctx->work);
+			return;
+		}
+		ctx->cur_step++;
+		/* fall-through */
+	default:
+		end_read_callbacks_bio(ctx->bio);
+	}
+}
+EXPORT_SYMBOL(read_callbacks);

As I mentioned, I think the work functions should be defined in this file rather than in fs/crypto/ or fs/verity/, since they're specific to the read_callbacks. fs/crypto/ and fs/verity/ should not be aware of read_callbacks at all. Moreover, the 'read_callbacks()' function should be static.

Quoted 16 lines
fs/read_callbacks.c new file +16 -0
+
+/**
+ * read_callbacks_end_bio() - Initiate the read callbacks state machine.
+ * @bio: bio on which read I/O operation has just been completed.
+ *
+ * Initiates the execution of the read callbacks state machine when the read
+ * operation has been completed successfully. If there was an error associated
+ * with the bio, this function frees the read callbacks context structure stored
+ * in bio->bi_private (if any).
+ *
+ * Return: 1 to indicate that the bio has been handled (including unlocking the
+ * pages); 0 otherwise.
+ */
+int read_callbacks_end_bio(struct bio *bio)
+{
+	if (!bio->bi_status && bio->bi_private) {

fs/read_callbacks.c line 115

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+		read_callbacks((struct read_callbacks_ctx *)(bio->bi_private));
+		return 1;
+	}
+
+	if (bio->bi_private)
+		put_read_callbacks_ctx((struct read_callbacks_ctx *)(bio->bi_private));
+
+	return 0;
+}
+EXPORT_SYMBOL(read_callbacks_end_bio);

Okay, several issues here...

First, the naming of this is really confusing, since it's actually *beginning* the read callbacks, not ending them; and it's basically the same name as end_read_callbacks_bio(), which actually *is* for ending the read callbacks. Since this is the endio hook, how about calling it read_callbacks_endio_bio(), and likewise read_callbacks_endio_bh()?

But more importantly, this doesn't need to have a return value, since the read_callbacks layer has to know how to end the bio (meaning unlock the pages and mark them uptodate or not) *anyway*, or at least know how to ask the filesystem to do it. So it should just do it if needed, rather than returning 0 and making the caller do it.

Also just assign 'ctx = bio->bi_private' at the start of the function; no need to access the field 4 times and have unnecessary casts.

So IMO, the below would be much better:

void read_callbacks_endio_bio(struct bio *bio, end_bio_func_t end_bio)

{
	struct read_callbacks_ctx *ctx = bio->bi_private;

	if (ctx) {
		if (!bio->bi_status) {
			ctx->end_bio = end_bio;
			read_callbacks(ctx);
			return;
		}
		free_read_callbacks_ctx(ctx);
	}
	end_bio(bio);
}

EXPORT_SYMBOL(read_callbacks_endio_bio);

And then the !CONFIG_FS_READ_CALLBACKS stub:

static inline void read_callbacks_endio_bio(struct bio *bio,
end_bio_func_t end_bio)

{
	end_bio(bio);
}

Similarly for the buffer_head versions.

Quoted 29 lines
fs/read_callbacks.c new file +29 -0
+
+/**
+ * read_callbacks_setup() - Initialize the read callbacks state machine
+ * @inode: The file on which read I/O is performed.
+ * @bio: bio holding page cache pages on which read I/O is performed.
+ * @page_op: Function to perform filesystem specific operations on a page.
+ *
+ * Based on the nature of the file's data (e.g. encrypted), this function
+ * computes the steps that need to be performed after data is read of the
+ * backing disk. This information is saved in a context structure. A pointer
+ * to the context structure is then stored in bio->bi_private for later
+ * usage.
+ *
+ * Return: 0 on success; An error code on failure.
+ */
+int read_callbacks_setup(struct inode *inode, struct bio *bio,
+			end_page_op_t page_op)
+{
+	struct read_callbacks_ctx *ctx = NULL;
+	unsigned int enabled_steps = 0;
+
+	if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode))
+		enabled_steps |= 1 << STEP_DECRYPT;
+
+	if (enabled_steps) {
+		ctx = mempool_alloc(read_callbacks_ctx_pool, GFP_NOFS);
+		if (!ctx)
+			return -ENOMEM;
+		ctx->bio = bio;

fs/read_callbacks.c line 154

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+		ctx->inode = inode;
+		ctx->enabled_steps = enabled_steps;
+		ctx->cur_step = STEP_INITIAL;
+		ctx->page_op = page_op;
+		bio->bi_private = ctx;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(read_callbacks_setup);

Please call it:

read_callbacks_setup_bio()

Then when you add buffer_head support later in the patchset, rather than adding a buffer_head argument to this function, add a new function:

read_callbacks_setup_bh()

So that all the users don't have to pass both the buffer_head and bio arguments.

These can use a common function get_read_callbacks_ctx() that creates a read_callbacks_ctx for the inode. E.g., the bio version could look like:

int read_callbacks_setup_bio(struct inode *inode, struct bio *bio)

{
	struct read_callbacks_ctx *ctx = get_read_callbacks_ctx(inode);

	if (ctx) {
		if (IS_ERR(ctx))
			return PTR_ERR(ctx);
		ctx->bio = bio;
		ctx->bh = NULL;
		bio->bi_private = ctx;
	}
	return 0;
}

EXPORT_SYMBOL(read_callbacks_setup_bio);

Also, a nit: can you move the read_callbacks_setup_*() functions to near the top of the file, since they're called first (before the endio functions)? Likewise in read_callbacks.h. It would nice to keep the functions in a logical order.

include/linux/read_callbacks.h line 3

Quoted 9 lines
include/linux/read_callbacks.h new file +3 -0
new file mode 100644
index 000000000000..aa1ec8ed7a6a
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _READ_CALLBACKS_H
+#define _READ_CALLBACKS_H

Headers should be self-contained, but this is missing some prerequisite headers and forward declarations. Try compiling a .c file with this header included first.

include/linux/read_callbacks.h line 14

Quoted 11 lines
include/linux/read_callbacks.h new file +11 -0
+
+typedef void (*end_page_op_t)(struct bio *bio, struct page *page);
+
+struct read_callbacks_ctx {
+	struct bio *bio;
+	struct inode *inode;
+	struct work_struct work;
+	unsigned int cur_step;
+	unsigned int enabled_steps;
+	end_page_op_t page_op;
+};

As I mentioned, I don't think read_callbacks_ctx should be exposed to filesystems like this. Instead just forward declare it here, and put the actual definition in fs/read_callbacks.c.

Quoted 5 lines
include/linux/read_callbacks.h new file +5 -0
+
+#ifdef CONFIG_FS_READ_CALLBACKS
+void read_callbacks(struct read_callbacks_ctx *ctx);
+int read_callbacks_end_bio(struct bio *bio);
+int read_callbacks_setup(struct inode *inode, struct bio *bio,

include/linux/read_callbacks.h line 29

Quoted 10 lines
include/linux/read_callbacks.h new file +10 -0
+			end_page_op_t page_op);
+#else
+static inline void read_callbacks(struct read_callbacks_ctx *ctx)
+{
+}
+
+static inline int read_callbacks_end_bio(struct bio *bio)
+{
+	return -EOPNOTSUPP;
+}

This stub needs to return 0, otherwise it breaks fs/mpage.c and f2fs for everyone when CONFIG_FS_READ_CALLBACKS is unset.

But as I mentioned elsewhere, I think this should actually call an end_bio() callback itself and return void, which would also avoid this issue.

include/linux/read_callbacks.h line 35

Quoted 6 lines
include/linux/read_callbacks.h new file +6 -0
+
+static inline int read_callbacks_setup(struct inode *inode,
+				struct bio *bio, end_page_op_t page_op)
+{
+	return -EOPNOTSUPP;
+}

Similarly here, this stub needs to return 0.

Thanks!

- Eric

maintainer
#5 lore raw

Re: [PATCH V3 2/7] Integrate read callbacks into Ext4 and F2FS

In reply to Chandan Rajendra

Hi Chandan,

Quoted 13 lines · Chandan Rajendra

This commit gets Ext4 and F2FS to make use of read callbacks API to perform decryption of file data read from the disk.

---
 fs/crypto/bio.c             |  30 +--------
 fs/crypto/crypto.c          |   1 +
 fs/crypto/fscrypt_private.h |   3 +
 fs/ext4/readpage.c          |  29 +++------
 fs/f2fs/data.c              | 124 +++++++-----------------------------
 fs/f2fs/super.c             |   9 +--
 fs/read_callbacks.c         |   1 -
 include/linux/fscrypt.h     |  18 ------
 8 files changed, 40 insertions(+), 175 deletions(-)

This patch changes many different components. It would be much easier to review, and might get more attention from the other ext4 and f2fs developers, if it were split into 3 patches:

a. Convert ext4 to use read_callbacks.
b. Convert f2fs to use read_callbacks.
c. Remove the functions from fs/crypto/ that became unused as a result of
patches (a) and (b). (Actually, this part probably should be merged with the patch that removes the fscrypt_ctx, and the patch renamed to something like "fscrypt: remove decryption I/O path helpers")

Any reason why this wouldn't work? AFAICS, you couldn't do it only because you made this patch change fscrypt_enqueue_decrypt_work() to be responsible for initializing the work function. But as per my comments on patch 1, I don't think we should do that, since it would make much more sense to put the work function in read_callbacks.c.

However, since you're converting ext4 to use mpage_readpages() anyway, I don't think we should bother with the intermediate change to ext4_mpage_readpages(). It's useless, and that intermediate state of the ext4 code inevitably won't get tested very well. So perhaps order the whole series as:

- fs: introduce read_callbacks
- fs/mpage.c: add decryption support via read_callbacks
- fs/buffer.c: add decryption support via read_callbacks
- f2fs: convert to use read_callbacks
- ext4: convert to use mpage_readpages[s]
- ext4: support encryption with subpage-sized blocks
- fscrypt: remove decryption I/O path helpers

That order would also give the flexibility to possibly apply the fs/ changes first, without having to update both ext4 and f2fs simultaneously with them.

fs/f2fs/data.c static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, · line 514

Quoted 7 lines
fs/f2fs/data.c +1 -2
@@ -557,8 +511,7 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct bio *bio;
-	struct bio_post_read_ctx *ctx;
-	unsigned int post_read_steps = 0;
+	int ret;

Nit: 'err' rather than 'ret', since this is 0 or a -errno value.

Quoted 13 lines
fs/f2fs/data.c +0 -13
-int __init f2fs_init_post_read_processing(void)
-{
-	bio_post_read_ctx_cache = KMEM_CACHE(bio_post_read_ctx, 0);
-	if (!bio_post_read_ctx_cache)
-		goto fail;
-	bio_post_read_ctx_pool =
-		mempool_create_slab_pool(NUM_PREALLOC_POST_READ_CTXS,
-					 bio_post_read_ctx_cache);
-	if (!bio_post_read_ctx_pool)
-		goto fail_free_cache;
-	return 0;
-
-fail_free_cache:

fs/f2fs/data.c void f2fs_clear_page_cache_dirty_tag(struct page *page) · line 2928

Quoted 10 lines
fs/f2fs/data.c +0 -10
-	kmem_cache_destroy(bio_post_read_ctx_cache);
-fail:
-	return -ENOMEM;
-}
-
-void __exit f2fs_destroy_post_read_processing(void)
-{
-	mempool_destroy(bio_post_read_ctx_pool);
-	kmem_cache_destroy(bio_post_read_ctx_cache);
-}

Need to remove the declarations of these functions from fs/f2fs/f2fs.h to.

fs/read_callbacks.c void read_callbacks(struct read_callbacks_ctx *ctx) · line 81

Quoted 12 lines
fs/read_callbacks.c +0 -1
index a4196e3de05f..4b7fc2a349cd 100644
@@ -76,7 +76,6 @@ void read_callbacks(struct read_callbacks_ctx *ctx)
 	switch (++ctx->cur_step) {
 	case STEP_DECRYPT:
 		if (ctx->enabled_steps & (1 << STEP_DECRYPT)) {
-			INIT_WORK(&ctx->work, fscrypt_decrypt_work);
 			fscrypt_enqueue_decrypt_work(&ctx->work);
 			return;
 		}

Again, I think the work initialization should remain here as:

	INIT_WORK(&ctx->work, decrypt_work);

rather than moving it to fs/crypto/.

Thanks!

- Eric

maintainer
#10 lore raw

Re: [PATCH V3 4/7] fs/mpage.c: Integrate read callbacks

In reply to Chandan Rajendra
Quoted 56 lines · Chandan Rajendra

This commit adds code to make do_mpage_readpage() to be "read callbacks" aware i.e. for files requiring decryption, do_mpage_readpage() now sets up the read callbacks state machine when allocating a bio and later starts execution of the state machine after file data is read from the underlying disk.

---
 fs/mpage.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
fs/mpage.c +10 -1
index 436a85260394..611ad122fc92 100644
@@ -30,6 +30,7 @@
 #include <linux/backing-dev.h>
 #include <linux/pagevec.h>
 #include <linux/cleancache.h>
+#include <linux/read_callbacks.h>
 #include "internal.h"
 
 /*
@@ -49,6 +50,8 @@ static void mpage_end_io(struct bio *bio)
 	struct bio_vec *bv;
 	struct bvec_iter_all iter_all;
 
+	if (read_callbacks_end_bio(bio))
+		return;
 	bio_for_each_segment_all(bv, bio, iter_all) {
 		struct page *page = bv->bv_page;
 		page_endio(page, bio_op(bio),
@@ -309,6 +312,12 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 					gfp);
 		if (args->bio == NULL)
 			goto confused;
+
+		if (read_callbacks_setup(inode, args->bio, NULL)) {
+			bio_put(args->bio);
+			args->bio = NULL;
+			goto confused;
+		}
 	}
 
 	length = first_hole << blkbits;
@@ -330,7 +339,7 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 confused:
 	if (args->bio)
 		args->bio = mpage_bio_submit(REQ_OP_READ, op_flags, args->bio);
-	if (!PageUptodate(page))
+	if (!PageUptodate(page) && !PageError(page))
 		block_read_full_page(page, args->get_block);
 	else
 		unlock_page(page);
Signature · 1 line
2.19.1

Why is the !PageError() check needed here?

- Eric

_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

maintainer
#14 lore raw

Re: [PATCH V3 6/7] Add decryption support for sub-pagesized blocks

In reply to Chandan Rajendra
Quoted 7 lines · Chandan Rajendra

To support decryption of sub-pagesized blocks this commit adds code to,
1. Track buffer head in "struct read_callbacks_ctx".
2. Pass buffer head argument to all read callbacks.
3. Add new fscrypt helper to decrypt the file data referred to by a buffer head.

Quoted 10 lines
---
 fs/buffer.c                    |  55 +++++++++------
 fs/crypto/bio.c                |  21 +++++-
 fs/f2fs/data.c                 |   2 +-
 fs/mpage.c                     |   2 +-
 fs/read_callbacks.c            | 118 +++++++++++++++++++++++++--------
 include/linux/buffer_head.h    |   1 +
 include/linux/read_callbacks.h |  13 +++-
 7 files changed, 158 insertions(+), 54 deletions(-)

This is another patch that unnecessarily changes way too many components at once. My suggestions elsewhere would resolve this, though:

- This patch changes fs/f2fs/data.c and fs/mpage.c only to pass a NULL buffer_head to read_callbacks_setup(). But as per my comments on patch 1, read_callbacks_setup() should be split into read_callbacks_setup_bio() and read_callbacks_end_bh().

- This patch changes fs/crypto/ only to add support for the buffer_head decryption work. But as per my comments on patch 1, that should be in read_callbacks.c instead.

And adding buffer_head support to fs/read_callbacks.c should be its own patch, *or* should simply be folded into the patch that adds fs/read_callbacks.c.

Then the only thing remaining in this patch would be updating fs/buffer.c to make it use the read_callbacks, which should be retitled to something like "fs/buffer.c: add decryption support via read_callbacks".

- Eric

maintainer
#8 lore raw

Re: [PATCH V3 3/7] fscrypt: remove struct fscrypt_ctx

In reply to Chandan Rajendra
Quoted 24 lines · Chandan Rajendra
fs/crypto/crypto.c +0 -24
-/**
- * fscrypt_get_ctx() - Get a decryption context
- * @gfp_flags:   The gfp flag for memory allocation
- *
- * Allocate and initialize a decryption context.
- *
- * Return: A new decryption context on success; an ERR_PTR() otherwise.
- */
-struct fscrypt_ctx *fscrypt_get_ctx(gfp_t gfp_flags)
-{
-	struct fscrypt_ctx *ctx;
-	unsigned long flags;
-
-	/*
-	 * First try getting a ctx from the free list so that we don't have to
-	 * call into the slab allocator.
-	 */
-	spin_lock_irqsave(&fscrypt_ctx_lock, flags);
-	ctx = list_first_entry_or_null(&fscrypt_free_ctxs,
-					struct fscrypt_ctx, free_list);
-	if (ctx)
-		list_del(&ctx->free_list);
-	spin_unlock_irqrestore(&fscrypt_ctx_lock, flags);
-	if (!ctx) {

fs/crypto/crypto.c void fscrypt_enqueue_decrypt_work(struct work_struct *work) · line 116

Quoted 10 lines
fs/crypto/crypto.c +0 -10
-		ctx = kmem_cache_zalloc(fscrypt_ctx_cachep, gfp_flags);
-		if (!ctx)
-			return ERR_PTR(-ENOMEM);
-		ctx->flags |= FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	} else {
-		ctx->flags &= ~FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	}
-	return ctx;
-}
-EXPORT_SYMBOL(fscrypt_get_ctx);

FS_CTX_REQUIRES_FREE_ENCRYPT_FL is no longer used, so should be removed.

- Eric

_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

maintainer
#17 lore raw
In reply to Chandan Rajendra
Quoted 14 lines · Chandan Rajendra

This patchset moves the "FS read I/O callbacks" code into a file of its own (i.e. fs/read_callbacks.c) and modifies the generic do_mpage_readpge() to make use of the functionality provided.

"FS read I/O callbacks" code implements the state machine that needs to be executed after reading data from files that are encrypted and/or have verity metadata associated with them.

With these changes in place, the patchset changes Ext4 to use mpage_readpage[s] instead of its own custom ext4_readpage[s]() functions. This is done to reduce duplication of code across filesystems. Also, "FS read I/O callbacks" source files will be built only if CONFIG_FS_ENCRYPTION is enabled.

Quoted 6 lines

The patchset also modifies fs/buffer.c to get file encryption/decryption to work with subpage-sized blocks.

The patches can also be obtained from
https://github.com/chandanr/linux.git at branch subpage-encryption-v3.

FWIW: while doing my review I put together an (untested) incremental patch that addresses my comments on the code, so I've provided it below in case you want to start with it when addressing my comments.

This is just a single diff against your subpage-encryption-v3 branch, so of course it would still need to be folded into the appropriate patches. Also see my suggestions in reply to patch 2 about how to better organize the series. I also left TODOs in kerneldoc comments that still need to be updated.

fs/Kconfig +1 -1
index d869859c88da18..d95897a0f3d052 100644
@@ -22,7 +22,7 @@ config FS_IOMAP
 	bool
 
 config FS_READ_CALLBACKS
-       bool
+	bool
 
 source "fs/ext2/Kconfig"
 source "fs/ext4/Kconfig"
fs/Makefile +1 -2
index a1a06f0db5c181..81854d8161dea7 100644
@@ -21,8 +21,6 @@ else
 obj-y +=	no-block.o
 endif
 
-obj-$(CONFIG_FS_READ_CALLBACKS) += read_callbacks.o
-
 obj-$(CONFIG_PROC_FS) += proc_namespace.o
 
 obj-y				+= notify/
@@ -55,6 +53,7 @@ obj-$(CONFIG_SYSCTL)		+= drop_caches.o
 
 obj-$(CONFIG_FHANDLE)		+= fhandle.o
 obj-$(CONFIG_FS_IOMAP)		+= iomap.o
+obj-$(CONFIG_FS_READ_CALLBACKS)	+= read_callbacks.o
 
 obj-y				+= quota/
 
fs/buffer.c +17 -23
index dcb67525dac91d..bfda6af7ea1042 100644
@@ -247,7 +247,7 @@ __find_get_block_slow(struct block_device *bdev, sector_t block)
 	return ret;
 }
 
-void end_buffer_async_read(struct buffer_head *bh)
+void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 {
 	unsigned long flags;
 	struct buffer_head *first;
@@ -255,7 +255,17 @@ void end_buffer_async_read(struct buffer_head *bh)
 	struct page *page;
 	int page_uptodate = 1;
 
+	BUG_ON(!buffer_async_read(bh));
+
 	page = bh->b_page;
+	if (uptodate) {
+		set_buffer_uptodate(bh);
+	} else {
+		clear_buffer_uptodate(bh);
+		buffer_io_error(bh, ", async page read");
+		SetPageError(page);
+	}
+
 	/*
 	 * Be _very_ careful from here on. Bad things can happen if
 	 * two buffer heads end IO at almost the same time and both
@@ -298,25 +308,9 @@ void end_buffer_async_read(struct buffer_head *bh)
  * I/O completion handler for block_read_full_page().  Pages are unlocked
  * after the I/O completes and the read callbacks (if any) have executed.
  */
-static void __end_buffer_async_read(struct buffer_head *bh, int uptodate)
+static void end_buffer_async_read_cbs(struct buffer_head *bh, int uptodate)
 {
-	struct page *page;
-
-	BUG_ON(!buffer_async_read(bh));
-
-	if (read_callbacks_end_bh(bh, uptodate))
-		return;
-
-	page = bh->b_page;
-	if (uptodate) {
-		set_buffer_uptodate(bh);
-	} else {
-		clear_buffer_uptodate(bh);
-		buffer_io_error(bh, ", async page read");
-		SetPageError(page);
-	}
-
-	end_buffer_async_read(bh);
+	read_callbacks_endio_bh(bh, uptodate);
 }
 
 /*
@@ -391,7 +385,7 @@ EXPORT_SYMBOL(end_buffer_async_write);
  */
 static void mark_buffer_async_read(struct buffer_head *bh)
 {
-	bh->b_end_io = __end_buffer_async_read;
+	bh->b_end_io = end_buffer_async_read_cbs;
 	set_buffer_async_read(bh);
 }
 
@@ -2307,10 +2301,10 @@ int block_read_full_page(struct page *page, get_block_t *get_block)
 	for (i = 0; i < nr; i++) {
 		bh = arr[i];
 		if (buffer_uptodate(bh)) {
-			__end_buffer_async_read(bh, 1);
+			end_buffer_async_read(bh, 1);
 		} else {
-			if (WARN_ON(read_callbacks_setup(inode, NULL, bh, NULL))) {
-				__end_buffer_async_read(bh, 0);
+			if (read_callbacks_setup_bh(inode, bh)) {
+				end_buffer_async_read(bh, 0);
 				continue;
 			}
 			submit_bh(REQ_OP_READ, 0, bh);
fs/crypto/Kconfig +0 -1
index 7752f99642808b..24ed99e2eca0b2 100644
@@ -9,7 +9,6 @@ config FS_ENCRYPTION
 	select CRYPTO_CTS
 	select CRYPTO_SHA256
 	select KEYS
-	select FS_READ_CALLBACKS if BLOCK
 	help
 	  Enable encryption of files and directories.  This
 	  feature is similar to ecryptfs, but it is more memory
fs/crypto/bio.c +0 -44

52 lines not on this page · raw message

fs/crypto/crypto.c +0 -1
index 18c6b664b042fb..56697ba58f8676 100644
@@ -46,7 +46,6 @@ struct kmem_cache *fscrypt_info_cachep;
 
 void fscrypt_enqueue_decrypt_work(struct work_struct *work)
 {
-	INIT_WORK(work, fscrypt_decrypt_work);
 	queue_work(fscrypt_read_workqueue, work);
 }
 EXPORT_SYMBOL(fscrypt_enqueue_decrypt_work);
fs/crypto/fscrypt_private.h +0 -5
index 0bd65921efbf1c..702403dc761464 100644
@@ -93,8 +93,6 @@ typedef enum {
 	FS_ENCRYPT,
 } fscrypt_direction_t;
 
-#define FS_CTX_REQUIRES_FREE_ENCRYPT_FL		0x00000001
-
 static inline bool fscrypt_valid_enc_modes(u32 contents_mode,
 					   u32 filenames_mode)
 {
@@ -113,9 +111,6 @@ static inline bool fscrypt_valid_enc_modes(u32 contents_mode,
 	return false;
 }
 
-/* bio.c */
-void fscrypt_decrypt_work(struct work_struct *work);
-
 /* crypto.c */
 extern struct kmem_cache *fscrypt_info_cachep;
 extern int fscrypt_initialize(unsigned int cop_flags);
fs/ext4/Kconfig +1 -0
index cbb5ca830e57f3..c232aab7260ffd 100644
@@ -39,6 +39,7 @@ config EXT4_FS
 	select CRYPTO
 	select CRYPTO_CRC32C
 	select FS_IOMAP
+	select FS_READ_CALLBACKS if FS_ENCRYPTION
 	help
 	  This is the next generation of the ext3 filesystem.
 
fs/f2fs/Kconfig +1 -0
index 110a38ca5d53a7..c66072dee9bb5d 100644
@@ -5,6 +5,7 @@ config F2FS_FS
 	select CRYPTO
 	select CRYPTO_CRC32
 	select F2FS_FS_XATTR if FS_ENCRYPTION
+	select FS_READ_CALLBACKS if FS_ENCRYPTION
 	help
 	  F2FS is based on Log-structured File System (LFS), which supports
 	  versatile "flash-friendly" features. The design has been focused on
fs/f2fs/data.c +12 -27
index 1e8b1eb68a90a3..cf17b19b292557 100644
@@ -65,20 +65,6 @@ static enum count_type __read_io_type(struct page *page)
 	return F2FS_RD_DATA;
 }
 
-static void f2fs_end_page_op(struct bio *bio, struct page *page)
-{
-	BUG_ON(!PageLocked(page));
-
-	/* PG_error was set if any post_read step failed */
-	if (bio->bi_status || PageError(page)) {
-		ClearPageUptodate(page);
-		/* will re-read again later */
-		ClearPageError(page);
-	}
-
-	dec_page_count(F2FS_P_SB(page), __read_io_type(page));
-}
-
 static void __read_end_io(struct bio *bio)
 {
 	struct page *page;
@@ -88,14 +74,16 @@ static void __read_end_io(struct bio *bio)
 	bio_for_each_segment_all(bv, bio, iter_all) {
 		page = bv->bv_page;
 
-		if (!bio->bi_status && !PageError(page))
+		if (bio->bi_status || read_callbacks_failed(page)) {
+			ClearPageUptodate(page);
+			/* will re-read again later */
+			ClearPageError(page);
+		} else {
 			SetPageUptodate(page);
-
-		f2fs_end_page_op(bio, page);
-
+		}
+		dec_page_count(F2FS_P_SB(page), __read_io_type(page));
 		unlock_page(page);
 	}
-
 	bio_put(bio);
 }
 
@@ -107,10 +95,7 @@ static void f2fs_read_end_io(struct bio *bio)
 		bio->bi_status = BLK_STS_IOERR;
 	}
 
-	if (read_callbacks_end_bio(bio))
-		return;
-
-	__read_end_io(bio);
+	read_callbacks_endio_bio(bio, __read_end_io);
 }
 
 static void f2fs_write_end_io(struct bio *bio)
@@ -511,7 +496,7 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct bio *bio;
-	int ret;
+	int err;
 
 	bio = f2fs_bio_alloc(sbi, min_t(int, nr_pages, BIO_MAX_PAGES), false);
 	if (!bio)
@@ -520,10 +505,10 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 	bio->bi_end_io = f2fs_read_end_io;
 	bio_set_op_attrs(bio, REQ_OP_READ, op_flag);
 
-	ret = read_callbacks_setup(inode, bio, NULL, f2fs_end_page_op);
-	if (ret) {
+	err = read_callbacks_setup_bio(inode, bio);
+	if (err) {
 		bio_put(bio);
-		return ERR_PTR(ret);
+		return ERR_PTR(err);
 	}
 
 	return bio;
fs/f2fs/f2fs.h +0 -2
index 06b89a9862ab2b..182c1d91d150b9 100644
@@ -3161,8 +3161,6 @@ void f2fs_destroy_checkpoint_caches(void);
 /*
  * data.c
  */
-int f2fs_init_post_read_processing(void);
-void f2fs_destroy_post_read_processing(void);
 void f2fs_submit_merged_write(struct f2fs_sb_info *sbi, enum page_type type);
 void f2fs_submit_merged_write_cond(struct f2fs_sb_info *sbi,
 				struct inode *inode, struct page *page,
fs/f2fs/super.c +0 -2
index e2e6a8c7236c7f..2c8123b1587166 100644
@@ -3627,9 +3627,7 @@ static int __init init_f2fs_fs(void)
 	err = register_filesystem(&f2fs_fs_type);
 	if (err)
 		goto free_shrinker;
-
 	f2fs_create_root_stats();
-
 	return 0;
 
 free_shrinker:
fs/mpage.c +22 -14
index 387c23b529eb90..bf0549d9ce656e 100644
@@ -33,6 +33,25 @@
 #include <linux/read_callbacks.h>
 #include "internal.h"
 
+static void __mpage_end_io(struct bio *bio)
+{
+	struct bio_vec *bv;
+	struct bvec_iter_all iter_all;
+	bool is_write = (bio_op(bio) != REQ_OP_READ);
+	int err = blk_status_to_errno(bio->bi_status);
+
+	bio_for_each_segment_all(bv, bio, iter_all) {
+		struct page *page = bv->bv_page;
+		int this_err = err;
+
+		if (!err && !is_write && read_callbacks_failed(page))
+			this_err = -EIO;
+
+		page_endio(page, is_write, this_err);
+	}
+	bio_put(bio);
+}
+
 /*
  * I/O completion handler for multipage BIOs.
  *
@@ -47,18 +66,7 @@
  */
 static void mpage_end_io(struct bio *bio)
 {
-	struct bio_vec *bv;
-	struct bvec_iter_all iter_all;
-
-	if (read_callbacks_end_bio(bio))
-		return;
-	bio_for_each_segment_all(bv, bio, iter_all) {
-		struct page *page = bv->bv_page;
-		page_endio(page, bio_op(bio),
-			   blk_status_to_errno(bio->bi_status));
-	}
-
-	bio_put(bio);
+	read_callbacks_endio_bio(bio, __mpage_end_io);
 }
 
 static struct bio *mpage_bio_submit(int op, int op_flags, struct bio *bio)
@@ -313,7 +321,7 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 		if (args->bio == NULL)
 			goto confused;
 
-		if (read_callbacks_setup(inode, args->bio, NULL, NULL)) {
+		if (read_callbacks_setup_bio(inode, args->bio)) {
 			bio_put(args->bio);
 			args->bio = NULL;
 			goto confused;
@@ -339,7 +347,7 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 confused:
 	if (args->bio)
 		args->bio = mpage_bio_submit(REQ_OP_READ, op_flags, args->bio);
-	if (!PageUptodate(page) && !PageError(page))
+	if (!PageUptodate(page))
 		block_read_full_page(page, args->get_block);
 	else
 		unlock_page(page);
fs/read_callbacks.c +141 -146

355 lines not on this page · raw message

include/linux/buffer_head.h +1 -1
index 42d0d63c7a3b38..5176a7a7e22aab 100644
@@ -165,7 +165,7 @@ void create_empty_buffers(struct page *, unsigned long,
 void end_buffer_read_sync(struct buffer_head *bh, int uptodate);
 void end_buffer_write_sync(struct buffer_head *bh, int uptodate);
 void end_buffer_async_write(struct buffer_head *bh, int uptodate);
-void end_buffer_async_read(struct buffer_head *bh);
+void end_buffer_async_read(struct buffer_head *bh, int uptodate);
 
 /* Things to do with buffers at mapping->private_list */
 void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode);
include/linux/fscrypt.h +1 -0
index 13012c589aa32e..6d2d16d79752bb 100644
@@ -402,6 +402,7 @@ static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
 	return !memcmp(de_name, fname->disk_name.name, fname->disk_name.len);
 }
 
+/* bio.c */
 static inline int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
 					sector_t pblk, unsigned int len)
 {
include/linux/read_callbacks.h +40 -26

84 lines not on this page · raw message

#3 lore raw

Re: [PATCH V3 1/7] FS: Introduce read callbacks

In reply to Eric Biggers
Quoted 446 lines · IST Eric Biggers

Hi Chandan,

Quoted 26 lines · Chandan Rajendra

Read callbacks implements a state machine to be executed after a buffered read I/O is completed. They help in further processing the file data read from the backing store. Currently, decryption is the only post processing step to be supported.

The execution of the state machine is to be initiated by the endio function associated with the read operation.

---
 fs/Kconfig                     |   3 +
 fs/Makefile                    |   2 +
 fs/crypto/Kconfig              |   1 +
 fs/crypto/bio.c                |  11 +++
 fs/read_callbacks.c            | 174 +++++++++++++++++++++++++++++++++
 include/linux/fscrypt.h        |   5 +
 include/linux/read_callbacks.h |  38 +++++++
 7 files changed, 234 insertions(+)
 create mode 100644 fs/read_callbacks.c
 create mode 100644 include/linux/read_callbacks.h
fs/Kconfig +0 -0
index f1046cf6ad85..d869859c88da 100644

fs/Kconfig if BLOCK · line 25

Quoted 5 lines
fs/Kconfig +2 -0
@@ -21,6 +21,9 @@ if BLOCK
 config FS_IOMAP
 	bool
 
+config FS_READ_CALLBACKS
+       bool

This should be intended with a tab, not spaces.

Quoted 9 lines
fs/Kconfig +1 -0
+
 source "fs/ext2/Kconfig"
 source "fs/ext4/Kconfig"
 source "fs/jbd2/Kconfig"
fs/Makefile +0 -0
index c9aea23aba56..a1a06f0db5c1 100644

fs/Makefile else · line 26

Quoted 6 lines
fs/Makefile +2 -0
@@ -21,6 +21,8 @@ else
 obj-y +=	no-block.o
 endif
 
+obj-$(CONFIG_FS_READ_CALLBACKS) += read_callbacks.o
+
 obj-$(CONFIG_PROC_FS) += proc_namespace.o

Nit: maybe move this to just below the line for iomap.o, to be consistent with where FS_READ_CALLBACKS is in the Kconfig file.

Quoted 30 lines
fs/Makefile +0 -0
 obj-y				+= notify/
fs/crypto/Kconfig +1 -0
index 24ed99e2eca0..7752f9964280 100644
@@ -9,6 +9,7 @@ config FS_ENCRYPTION
 	select CRYPTO_CTS
 	select CRYPTO_SHA256
 	select KEYS
+	select FS_READ_CALLBACKS if BLOCK
 	help
 	  Enable encryption of files and directories.  This
 	  feature is similar to ecryptfs, but it is more memory
fs/crypto/bio.c +1 -0
index 82da2510721f..f677ff93d464 100644
@@ -24,6 +24,7 @@
 #include <linux/module.h>
 #include <linux/bio.h>
 #include <linux/namei.h>
+#include <linux/read_callbacks.h>
 #include "fscrypt_private.h"
 
 static void __fscrypt_decrypt_bio(struct bio *bio, bool done)
@@ -68,6 +69,16 @@ void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx, struct bio *bio)
 }
 EXPORT_SYMBOL(fscrypt_enqueue_decrypt_bio);
 

fs/crypto/bio.c void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx, struct bio *bio) · line 81

Quoted 10 lines
fs/crypto/bio.c +10 -0
+void fscrypt_decrypt_work(struct work_struct *work)
+{
+	struct read_callbacks_ctx *ctx =
+		container_of(work, struct read_callbacks_ctx, work);
+
+	fscrypt_decrypt_bio(ctx->bio);
+
+	read_callbacks(ctx);
+}
+

This seems like a layering violation, since it means that read_callbacks.c calls fs/crypto/ *and* fs/crypto/ calls read_callbacks.c. I don't think fs/crypto/ should be aware of read_callbacks at all. Instead we should have a clear ordering between the components: the filesystem uses read_callbacks.c and fs/crypto/, and read_callbacks.c uses fs/crypto/. So how about:

- Move fscrypt_decrypt_work(), fscrypt_decrypt_bh(), and fscrypt_decrypt_bio() into fs/read_callbacks.c and remove the "fscrypt_" prefix from them.

- Instead of selecting FS_READ_CALLBACKS from FS_ENCRYPTION, select it from EXT4_FS and F2FS_FS (if FS_ENCRYPTION). I.e., it's really the *filesystems* themselves that use read_callbacks, not fs/crypto/.

- Move the definition of read_callbacks_ctx into fs/read_callbacks.c, and make read_callbacks() static, so these are private to the read_callbacks component.

Quoted 27 lines
fs/crypto/bio.c +0 -0
 int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
 				sector_t pblk, unsigned int len)
 {
fs/read_callbacks.c new file +18 -0
new file mode 100644
index 000000000000..a4196e3de05f
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * This file tracks the state machine that needs to be executed after reading
+ * data from files that are encrypted and/or have verity metadata associated
+ * with them.
+ */
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/bio.h>
+#include <linux/fscrypt.h>
+#include <linux/read_callbacks.h>
+
+#define NUM_PREALLOC_READ_CALLBACK_CTXS	128
+
+static struct kmem_cache *read_callbacks_ctx_cache;
+static mempool_t *read_callbacks_ctx_pool;
+

fs/read_callbacks.c line 28

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+/* Read callback state machine steps */
+enum read_callbacks_step {
+	STEP_INITIAL = 0,
+	STEP_DECRYPT,
+};
+
+static void put_read_callbacks_ctx(struct read_callbacks_ctx *ctx)
+{
+	mempool_free(ctx, read_callbacks_ctx_pool);
+}

Maybe call this free_read_callbacks_ctx(), so that it doesn't sound like it's doing reference counting.

Quoted 20 lines
fs/read_callbacks.c new file +20 -0
+
+static void end_read_callbacks_bio(struct bio *bio)
+{
+	struct read_callbacks_ctx *ctx;
+	struct page *page;
+	struct bio_vec *bv;
+	struct bvec_iter_all iter_all;
+
+	ctx = bio->bi_private;
+
+	bio_for_each_segment_all(bv, bio, iter_all) {
+		page = bv->bv_page;
+
+		if (bio->bi_status || PageError(page)) {
+			ClearPageUptodate(page);
+			SetPageError(page);
+		} else {
+			SetPageUptodate(page);
+		}
+

fs/read_callbacks.c line 58

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+		if (ctx->page_op)
+			ctx->page_op(bio, page);
+
+		unlock_page(page);
+	}
+
+	put_read_callbacks_ctx(ctx);
+
+	bio_put(bio);
+}

The filesystem itself (or fs/mpage.c) actually has to implement almost all this logic as well anyway, because CONFIG_FS_READ_CALLBACKS may be unset. And the ->page_op() callback, which exists only because f2fs needs to update some counter, is very ugly.

IMO, it would be simpler to just make this whole function filesystem-specific, as a 'typedef void (*end_bio_func_t)(struct bio *bio);' which gets passed to the read_callbacks endio hook.

Of course, each end_bio_func_t would have to check PageError() to check whether any read_callbacks step failed. But to make that a bit easier and to make it get compiled out when CONFIG_FS_READ_CALLBACKS is unset, there could be a helper function in read_callbacks.h:

	#ifdef CONFIG_FS_READ_CALLBACKS
	static inline bool read_callbacks_failed(struct page *page)
	{
		return PageError(page);
	}
	#else
	static inline bool read_callbacks_failed(struct page *page)
	{
		return false;
	}
	#endif
Quoted 21 lines
fs/read_callbacks.c new file +21 -0
+
+/**
+ * read_callbacks() - Execute the read callbacks state machine.
+ * @ctx: The context structure tracking the current state.
+ *
+ * For each state, this function enqueues a work into appropriate subsystem's
+ * work queue. After performing further processing of the data in the bio's
+ * pages, the subsystem should invoke read_calbacks() to continue with the next
+ * state in the state machine.
+ */
+void read_callbacks(struct read_callbacks_ctx *ctx)
+{
+	/*
+	 * We use different work queues for decryption and for verity because
+	 * verity may require reading metadata pages that need decryption, and
+	 * we shouldn't recurse to the same workqueue.
+	 */
+	switch (++ctx->cur_step) {
+	case STEP_DECRYPT:
+		if (ctx->enabled_steps & (1 << STEP_DECRYPT)) {
+			INIT_WORK(&ctx->work, fscrypt_decrypt_work);

fs/read_callbacks.c line 89

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+			fscrypt_enqueue_decrypt_work(&ctx->work);
+			return;
+		}
+		ctx->cur_step++;
+		/* fall-through */
+	default:
+		end_read_callbacks_bio(ctx->bio);
+	}
+}
+EXPORT_SYMBOL(read_callbacks);

As I mentioned, I think the work functions should be defined in this file rather than in fs/crypto/ or fs/verity/, since they're specific to the read_callbacks. fs/crypto/ and fs/verity/ should not be aware of read_callbacks at all. Moreover, the 'read_callbacks()' function should be static.

Quoted 16 lines
fs/read_callbacks.c new file +16 -0
+
+/**
+ * read_callbacks_end_bio() - Initiate the read callbacks state machine.
+ * @bio: bio on which read I/O operation has just been completed.
+ *
+ * Initiates the execution of the read callbacks state machine when the read
+ * operation has been completed successfully. If there was an error associated
+ * with the bio, this function frees the read callbacks context structure stored
+ * in bio->bi_private (if any).
+ *
+ * Return: 1 to indicate that the bio has been handled (including unlocking the
+ * pages); 0 otherwise.
+ */
+int read_callbacks_end_bio(struct bio *bio)
+{
+	if (!bio->bi_status && bio->bi_private) {

fs/read_callbacks.c line 115

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+		read_callbacks((struct read_callbacks_ctx *)(bio->bi_private));
+		return 1;
+	}
+
+	if (bio->bi_private)
+		put_read_callbacks_ctx((struct read_callbacks_ctx *)(bio->bi_private));
+
+	return 0;
+}
+EXPORT_SYMBOL(read_callbacks_end_bio);

Okay, several issues here...

First, the naming of this is really confusing, since it's actually *beginning* the read callbacks, not ending them; and it's basically the same name as end_read_callbacks_bio(), which actually *is* for ending the read callbacks. Since this is the endio hook, how about calling it read_callbacks_endio_bio(), and likewise read_callbacks_endio_bh()?

But more importantly, this doesn't need to have a return value, since the read_callbacks layer has to know how to end the bio (meaning unlock the pages and mark them uptodate or not) *anyway*, or at least know how to ask the filesystem to do it. So it should just do it if needed, rather than returning 0 and making the caller do it.

Also just assign 'ctx = bio->bi_private' at the start of the function; no need to access the field 4 times and have unnecessary casts.

So IMO, the below would be much better:

void read_callbacks_endio_bio(struct bio *bio, end_bio_func_t end_bio)

{
	struct read_callbacks_ctx *ctx = bio->bi_private;

	if (ctx) {
		if (!bio->bi_status) {
			ctx->end_bio = end_bio;
			read_callbacks(ctx);
			return;
		}
		free_read_callbacks_ctx(ctx);
	}
	end_bio(bio);
}

EXPORT_SYMBOL(read_callbacks_endio_bio);

And then the !CONFIG_FS_READ_CALLBACKS stub:

static inline void read_callbacks_endio_bio(struct bio *bio,
end_bio_func_t end_bio)

{
	end_bio(bio);
}

Similarly for the buffer_head versions.

Quoted 29 lines
fs/read_callbacks.c new file +29 -0
+
+/**
+ * read_callbacks_setup() - Initialize the read callbacks state machine
+ * @inode: The file on which read I/O is performed.
+ * @bio: bio holding page cache pages on which read I/O is performed.
+ * @page_op: Function to perform filesystem specific operations on a page.
+ *
+ * Based on the nature of the file's data (e.g. encrypted), this function
+ * computes the steps that need to be performed after data is read of the
+ * backing disk. This information is saved in a context structure. A pointer
+ * to the context structure is then stored in bio->bi_private for later
+ * usage.
+ *
+ * Return: 0 on success; An error code on failure.
+ */
+int read_callbacks_setup(struct inode *inode, struct bio *bio,
+			end_page_op_t page_op)
+{
+	struct read_callbacks_ctx *ctx = NULL;
+	unsigned int enabled_steps = 0;
+
+	if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode))
+		enabled_steps |= 1 << STEP_DECRYPT;
+
+	if (enabled_steps) {
+		ctx = mempool_alloc(read_callbacks_ctx_pool, GFP_NOFS);
+		if (!ctx)
+			return -ENOMEM;
+		ctx->bio = bio;

fs/read_callbacks.c line 154

Quoted 10 lines
fs/read_callbacks.c new file +10 -0
+		ctx->inode = inode;
+		ctx->enabled_steps = enabled_steps;
+		ctx->cur_step = STEP_INITIAL;
+		ctx->page_op = page_op;
+		bio->bi_private = ctx;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(read_callbacks_setup);

Please call it:

read_callbacks_setup_bio()

Then when you add buffer_head support later in the patchset, rather than adding a buffer_head argument to this function, add a new function:

read_callbacks_setup_bh()

So that all the users don't have to pass both the buffer_head and bio arguments.

These can use a common function get_read_callbacks_ctx() that creates a read_callbacks_ctx for the inode. E.g., the bio version could look like:

int read_callbacks_setup_bio(struct inode *inode, struct bio *bio)

{
	struct read_callbacks_ctx *ctx = get_read_callbacks_ctx(inode);

	if (ctx) {
		if (IS_ERR(ctx))
			return PTR_ERR(ctx);
		ctx->bio = bio;
		ctx->bh = NULL;
		bio->bi_private = ctx;
	}
	return 0;
}

EXPORT_SYMBOL(read_callbacks_setup_bio);

Also, a nit: can you move the read_callbacks_setup_*() functions to near the top of the file, since they're called first (before the endio functions)? Likewise in read_callbacks.h. It would nice to keep the functions in a logical order.

include/linux/read_callbacks.h line 3

Quoted 9 lines
include/linux/read_callbacks.h new file +3 -0
new file mode 100644
index 000000000000..aa1ec8ed7a6a
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _READ_CALLBACKS_H
+#define _READ_CALLBACKS_H

Headers should be self-contained, but this is missing some prerequisite headers and forward declarations. Try compiling a .c file with this header included first.

include/linux/read_callbacks.h line 14

Quoted 11 lines
include/linux/read_callbacks.h new file +11 -0
+
+typedef void (*end_page_op_t)(struct bio *bio, struct page *page);
+
+struct read_callbacks_ctx {
+	struct bio *bio;
+	struct inode *inode;
+	struct work_struct work;
+	unsigned int cur_step;
+	unsigned int enabled_steps;
+	end_page_op_t page_op;
+};

As I mentioned, I don't think read_callbacks_ctx should be exposed to filesystems like this. Instead just forward declare it here, and put the actual definition in fs/read_callbacks.c.

Quoted 5 lines
include/linux/read_callbacks.h new file +5 -0
+
+#ifdef CONFIG_FS_READ_CALLBACKS
+void read_callbacks(struct read_callbacks_ctx *ctx);
+int read_callbacks_end_bio(struct bio *bio);
+int read_callbacks_setup(struct inode *inode, struct bio *bio,

include/linux/read_callbacks.h line 29

Quoted 10 lines
include/linux/read_callbacks.h new file +10 -0
+			end_page_op_t page_op);
+#else
+static inline void read_callbacks(struct read_callbacks_ctx *ctx)
+{
+}
+
+static inline int read_callbacks_end_bio(struct bio *bio)
+{
+	return -EOPNOTSUPP;
+}

This stub needs to return 0, otherwise it breaks fs/mpage.c and f2fs for everyone when CONFIG_FS_READ_CALLBACKS is unset.

But as I mentioned elsewhere, I think this should actually call an end_bio() callback itself and return void, which would also avoid this issue.

Quoted 9 lines

include/linux/read_callbacks.h line 35

Quoted 6 lines
include/linux/read_callbacks.h new file +6 -0
+
+static inline int read_callbacks_setup(struct inode *inode,
+				struct bio *bio, end_page_op_t page_op)
+{
+	return -EOPNOTSUPP;
+}

Similarly here, this stub needs to return 0.

I agree with all your comments. I will fix them up and post the next version.

Signature · 1 line
chandan
#6 lore raw

Re: [PATCH V3 2/7] Integrate read callbacks into Ext4 and F2FS

In reply to Eric Biggers
Quoted 28 lines · IST Eric Biggers

Hi Chandan,

Quoted 13 lines · Chandan Rajendra

This commit gets Ext4 and F2FS to make use of read callbacks API to perform decryption of file data read from the disk.

---
 fs/crypto/bio.c             |  30 +--------
 fs/crypto/crypto.c          |   1 +
 fs/crypto/fscrypt_private.h |   3 +
 fs/ext4/readpage.c          |  29 +++------
 fs/f2fs/data.c              | 124 +++++++-----------------------------
 fs/f2fs/super.c             |   9 +--
 fs/read_callbacks.c         |   1 -
 include/linux/fscrypt.h     |  18 ------
 8 files changed, 40 insertions(+), 175 deletions(-)

This patch changes many different components. It would be much easier to review, and might get more attention from the other ext4 and f2fs developers, if it were split into 3 patches:

a. Convert ext4 to use read_callbacks.
b. Convert f2fs to use read_callbacks.
c. Remove the functions from fs/crypto/ that became unused as a result of
patches (a) and (b). (Actually, this part probably should be merged with the patch that removes the fscrypt_ctx, and the patch renamed to something like "fscrypt: remove decryption I/O path helpers")

Quoted 5 lines

Any reason why this wouldn't work? AFAICS, you couldn't do it only because you made this patch change fscrypt_enqueue_decrypt_work() to be responsible for initializing the work function. But as per my comments on patch 1, I don't think we should do that, since it would make much more sense to put the work function in read_callbacks.c.

Yes, you are right about that. I will make the changes suggested by you.

Quoted 76 lines

However, since you're converting ext4 to use mpage_readpages() anyway, I don't think we should bother with the intermediate change to ext4_mpage_readpages(). It's useless, and that intermediate state of the ext4 code inevitably won't get tested very well. So perhaps order the whole series as:

- fs: introduce read_callbacks
- fs/mpage.c: add decryption support via read_callbacks
- fs/buffer.c: add decryption support via read_callbacks
- f2fs: convert to use read_callbacks
- ext4: convert to use mpage_readpages[s]
- ext4: support encryption with subpage-sized blocks
- fscrypt: remove decryption I/O path helpers

That order would also give the flexibility to possibly apply the fs/ changes first, without having to update both ext4 and f2fs simultaneously with them.

fs/f2fs/data.c static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, · line 514

Quoted 7 lines
fs/f2fs/data.c +1 -2
@@ -557,8 +511,7 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct bio *bio;
-	struct bio_post_read_ctx *ctx;
-	unsigned int post_read_steps = 0;
+	int ret;

Nit: 'err' rather than 'ret', since this is 0 or a -errno value.

Quoted 13 lines
fs/f2fs/data.c +0 -13
-int __init f2fs_init_post_read_processing(void)
-{
-	bio_post_read_ctx_cache = KMEM_CACHE(bio_post_read_ctx, 0);
-	if (!bio_post_read_ctx_cache)
-		goto fail;
-	bio_post_read_ctx_pool =
-		mempool_create_slab_pool(NUM_PREALLOC_POST_READ_CTXS,
-					 bio_post_read_ctx_cache);
-	if (!bio_post_read_ctx_pool)
-		goto fail_free_cache;
-	return 0;
-
-fail_free_cache:

fs/f2fs/data.c void f2fs_clear_page_cache_dirty_tag(struct page *page) · line 2928

Quoted 10 lines
fs/f2fs/data.c +0 -10
-	kmem_cache_destroy(bio_post_read_ctx_cache);
-fail:
-	return -ENOMEM;
-}
-
-void __exit f2fs_destroy_post_read_processing(void)
-{
-	mempool_destroy(bio_post_read_ctx_pool);
-	kmem_cache_destroy(bio_post_read_ctx_cache);
-}

Need to remove the declarations of these functions from fs/f2fs/f2fs.h to.

fs/read_callbacks.c void read_callbacks(struct read_callbacks_ctx *ctx) · line 81

Quoted 12 lines
fs/read_callbacks.c +0 -1
index a4196e3de05f..4b7fc2a349cd 100644
@@ -76,7 +76,6 @@ void read_callbacks(struct read_callbacks_ctx *ctx)
 	switch (++ctx->cur_step) {
 	case STEP_DECRYPT:
 		if (ctx->enabled_steps & (1 << STEP_DECRYPT)) {
-			INIT_WORK(&ctx->work, fscrypt_decrypt_work);
 			fscrypt_enqueue_decrypt_work(&ctx->work);
 			return;
 		}

Again, I think the work initialization should remain here as:

	INIT_WORK(&ctx->work, decrypt_work);

rather than moving it to fs/crypto/.

Thanks!

- Eric

Signature · 1 line
chandan
#11 lore raw

Re: [PATCH V3 4/7] fs/mpage.c: Integrate read callbacks

In reply to Eric Biggers
Quoted 45 lines · IST Eric Biggers
Quoted 44 lines · Chandan Rajendra

This commit adds code to make do_mpage_readpage() to be "read callbacks" aware i.e. for files requiring decryption, do_mpage_readpage() now sets up the read callbacks state machine when allocating a bio and later starts execution of the state machine after file data is read from the underlying disk.

---
 fs/mpage.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
fs/mpage.c +9 -0
index 436a85260394..611ad122fc92 100644
@@ -30,6 +30,7 @@
 #include <linux/backing-dev.h>
 #include <linux/pagevec.h>
 #include <linux/cleancache.h>
+#include <linux/read_callbacks.h>
 #include "internal.h"
 
 /*
@@ -49,6 +50,8 @@ static void mpage_end_io(struct bio *bio)
 	struct bio_vec *bv;
 	struct bvec_iter_all iter_all;
 
+	if (read_callbacks_end_bio(bio))
+		return;
 	bio_for_each_segment_all(bv, bio, iter_all) {
 		struct page *page = bv->bv_page;
 		page_endio(page, bio_op(bio),
@@ -309,6 +312,12 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 					gfp);
 		if (args->bio == NULL)
 			goto confused;
+
+		if (read_callbacks_setup(inode, args->bio, NULL)) {
+			bio_put(args->bio);
+			args->bio = NULL;
+			goto confused;
+		}
 	}
 
Quoted 13 lines

fs/mpage.c static struct bio *do_mpage_readpage(struct mpage_readpage_args *args) · line 345

Quoted 10 lines
fs/mpage.c +1 -1
 	length = first_hole << blkbits;
@@ -330,7 +339,7 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
 confused:
 	if (args->bio)
 		args->bio = mpage_bio_submit(REQ_OP_READ, op_flags, args->bio);
-	if (!PageUptodate(page))
+	if (!PageUptodate(page) && !PageError(page))
 		block_read_full_page(page, args->get_block);
 	else
 		unlock_page(page);

Why is the !PageError() check needed here?

Sorry, Its a remnant from when I was debugging the patchset. I will remove this.

Signature · 1 line
chandan
#15 lore raw

Re: [PATCH V3 6/7] Add decryption support for sub-pagesized blocks

In reply to Eric Biggers
Quoted 31 lines · IST Eric Biggers
Quoted 7 lines · Chandan Rajendra

To support decryption of sub-pagesized blocks this commit adds code to,
1. Track buffer head in "struct read_callbacks_ctx".
2. Pass buffer head argument to all read callbacks.
3. Add new fscrypt helper to decrypt the file data referred to by a buffer head.

Quoted 10 lines
---
 fs/buffer.c                    |  55 +++++++++------
 fs/crypto/bio.c                |  21 +++++-
 fs/f2fs/data.c                 |   2 +-
 fs/mpage.c                     |   2 +-
 fs/read_callbacks.c            | 118 +++++++++++++++++++++++++--------
 include/linux/buffer_head.h    |   1 +
 include/linux/read_callbacks.h |  13 +++-
 7 files changed, 158 insertions(+), 54 deletions(-)

This is another patch that unnecessarily changes way too many components at once. My suggestions elsewhere would resolve this, though:

- This patch changes fs/f2fs/data.c and fs/mpage.c only to pass a NULL buffer_head to read_callbacks_setup(). But as per my comments on patch 1, read_callbacks_setup() should be split into read_callbacks_setup_bio() and read_callbacks_end_bh().

- This patch changes fs/crypto/ only to add support for the buffer_head decryption work. But as per my comments on patch 1, that should be in read_callbacks.c instead.

Quoted 7 lines

And adding buffer_head support to fs/read_callbacks.c should be its own patch, *or* should simply be folded into the patch that adds fs/read_callbacks.c.

Then the only thing remaining in this patch would be updating fs/buffer.c to make it use the read_callbacks, which should be retitled to something like "fs/buffer.c: add decryption support via read_callbacks".

I agree.

Signature · 1 line
chandan
#18 lore raw
In reply to Eric Biggers
Quoted 22 lines · IST Eric Biggers
Quoted 14 lines · Chandan Rajendra

This patchset moves the "FS read I/O callbacks" code into a file of its own (i.e. fs/read_callbacks.c) and modifies the generic do_mpage_readpge() to make use of the functionality provided.

"FS read I/O callbacks" code implements the state machine that needs to be executed after reading data from files that are encrypted and/or have verity metadata associated with them.

With these changes in place, the patchset changes Ext4 to use mpage_readpage[s] instead of its own custom ext4_readpage[s]() functions. This is done to reduce duplication of code across filesystems. Also, "FS read I/O callbacks" source files will be built only if CONFIG_FS_ENCRYPTION is enabled.

Quoted 6 lines

The patchset also modifies fs/buffer.c to get file encryption/decryption to work with subpage-sized blocks.

The patches can also be obtained from
https://github.com/chandanr/linux.git at branch subpage-encryption-v3.

Quoted 9 lines

FWIW: while doing my review I put together an (untested) incremental patch that addresses my comments on the code, so I've provided it below in case you want to start with it when addressing my comments.

This is just a single diff against your subpage-encryption-v3 branch, so of course it would still need to be folded into the appropriate patches. Also see my suggestions in reply to patch 2 about how to better organize the series. I also left TODOs in kerneldoc comments that still need to be updated.

Thanks for all your help. I will post the next version of the patchset addressing all your review comments.

Signature · 1 line
chandan