1 files changed, 126 insertions, 233 deletions

diff --git a/fs/direct-io.c b/fs/direct-io.c
index dbc1a1f080ce..2267f5ae7f77 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * fs/direct-io.c
  *
@@ -36,11 +37,12 @@
 #include <linux/rwsem.h>
 #include <linux/uio.h>
 #include <linux/atomic.h>
-#include <linux/prefetch.h>
+
+#include "internal.h"
 
 /*
- * How many user pages to map in one call to get_user_pages().  This determines
- * the size of a structure in the slab cache
+ * How many user pages to map in one call to iov_iter_extract_pages().  This
+ * determines the size of a structure in the slab cache
  */
 #define DIO_PAGES	64
 
@@ -83,7 +85,6 @@ struct dio_submit {
 	sector_t final_block_in_request;/* doesn't change */
 	int boundary;			/* prev block is at a boundary */
 	get_block_t *get_block;		/* block mapping function */
-	dio_submit_t *submit_io;	/* IO submition function */
 
 	loff_t logical_offset_in_bio;	/* current first logical block in bio */
 	sector_t final_block_in_bio;	/* current final block in bio + 1 */
@@ -114,19 +115,18 @@ struct dio_submit {
 /* dio_state communicated between submission path and end_io */
 struct dio {
 	int flags;			/* doesn't change */
-	int op;
-	int op_flags;
-	blk_qc_t bio_cookie;
+	blk_opf_t opf;			/* request operation type and flags */
 	struct gendisk *bio_disk;
 	struct inode *inode;
 	loff_t i_size;			/* i_size when submitted */
 	dio_iodone_t *end_io;		/* IO completion function */
+	bool is_pinned;			/* T if we have pins on the pages */
 
 	void *private;			/* copy from map_bh.b_private */
 
 	/* BIO completion state */
 	spinlock_t bio_lock;		/* protects BIO fields below */
-	int page_errors;		/* errno from get_user_pages() */
+	int page_errors;		/* err from iov_iter_extract_pages() */
 	int is_async;			/* is IO async ? */
 	bool defer_completion;		/* defer AIO completion to workqueue? */
 	bool should_dirty;		/* if pages should be dirtied */
@@ -150,7 +150,7 @@ struct dio {
 	};
 } ____cacheline_aligned_in_smp;
 
-static struct kmem_cache *dio_cache __read_mostly;
+static struct kmem_cache *dio_cache __ro_after_init;
 
 /*
  * How many pages are in the queue?
@@ -165,13 +165,14 @@ static inline unsigned dio_pages_present(struct dio_submit *sdio)
  */
 static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio)
 {
+	struct page **pages = dio->pages;
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
 	ssize_t ret;
 
-	ret = iov_iter_get_pages(sdio->iter, dio->pages, LONG_MAX, DIO_PAGES,
-				&sdio->from);
+	ret = iov_iter_extract_pages(sdio->iter, &pages, LONG_MAX,
+				     DIO_PAGES, 0, &sdio->from);
 
-	if (ret < 0 && sdio->blocks_available && (dio->op == REQ_OP_WRITE)) {
-		struct page *page = ZERO_PAGE(0);
+	if (ret < 0 && sdio->blocks_available && dio_op == REQ_OP_WRITE) {
 		/*
 		 * A memory fault, but the filesystem has some outstanding
 		 * mapped blocks.  We need to use those blocks up to avoid
@@ -179,8 +180,7 @@ static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio)
 		 */
 		if (dio->page_errors == 0)
 			dio->page_errors = ret;
-		get_page(page);
-		dio->pages[0] = page;
+		dio->pages[0] = ZERO_PAGE(0);
 		sdio->head = 0;
 		sdio->tail = 1;
 		sdio->from = 0;
@@ -189,7 +189,6 @@ static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio)
 	}
 
 	if (ret >= 0) {
-		iov_iter_advance(sdio->iter, ret);
 		ret += sdio->from;
 		sdio->head = 0;
 		sdio->tail = (ret + PAGE_SIZE - 1) / PAGE_SIZE;
@@ -201,9 +200,9 @@ static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio)
 
 /*
  * Get another userspace page.  Returns an ERR_PTR on error.  Pages are
- * buffered inside the dio so that we can call get_user_pages() against a
- * decent number of pages, less frequently.  To provide nicer use of the
- * L1 cache.
+ * buffered inside the dio so that we can call iov_iter_extract_pages()
+ * against a decent number of pages, less frequently.  To provide nicer use of
+ * the L1 cache.
  */
 static inline struct page *dio_get_page(struct dio *dio,
 					struct dio_submit *sdio)
@@ -219,30 +218,20 @@ static inline struct page *dio_get_page(struct dio *dio,
 	return dio->pages[sdio->head];
 }
 
-/*
- * Warn about a page cache invalidation failure during a direct io write.
- */
-void dio_warn_stale_pagecache(struct file *filp)
+static void dio_pin_page(struct dio *dio, struct page *page)
 {
-	static DEFINE_RATELIMIT_STATE(_rs, 86400 * HZ, DEFAULT_RATELIMIT_BURST);
-	char pathname[128];
-	struct inode *inode = file_inode(filp);
-	char *path;
-
-	errseq_set(&inode->i_mapping->wb_err, -EIO);
-	if (__ratelimit(&_rs)) {
-		path = file_path(filp, pathname, sizeof(pathname));
-		if (IS_ERR(path))
-			path = "(unknown)";
-		pr_crit("Page cache invalidation failure on direct I/O.  Possible data corruption due to collision with buffered I/O!\n");
-		pr_crit("File: %s PID: %d Comm: %.20s\n", path, current->pid,
-			current->comm);
-	}
+	if (dio->is_pinned)
+		folio_add_pin(page_folio(page));
+}
+
+static void dio_unpin_page(struct dio *dio, struct page *page)
+{
+	if (dio->is_pinned)
+		unpin_user_page(page);
 }
 
-/**
+/*
  * dio_complete() - called when all DIO BIO I/O has been completed
- * @offset: the byte offset in the file of the completed operation
  *
  * This drops i_dio_count, lets interested parties know that a DIO operation
  * has completed, and calculates the resulting return code for the operation.
@@ -254,6 +243,7 @@ void dio_warn_stale_pagecache(struct file *filp)
  */
 static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
 {
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
 	loff_t offset = dio->iocb->ki_pos;
 	ssize_t transferred = 0;
 	int err;
@@ -271,7 +261,7 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
 		transferred = dio->result;
 
 		/* Check for short read case */
-		if ((dio->op == REQ_OP_READ) &&
+		if (dio_op == REQ_OP_READ &&
 		    ((offset + transferred) > dio->i_size))
 			transferred = dio->i_size - offset;
 		/* ignore EFAULT if some IO has been done */
@@ -306,14 +296,8 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
 	 * zeros from unwritten extents.
 	 */
 	if (flags & DIO_COMPLETE_INVALIDATE &&
-	    ret > 0 && dio->op == REQ_OP_WRITE &&
-	    dio->inode->i_mapping->nrpages) {
-		err = invalidate_inode_pages2_range(dio->inode->i_mapping,
-					offset >> PAGE_SHIFT,
-					(offset + ret - 1) >> PAGE_SHIFT);
-		if (err)
-			dio_warn_stale_pagecache(dio->iocb->ki_filp);
-	}
+	    ret > 0 && dio_op == REQ_OP_WRITE)
+		kiocb_invalidate_post_direct_write(dio->iocb, ret);
 
 	inode_dio_end(dio->inode);
 
@@ -325,9 +309,9 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
 		 */
 		dio->iocb->ki_pos += transferred;
 
-		if (ret > 0 && dio->op == REQ_OP_WRITE)
+		if (ret > 0 && dio_op == REQ_OP_WRITE)
 			ret = generic_write_sync(dio->iocb, ret);
-		dio->iocb->ki_complete(dio->iocb, ret, 0);
+		dio->iocb->ki_complete(dio->iocb, ret);
 	}
 
 	kmem_cache_free(dio_cache, dio);
@@ -349,6 +333,7 @@ static blk_status_t dio_bio_complete(struct dio *dio, struct bio *bio);
 static void dio_bio_end_aio(struct bio *bio)
 {
 	struct dio *dio = bio->bi_private;
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
 	unsigned long remaining;
 	unsigned long flags;
 	bool defer_completion = false;
@@ -373,7 +358,7 @@ static void dio_bio_end_aio(struct bio *bio)
 		 */
 		if (dio->result)
 			defer_completion = dio->defer_completion ||
-					   (dio->op == REQ_OP_WRITE &&
+					   (dio_op == REQ_OP_WRITE &&
 					    dio->inode->i_mapping->nrpages);
 		if (defer_completion) {
 			INIT_WORK(&dio->complete_work, dio_aio_complete_work);
@@ -405,25 +390,6 @@ static void dio_bio_end_io(struct bio *bio)
 	spin_unlock_irqrestore(&dio->bio_lock, flags);
 }
 
-/**
- * dio_end_io - handle the end io action for the given bio
- * @bio: The direct io bio thats being completed
- *
- * This is meant to be called by any filesystem that uses their own dio_submit_t
- * so that the DIO specific endio actions are dealt with after the filesystem
- * has done it's completion work.
- */
-void dio_end_io(struct bio *bio)
-{
-	struct dio *dio = bio->bi_private;
-
-	if (dio->is_async)
-		dio_bio_end_aio(bio);
-	else
-		dio_bio_end_io(bio);
-}
-EXPORT_SYMBOL_GPL(dio_end_io);
-
 static inline void
 dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
 	      struct block_device *bdev,
@@ -435,17 +401,15 @@ dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
 	 * bio_alloc() is guaranteed to return a bio when allowed to sleep and
 	 * we request a valid number of vectors.
 	 */
-	bio = bio_alloc(GFP_KERNEL, nr_vecs);
-
-	bio_set_dev(bio, bdev);
+	bio = bio_alloc(bdev, nr_vecs, dio->opf, GFP_KERNEL);
 	bio->bi_iter.bi_sector = first_sector;
-	bio_set_op_attrs(bio, dio->op, dio->op_flags);
 	if (dio->is_async)
 		bio->bi_end_io = dio_bio_end_aio;
 	else
 		bio->bi_end_io = dio_bio_end_io;
-
-	bio->bi_write_hint = dio->iocb->ki_hint;
+	if (dio->is_pinned)
+		bio_set_flag(bio, BIO_PAGE_PINNED);
+	bio->bi_write_hint = file_inode(dio->iocb->ki_filp)->i_write_hint;
 
 	sdio->bio = bio;
 	sdio->logical_offset_in_bio = sdio->cur_page_fs_offset;
@@ -460,6 +424,7 @@ dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
  */
 static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio)
 {
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
 	struct bio *bio = sdio->bio;
 	unsigned long flags;
 
@@ -469,16 +434,12 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio)
 	dio->refcount++;
 	spin_unlock_irqrestore(&dio->bio_lock, flags);
 
-	if (dio->is_async && dio->op == REQ_OP_READ && dio->should_dirty)
+	if (dio->is_async && dio_op == REQ_OP_READ && dio->should_dirty)
 		bio_set_pages_dirty(bio);
 
-	dio->bio_disk = bio->bi_disk;
+	dio->bio_disk = bio->bi_bdev->bd_disk;
 
-	if (sdio->submit_io) {
-		sdio->submit_io(bio, dio->inode, sdio->logical_offset_in_bio);
-		dio->bio_cookie = BLK_QC_T_NONE;
-	} else
-		dio->bio_cookie = submit_bio(bio);
+	submit_bio(bio);
 
 	sdio->bio = NULL;
 	sdio->boundary = 0;
@@ -490,15 +451,17 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio)
  */
 static inline void dio_cleanup(struct dio *dio, struct dio_submit *sdio)
 {
-	while (sdio->head < sdio->tail)
-		put_page(dio->pages[sdio->head++]);
+	if (dio->is_pinned)
+		unpin_user_pages(dio->pages + sdio->head,
+				 sdio->tail - sdio->head);
+	sdio->head = sdio->tail;
 }
 
 /*
  * Wait for the next BIO to complete.  Remove it and return it.  NULL is
  * returned once all BIOs have been completed.  This must only be called once
  * all bios have been issued so that dio->refcount can only decrease.  This
- * requires that that the caller hold a reference on the dio.
+ * requires that the caller hold a reference on the dio.
  */
 static struct bio *dio_await_one(struct dio *dio)
 {
@@ -517,9 +480,7 @@ static struct bio *dio_await_one(struct dio *dio)
 		__set_current_state(TASK_UNINTERRUPTIBLE);
 		dio->waiter = current;
 		spin_unlock_irqrestore(&dio->bio_lock, flags);
-		if (!(dio->iocb->ki_flags & IOCB_HIPRI) ||
-		    !blk_poll(dio->bio_disk->queue, dio->bio_cookie, true))
-			io_schedule();
+		blk_io_schedule();
 		/* wake up sets us TASK_RUNNING */
 		spin_lock_irqsave(&dio->bio_lock, flags);
 		dio->waiter = NULL;
@@ -537,9 +498,9 @@ static struct bio *dio_await_one(struct dio *dio)
  */
 static blk_status_t dio_bio_complete(struct dio *dio, struct bio *bio)
 {
-	struct bio_vec *bvec;
-	unsigned i;
 	blk_status_t err = bio->bi_status;
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
+	bool should_dirty = dio_op == REQ_OP_READ && dio->should_dirty;
 
 	if (err) {
 		if (err == BLK_STS_AGAIN && (bio->bi_opf & REQ_NOWAIT))
@@ -548,17 +509,10 @@ static blk_status_t dio_bio_complete(struct dio *dio, struct bio *bio)
 			dio->io_error = -EIO;
 	}
 
-	if (dio->is_async && dio->op == REQ_OP_READ && dio->should_dirty) {
+	if (dio->is_async && should_dirty) {
 		bio_check_pages_dirty(bio);	/* transfers ownership */
 	} else {
-		bio_for_each_segment_all(bvec, bio, i) {
-			struct page *page = bvec->bv_page;
-
-			if (dio->op == REQ_OP_READ && !PageCompound(page) &&
-					dio->should_dirty)
-				set_page_dirty_lock(page);
-			put_page(page);
-		}
+		bio_release_pages(bio, should_dirty);
 		bio_put(bio);
 	}
 	return err;
@@ -611,30 +565,6 @@ static inline int dio_bio_reap(struct dio *dio, struct dio_submit *sdio)
 	return ret;
 }
 
-/*
- * Create workqueue for deferred direct IO completions. We allocate the
- * workqueue when it's first needed. This avoids creating workqueue for
- * filesystems that don't need it and also allows us to create the workqueue
- * late enough so the we can include s_id in the name of the workqueue.
- */
-int sb_init_dio_done_wq(struct super_block *sb)
-{
-	struct workqueue_struct *old;
-	struct workqueue_struct *wq = alloc_workqueue("dio/%s",
-						      WQ_MEM_RECLAIM, 0,
-						      sb->s_id);
-	if (!wq)
-		return -ENOMEM;
-	/*
-	 * This has to be atomic as more DIOs can race to create the workqueue
-	 */
-	old = cmpxchg(&sb->s_dio_done_wq, NULL, wq);
-	/* Someone created workqueue before us? Free ours... */
-	if (old)
-		destroy_workqueue(wq);
-	return 0;
-}
-
 static int dio_set_defer_completion(struct dio *dio)
 {
 	struct super_block *sb = dio->inode->i_sb;
@@ -673,12 +603,14 @@ static int dio_set_defer_completion(struct dio *dio)
 static int get_more_blocks(struct dio *dio, struct dio_submit *sdio,
 			   struct buffer_head *map_bh)
 {
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
 	int ret;
 	sector_t fs_startblk;	/* Into file, in filesystem-sized blocks */
 	sector_t fs_endblk;	/* Into file, in filesystem-sized blocks */
 	unsigned long fs_count;	/* Number of filesystem-sized blocks */
 	int create;
 	unsigned int i_blkbits = sdio->blkbits + sdio->blkfactor;
+	loff_t i_size;
 
 	/*
 	 * If there was a memory error and we've overwritten all the
@@ -706,10 +638,10 @@ static int get_more_blocks(struct dio *dio, struct dio_submit *sdio,
 		 * which may decide to handle it or also return an unmapped
 		 * buffer head.
 		 */
-		create = dio->op == REQ_OP_WRITE;
+		create = dio_op == REQ_OP_WRITE;
 		if (dio->flags & DIO_SKIP_HOLES) {
-			if (fs_startblk <= ((i_size_read(dio->inode) - 1) >>
-							i_blkbits))
+			i_size = i_size_read(dio->inode);
+			if (i_size && fs_startblk <= (i_size - 1) >> i_blkbits)
 				create = 0;
 		}
 
@@ -738,7 +670,7 @@ static inline int dio_new_bio(struct dio *dio, struct dio_submit *sdio,
 	if (ret)
 		goto out;
 	sector = start_sector << (sdio->blkbits - 9);
-	nr_pages = min(sdio->pages_in_io, BIO_MAX_PAGES);
+	nr_pages = bio_max_segs(sdio->pages_in_io);
 	BUG_ON(nr_pages <= 0);
 	dio_bio_alloc(dio, sdio, map_bh->b_bdev, sector, nr_pages);
 	sdio->boundary = 0;
@@ -753,7 +685,7 @@ out:
  *
  * Return zero on success.  Non-zero means the caller needs to start a new BIO.
  */
-static inline int dio_bio_add_page(struct dio_submit *sdio)
+static inline int dio_bio_add_page(struct dio *dio, struct dio_submit *sdio)
 {
 	int ret;
 
@@ -765,7 +697,7 @@ static inline int dio_bio_add_page(struct dio_submit *sdio)
 		 */
 		if ((sdio->cur_page_len + sdio->cur_page_offset) == PAGE_SIZE)
 			sdio->pages_in_io--;
-		get_page(sdio->cur_page);
+		dio_pin_page(dio, sdio->cur_page);
 		sdio->final_block_in_bio = sdio->cur_page_block +
 			(sdio->cur_page_len >> sdio->blkbits);
 		ret = 0;
@@ -820,11 +752,11 @@ static inline int dio_send_cur_page(struct dio *dio, struct dio_submit *sdio,
 			goto out;
 	}
 
-	if (dio_bio_add_page(sdio) != 0) {
+	if (dio_bio_add_page(dio, sdio) != 0) {
 		dio_bio_submit(dio, sdio);
 		ret = dio_new_bio(dio, sdio, sdio->cur_page_block, map_bh);
 		if (ret == 0) {
-			ret = dio_bio_add_page(sdio);
+			ret = dio_bio_add_page(dio, sdio);
 			BUG_ON(ret != 0);
 		}
 	}
@@ -854,9 +786,11 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page,
 		    unsigned offset, unsigned len, sector_t blocknr,
 		    struct buffer_head *map_bh)
 {
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
 	int ret = 0;
+	int boundary = sdio->boundary;	/* dio_send_cur_page may clear it */
 
-	if (dio->op == REQ_OP_WRITE) {
+	if (dio_op == REQ_OP_WRITE) {
 		/*
 		 * Read accounting is performed in submit_bio()
 		 */
@@ -879,13 +813,13 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page,
 	 */
 	if (sdio->cur_page) {
 		ret = dio_send_cur_page(dio, sdio, map_bh);
-		put_page(sdio->cur_page);
+		dio_unpin_page(dio, sdio->cur_page);
 		sdio->cur_page = NULL;
 		if (ret)
 			return ret;
 	}
 
-	get_page(page);		/* It is in dio */
+	dio_pin_page(dio, page);		/* It is in dio */
 	sdio->cur_page = page;
 	sdio->cur_page_offset = offset;
 	sdio->cur_page_len = len;
@@ -893,14 +827,14 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page,
 	sdio->cur_page_fs_offset = sdio->block_in_file << sdio->blkbits;
 out:
 	/*
-	 * If sdio->boundary then we want to schedule the IO now to
+	 * If boundary then we want to schedule the IO now to
 	 * avoid metadata seeks.
 	 */
-	if (sdio->boundary) {
+	if (boundary) {
 		ret = dio_send_cur_page(dio, sdio, map_bh);
 		if (sdio->bio)
 			dio_bio_submit(dio, sdio);
-		put_page(sdio->cur_page);
+		dio_unpin_page(dio, sdio->cur_page);
 		sdio->cur_page = NULL;
 	}
 	return ret;
@@ -969,6 +903,7 @@ static inline void dio_zero_block(struct dio *dio, struct dio_submit *sdio,
 static int do_direct_IO(struct dio *dio, struct dio_submit *sdio,
 			struct buffer_head *map_bh)
 {
+	const enum req_op dio_op = dio->opf & REQ_OP_MASK;
 	const unsigned blkbits = sdio->blkbits;
 	const unsigned i_blkbits = blkbits + sdio->blkfactor;
 	int ret = 0;
@@ -1000,7 +935,7 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio,
 
 				ret = get_more_blocks(dio, sdio, map_bh);
 				if (ret) {
-					put_page(page);
+					dio_unpin_page(dio, page);
 					goto out;
 				}
 				if (!buffer_mapped(map_bh))
@@ -1044,8 +979,8 @@ do_holes:
 				loff_t i_size_aligned;
 
 				/* AKPM: eargh, -ENOTBLK is a hack */
-				if (dio->op == REQ_OP_WRITE) {
-					put_page(page);
+				if (dio_op == REQ_OP_WRITE) {
+					dio_unpin_page(dio, page);
 					return -ENOTBLK;
 				}
 
@@ -1058,10 +993,10 @@ do_holes:
 				if (sdio->block_in_file >=
 						i_size_aligned >> blkbits) {
 					/* We hit eof */
-					put_page(page);
+					dio_unpin_page(dio, page);
 					goto out;
 				}
-				zero_user(page, from, 1 << blkbits);
+				memzero_page(page, from, 1 << blkbits);
 				sdio->block_in_file++;
 				from += 1 << blkbits;
 				dio->result += 1 << blkbits;
@@ -1098,7 +1033,7 @@ do_holes:
 						  sdio->next_block_for_io,
 						  map_bh);
 			if (ret) {
-				put_page(page);
+				dio_unpin_page(dio, page);
 				goto out;
 			}
 			sdio->next_block_for_io += this_chunk_blocks;
@@ -1113,8 +1048,8 @@ next_block:
 				break;
 		}
 
-		/* Drop the ref which was taken in get_user_pages() */
-		put_page(page);
+		/* Drop the pin which was taken in get_user_pages() */
+		dio_unpin_page(dio, page);
 	}
 out:
 	return ret;
@@ -1148,8 +1083,8 @@ static inline int drop_refcount(struct dio *dio)
  * The locking rules are governed by the flags parameter:
  *  - if the flags value contains DIO_LOCKING we use a fancy locking
  *    scheme for dumb filesystems.
- *    For writes this function is called under i_mutex and returns with
- *    i_mutex held, for reads, i_mutex is not held on entry, but it is
+ *    For writes this function is called under i_rwsem and returns with
+ *    i_rwsem held, for reads, i_rwsem is not held on entry, but it is
  *    taken and dropped again before returning.
  *  - if the flags value does NOT contain DIO_LOCKING we don't use any
  *    internal locking but rather rely on the filesystem to synchronize
@@ -1159,7 +1094,7 @@ static inline int drop_refcount(struct dio *dio)
  * counter before starting direct I/O, and decrement it once we are done.
  * Truncate can wait for it to reach zero to provide exclusion.  It is
  * expected that filesystem provide exclusion between new direct I/O
- * and truncates.  For DIO_LOCKING filesystems this is done by i_mutex,
+ * and truncates.  For DIO_LOCKING filesystems this is done by i_rwsem,
  * but other filesystems need to take care of this on their own.
  *
  * NOTE: if you pass "sdio" to anything by pointer make sure that function
@@ -1167,11 +1102,10 @@ static inline int drop_refcount(struct dio *dio)
  * individual fields and will generate much worse code. This is important
  * for the whole file.
  */
-static inline ssize_t
-do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
-		      struct block_device *bdev, struct iov_iter *iter,
-		      get_block_t get_block, dio_iodone_t end_io,
-		      dio_submit_t submit_io, int flags)
+ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
+		struct block_device *bdev, struct iov_iter *iter,
+		get_block_t get_block, dio_iodone_t end_io,
+		int flags)
 {
 	unsigned i_blkbits = READ_ONCE(inode->i_blkbits);
 	unsigned blkbits = i_blkbits;
@@ -1181,32 +1115,18 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	loff_t offset = iocb->ki_pos;
 	const loff_t end = offset + count;
 	struct dio *dio;
-	struct dio_submit sdio = { 0, };
+	struct dio_submit sdio = { NULL, };
 	struct buffer_head map_bh = { 0, };
 	struct blk_plug plug;
 	unsigned long align = offset | iov_iter_alignment(iter);
 
-	/*
-	 * Avoid references to bdev if not absolutely needed to give
-	 * the early prefetch in the caller enough time.
-	 */
-
-	if (align & blocksize_mask) {
-		if (bdev)
-			blkbits = blksize_bits(bdev_logical_block_size(bdev));
-		blocksize_mask = (1 << blkbits) - 1;
-		if (align & blocksize_mask)
-			goto out;
-	}
-
 	/* watch out for a 0 len io from a tricksy fs */
 	if (iov_iter_rw(iter) == READ && !count)
 		return 0;
 
 	dio = kmem_cache_alloc(dio_cache, GFP_KERNEL);
-	retval = -ENOMEM;
 	if (!dio)
-		goto out;
+		return -ENOMEM;
 	/*
 	 * Believe it or not, zeroing out the page array caused a .5%
 	 * performance regression in a database benchmark.  So, we take
@@ -1215,32 +1135,33 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	memset(dio, 0, offsetof(struct dio, pages));
 
 	dio->flags = flags;
-	if (dio->flags & DIO_LOCKING) {
-		if (iov_iter_rw(iter) == READ) {
-			struct address_space *mapping =
-					iocb->ki_filp->f_mapping;
-
-			/* will be released by direct_io_worker */
-			inode_lock(inode);
-
-			retval = filemap_write_and_wait_range(mapping, offset,
-							      end - 1);
-			if (retval) {
-				inode_unlock(inode);
-				kmem_cache_free(dio_cache, dio);
-				goto out;
-			}
-		}
+	if (dio->flags & DIO_LOCKING && iov_iter_rw(iter) == READ) {
+		/* will be released by direct_io_worker */
+		inode_lock(inode);
 	}
+	dio->is_pinned = iov_iter_extract_will_pin(iter);
 
 	/* Once we sampled i_size check for reads beyond EOF */
 	dio->i_size = i_size_read(inode);
 	if (iov_iter_rw(iter) == READ && offset >= dio->i_size) {
-		if (dio->flags & DIO_LOCKING)
-			inode_unlock(inode);
-		kmem_cache_free(dio_cache, dio);
 		retval = 0;
-		goto out;
+		goto fail_dio;
+	}
+
+	if (align & blocksize_mask) {
+		if (bdev)
+			blkbits = blksize_bits(bdev_logical_block_size(bdev));
+		blocksize_mask = (1 << blkbits) - 1;
+		if (align & blocksize_mask)
+			goto fail_dio;
+	}
+
+	if (dio->flags & DIO_LOCKING && iov_iter_rw(iter) == READ) {
+		struct address_space *mapping = iocb->ki_filp->f_mapping;
+
+		retval = filemap_write_and_wait_range(mapping, offset, end - 1);
+		if (retval)
+			goto fail_dio;
 	}
 
 	/*
@@ -1258,15 +1179,12 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 
 	dio->inode = inode;
 	if (iov_iter_rw(iter) == WRITE) {
-		dio->op = REQ_OP_WRITE;
-		dio->op_flags = REQ_SYNC | REQ_IDLE;
+		dio->opf = REQ_OP_WRITE | REQ_SYNC | REQ_IDLE;
 		if (iocb->ki_flags & IOCB_NOWAIT)
-			dio->op_flags |= REQ_NOWAIT;
+			dio->opf |= REQ_NOWAIT;
 	} else {
-		dio->op = REQ_OP_READ;
+		dio->opf = REQ_OP_READ;
 	}
-	if (iocb->ki_flags & IOCB_HIPRI)
-		dio->op_flags |= REQ_HIPRI;
 
 	/*
 	 * For AIO O_(D)SYNC writes we need to defer completions to a workqueue
@@ -1274,7 +1192,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	 */
 	if (dio->is_async && iov_iter_rw(iter) == WRITE) {
 		retval = 0;
-		if (iocb->ki_flags & IOCB_DSYNC)
+		if (iocb_is_dsync(iocb))
 			retval = dio_set_defer_completion(dio);
 		else if (!dio->inode->i_sb->s_dio_done_wq) {
 			/*
@@ -1284,14 +1202,8 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 			 */
 			retval = sb_init_dio_done_wq(dio->inode->i_sb);
 		}
-		if (retval) {
-			/*
-			 * We grab i_mutex only for reads so we don't have
-			 * to release it here
-			 */
-			kmem_cache_free(dio_cache, dio);
-			goto out;
-		}
+		if (retval)
+			goto fail_dio;
 	}
 
 	/*
@@ -1299,14 +1211,12 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	 */
 	inode_dio_begin(inode);
 
-	retval = 0;
 	sdio.blkbits = blkbits;
 	sdio.blkfactor = i_blkbits - blkbits;
 	sdio.block_in_file = offset >> blkbits;
 
 	sdio.get_block = get_block;
 	dio->end_io = end_io;
-	sdio.submit_io = submit_io;
 	sdio.final_block_in_bio = -1;
 	sdio.next_block_for_io = -1;
 
@@ -1315,7 +1225,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	spin_lock_init(&dio->bio_lock);
 	dio->refcount = 1;
 
-	dio->should_dirty = iter_is_iovec(iter) && iov_iter_rw(iter) == READ;
+	dio->should_dirty = user_backed_iter(iter) && iov_iter_rw(iter) == READ;
 	sdio.iter = iter;
 	sdio.final_block_in_request = end >> blkbits;
 
@@ -1337,7 +1247,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	if (retval == -ENOTBLK) {
 		/*
 		 * The remaining part of the request will be
-		 * be handled by buffered I/O when we return
+		 * handled by buffered I/O when we return
 		 */
 		retval = 0;
 	}
@@ -1353,7 +1263,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 		ret2 = dio_send_cur_page(dio, &sdio, &map_bh);
 		if (retval == 0)
 			retval = ret2;
-		put_page(sdio.cur_page);
+		dio_unpin_page(dio, sdio.cur_page);
 		sdio.cur_page = NULL;
 	}
 	if (sdio.bio)
@@ -1369,7 +1279,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 
 	/*
 	 * All block lookups have been performed. For READ requests
-	 * we can let i_mutex go now that its achieved its purpose
+	 * we can let i_rwsem go now that its achieved its purpose
 	 * of protecting us from looking up uninitialized blocks.
 	 */
 	if (iov_iter_rw(iter) == READ && (dio->flags & DIO_LOCKING))
@@ -1394,32 +1304,15 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	} else
 		BUG_ON(retval != -EIOCBQUEUED);
 
-out:
 	return retval;
-}
 
-ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
-			     struct block_device *bdev, struct iov_iter *iter,
-			     get_block_t get_block,
-			     dio_iodone_t end_io, dio_submit_t submit_io,
-			     int flags)
-{
-	/*
-	 * The block device state is needed in the end to finally
-	 * submit everything.  Since it's likely to be cache cold
-	 * prefetch it here as first thing to hide some of the
-	 * latency.
-	 *
-	 * Attempt to prefetch the pieces we likely need later.
-	 */
-	prefetch(&bdev->bd_disk->part_tbl);
-	prefetch(bdev->bd_queue);
-	prefetch((char *)bdev->bd_queue + SMP_CACHE_BYTES);
+fail_dio:
+	if (dio->flags & DIO_LOCKING && iov_iter_rw(iter) == READ)
+		inode_unlock(inode);
 
-	return do_blockdev_direct_IO(iocb, inode, bdev, iter, get_block,
-				     end_io, submit_io, flags);
+	kmem_cache_free(dio_cache, dio);
+	return retval;
 }
-
 EXPORT_SYMBOL(__blockdev_direct_IO);
 
 static __init int dio_init(void)