xref: /linux/fs/buffer.c (revision 6b3f7af57881f6d6250c6dcc4d910fe8e855a607)

// SPDX-License-Identifier: GPL-2.0-only
/*
 *  linux/fs/buffer.c
 *
 *  Copyright (C) 1991, 1992, 2002  Linus Torvalds
 */

/*
 * Start bdflush() with kernel_thread not syscall - Paul Gortmaker, 12/95
 *
 * Removed a lot of unnecessary code and simplified things now that
 * the buffer cache isn't our primary cache - Andrew Tridgell 12/96
 *
 * Speed up hash, lru, and free list operations.  Use gfp() for allocating
 * hash table, use SLAB cache for buffer heads. SMP threading.  -DaveM
 *
 * Added 32k buffer block sizes - these are required older ARM systems. - RMK
 *
 * async buffer flushing, 1999 Andrea Arcangeli <andrea@suse.de>
 */

#include <linux/kernel.h>
#include <linux/sched/signal.h>
#include <linux/syscalls.h>
#include <linux/fs.h>
#include <linux/iomap.h>
#include <linux/mm.h>
#include <linux/percpu.h>
#include <linux/slab.h>
#include <linux/capability.h>
#include <linux/blkdev.h>
#include <linux/blk-crypto.h>
#include <linux/file.h>
#include <linux/quotaops.h>
#include <linux/highmem.h>
#include <linux/export.h>
#include <linux/backing-dev.h>
#include <linux/writeback.h>
#include <linux/hash.h>
#include <linux/suspend.h>
#include <linux/buffer_head.h>
#include <linux/task_io_accounting_ops.h>
#include <linux/bio.h>
#include <linux/cpu.h>
#include <linux/bitops.h>
#include <linux/mpage.h>
#include <linux/bit_spinlock.h>
#include <linux/folio_batch.h>
#include <linux/sched/mm.h>
#include <trace/events/block.h>
#include <linux/fscrypt.h>
#include <linux/fsverity.h>
#include <linux/sched/isolation.h>

#include "internal.h"

#define BH_ENTRY(list) list_entry((list), struct buffer_head, b_assoc_buffers)

inline void touch_buffer(struct buffer_head *bh)
{
	trace_block_touch_buffer(bh);
	folio_mark_accessed(bh->b_folio);
}
EXPORT_SYMBOL(touch_buffer);

void __lock_buffer(struct buffer_head *bh)
{
	wait_on_bit_lock_io(&bh->b_state, BH_Lock, TASK_UNINTERRUPTIBLE);
}
EXPORT_SYMBOL(__lock_buffer);

void unlock_buffer(struct buffer_head *bh)
{
	clear_and_wake_up_bit(BH_Lock, &bh->b_state);
}
EXPORT_SYMBOL(unlock_buffer);

/*
 * Returns if the folio has dirty or writeback buffers. If all the buffers
 * are unlocked and clean then the folio_test_dirty information is stale. If
 * any of the buffers are locked, it is assumed they are locked for IO.
 */
void buffer_check_dirty_writeback(struct folio *folio,
				     bool *dirty, bool *writeback)
{
	struct buffer_head *head, *bh;
	*dirty = false;
	*writeback = false;

	BUG_ON(!folio_test_locked(folio));

	head = folio_buffers(folio);
	if (!head)
		return;

	if (folio_test_writeback(folio))
		*writeback = true;

	bh = head;
	do {
		if (buffer_locked(bh))
			*writeback = true;

		if (buffer_dirty(bh))
			*dirty = true;

		bh = bh->b_this_page;
	} while (bh != head);
}

/*
 * Block until a buffer comes unlocked.  This doesn't stop it
 * from becoming locked again - you have to lock it yourself
 * if you want to preserve its state.
 */
void __wait_on_buffer(struct buffer_head * bh)
{
	wait_on_bit_io(&bh->b_state, BH_Lock, TASK_UNINTERRUPTIBLE);
}
EXPORT_SYMBOL(__wait_on_buffer);

static void buffer_io_error(struct buffer_head *bh, char *msg)
{
	if (!test_bit(BH_Quiet, &bh->b_state))
		printk_ratelimited(KERN_ERR
			"Buffer I/O error on dev %pg, logical block %llu%s\n",
			bh->b_bdev, (unsigned long long)bh->b_blocknr, msg);
}

/**
 * bio_endio_bh - Discard the bio used to submit a buffer.
 * @bio: The bio.
 * @bhp: Where to return the buffer_head.
 *
 * Call this in your bio_end_io handler to retrieve the buffer_head
 * submitted in bh_submit().  If you did not call bh_submit(), do not
 * call this function; it will return garbage.
 *
 * This function consumes the bio refcount which will probably free the
 * bio.
 *
 * Return: True if the I/O succeeded.
 */
bool bio_endio_bh(struct bio *bio, struct buffer_head **bhp)
{
	bool success = bio->bi_status == BLK_STS_OK;
	struct buffer_head *bh = bio->bi_private;

	if (unlikely(bio_flagged(bio, BIO_QUIET)))
		set_bit(BH_Quiet, &bh->b_state);
	bio_put(bio);

	*bhp = bh;
	return success;
}
EXPORT_SYMBOL(bio_endio_bh);

/**
 * end_buffer_read_sync - Handle buffer reads finishing
 * @bh: The buffer.
 * @uptodate: True if the read was successful.
 *
 * If a buffer is read through a mechanism that isn't bh_submit(), you
 * can call this function to finish the read.
 */
void end_buffer_read_sync(struct buffer_head *bh, int uptodate)
{
	if (uptodate) {
		set_buffer_uptodate(bh);
	} else {
		/* This happens, due to failed read-ahead attempts. */
		clear_buffer_uptodate(bh);
	}
	unlock_buffer(bh);
}
EXPORT_SYMBOL(end_buffer_read_sync);

/**
 * bh_end_read - I/O end handler for reads
 * @bio: The bio being completed.
 *
 * Pass this function to bh_submit() if you're reading into the buffer,
 * unless you need your own special I/O end handler.
 */
void bh_end_read(struct bio *bio)
{
	struct buffer_head *bh;
	bool uptodate = bio_endio_bh(bio, &bh);
	end_buffer_read_sync(bh, uptodate);
}
EXPORT_SYMBOL(bh_end_read);

/**
 * bh_end_write - I/O end handler for writes
 * @bio: The bio being completed.
 *
 * Pass this function to bh_submit() if you're writing from the buffer,
 * unless you need your own special I/O end handler.
 */
void bh_end_write(struct bio *bio)
{
	struct buffer_head *bh;
	bool success = bio_endio_bh(bio, &bh);

	if (success) {
		set_buffer_uptodate(bh);
	} else {
		buffer_io_error(bh, ", lost sync page write");
		mark_buffer_write_io_error(bh);
		clear_buffer_uptodate(bh);
	}
	unlock_buffer(bh);
}
EXPORT_SYMBOL(bh_end_write);

static struct buffer_head *
__find_get_block_slow(struct block_device *bdev, sector_t block, bool atomic)
{
	struct address_space *bd_mapping = bdev->bd_mapping;
	const int blkbits = bd_mapping->host->i_blkbits;
	struct buffer_head *ret = NULL;
	pgoff_t index;
	struct buffer_head *bh;
	struct buffer_head *head;
	struct folio *folio;
	int all_mapped = 1;
	static DEFINE_RATELIMIT_STATE(last_warned, HZ, 1);

	index = ((loff_t)block << blkbits) / PAGE_SIZE;
	folio = __filemap_get_folio(bd_mapping, index, FGP_ACCESSED, 0);
	if (IS_ERR(folio))
		goto out;

	/*
	 * Folio lock protects the buffers. Callers that cannot block
	 * will fallback to serializing vs try_to_free_buffers() via
	 * the i_private_lock.
	 */
	if (atomic)
		spin_lock(&bd_mapping->i_private_lock);
	else
		folio_lock(folio);

	head = folio_buffers(folio);
	if (!head)
		goto out_unlock;
	/*
	 * Upon a noref migration, the folio lock serializes here;
	 * otherwise bail.
	 */
	if (test_bit_acquire(BH_Migrate, &head->b_state)) {
		WARN_ON(!atomic);
		goto out_unlock;
	}

	bh = head;
	do {
		if (!buffer_mapped(bh))
			all_mapped = 0;
		else if (bh->b_blocknr == block) {
			ret = bh;
			get_bh(bh);
			goto out_unlock;
		}
		bh = bh->b_this_page;
	} while (bh != head);

	/* we might be here because some of the buffers on this page are
	 * not mapped.  This is due to various races between
	 * file io on the block device and getblk.  It gets dealt with
	 * elsewhere, don't buffer_error if we had some unmapped buffers
	 */
	ratelimit_set_flags(&last_warned, RATELIMIT_MSG_ON_RELEASE);
	if (all_mapped && __ratelimit(&last_warned)) {
		printk("__find_get_block_slow() failed. block=%llu, "
		       "b_blocknr=%llu, b_state=0x%08lx, b_size=%zu, "
		       "device %pg blocksize: %d\n",
		       (unsigned long long)block,
		       (unsigned long long)bh->b_blocknr,
		       bh->b_state, bh->b_size, bdev,
		       1 << blkbits);
	}
out_unlock:
	if (atomic)
		spin_unlock(&bd_mapping->i_private_lock);
	else
		folio_unlock(folio);
	folio_put(folio);
out:
	return ret;
}

static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
{
	unsigned long flags;
	struct buffer_head *first;
	struct buffer_head *tmp;
	struct folio *folio;
	int folio_uptodate = 1;

	BUG_ON(!buffer_async_read(bh));

	folio = bh->b_folio;
	if (uptodate) {
		set_buffer_uptodate(bh);
	} else {
		clear_buffer_uptodate(bh);
		buffer_io_error(bh, ", async page read");
	}

	/*
	 * Be _very_ careful from here on. Bad things can happen if
	 * two buffer heads end IO at almost the same time and both
	 * decide that the page is now completely done.
	 */
	first = folio_buffers(folio);
	spin_lock_irqsave(&first->b_uptodate_lock, flags);
	clear_buffer_async_read(bh);
	unlock_buffer(bh);
	tmp = bh;
	do {
		if (!buffer_uptodate(tmp))
			folio_uptodate = 0;
		if (buffer_async_read(tmp)) {
			BUG_ON(!buffer_locked(tmp));
			goto still_busy;
		}
		tmp = tmp->b_this_page;
	} while (tmp != bh);
	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);

	folio_end_read(folio, folio_uptodate);
	return;

still_busy:
	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
}

struct postprocess_bh_ctx {
	struct work_struct work;
	struct buffer_head *bh;
	struct fsverity_info *vi;
};

static void verify_bh(struct work_struct *work)
{
	struct postprocess_bh_ctx *ctx =
		container_of(work, struct postprocess_bh_ctx, work);
	struct buffer_head *bh = ctx->bh;
	bool valid;

	valid = fsverity_verify_blocks(ctx->vi, bh->b_folio, bh->b_size,
				       bh_offset(bh));
	end_buffer_async_read(bh, valid);
	kfree(ctx);
}

static void decrypt_bh(struct work_struct *work)
{
	struct postprocess_bh_ctx *ctx =
		container_of(work, struct postprocess_bh_ctx, work);
	struct buffer_head *bh = ctx->bh;
	int err;

	err = fscrypt_decrypt_pagecache_blocks(bh->b_folio, bh->b_size,
					       bh_offset(bh));
	if (err == 0 && ctx->vi) {
		/*
		 * 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.
		 */
		INIT_WORK(&ctx->work, verify_bh);
		fsverity_enqueue_verify_work(&ctx->work);
		return;
	}
	end_buffer_async_read(bh, err == 0);
	kfree(ctx);
}

/*
 * I/O completion handler for block_read_full_folio() - folios
 * which come unlocked at the end of I/O.
 */
static void bh_end_async_read(struct bio *bio)
{
	struct buffer_head *bh;
	bool uptodate = bio_endio_bh(bio, &bh);
	struct inode *inode = bh->b_folio->mapping->host;
	bool decrypt = fscrypt_inode_uses_fs_layer_crypto(inode);
	struct fsverity_info *vi = NULL;

	/* needed by ext4 */
	if (bh->b_folio->index < DIV_ROUND_UP(inode->i_size, PAGE_SIZE))
		vi = fsverity_get_info(inode);

	/* Decrypt (with fscrypt) and/or verify (with fsverity) if needed. */
	if (uptodate && (decrypt || vi)) {
		struct postprocess_bh_ctx *ctx = kmalloc_obj(*ctx, GFP_ATOMIC);

		if (ctx) {
			ctx->bh = bh;
			ctx->vi = vi;
			if (decrypt) {
				INIT_WORK(&ctx->work, decrypt_bh);
				fscrypt_enqueue_decrypt_work(&ctx->work);
			} else {
				INIT_WORK(&ctx->work, verify_bh);
				fsverity_enqueue_verify_work(&ctx->work);
			}
			return;
		}
		uptodate = false;
	}
	end_buffer_async_read(bh, uptodate);
}

/**
 * bh_end_async_write - I/O end handler for async folio writes
 * @bio: The bio being completed.
 *
 * Pass this function to bh_submit() if you're doing the equivalent of
 * block_write_full_folio().  That is, the folio is unlocked, and will
 * have its writeback flag cleared once all async write buffers have
 * completed.
 */
void bh_end_async_write(struct bio *bio)
{
	struct buffer_head *bh;
	bool success = bio_endio_bh(bio, &bh);
	unsigned long flags;
	struct buffer_head *first;
	struct buffer_head *tmp;
	struct folio *folio;

	BUG_ON(!buffer_async_write(bh));

	folio = bh->b_folio;
	if (success) {
		set_buffer_uptodate(bh);
	} else {
		buffer_io_error(bh, ", lost async page write");
		mark_buffer_write_io_error(bh);
		clear_buffer_uptodate(bh);
	}

	first = folio_buffers(folio);
	spin_lock_irqsave(&first->b_uptodate_lock, flags);

	clear_buffer_async_write(bh);
	unlock_buffer(bh);
	tmp = bh->b_this_page;
	while (tmp != bh) {
		if (buffer_async_write(tmp)) {
			BUG_ON(!buffer_locked(tmp));
			goto still_busy;
		}
		tmp = tmp->b_this_page;
	}
	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
	folio_end_writeback(folio);
	return;

still_busy:
	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
}
EXPORT_SYMBOL(bh_end_async_write);


/*
 * fs/buffer.c contains helper functions for buffer-backed address space's
 * fsync functions.  A common requirement for buffer-based filesystems is
 * that certain data from the backing blockdev needs to be written out for
 * a successful fsync().  For example, ext2 indirect blocks need to be
 * written back and waited upon before fsync() returns.
 *
 * The functions mmb_mark_buffer_dirty(), mmb_sync(), mmb_has_buffers()
 * and mmb_invalidate() are provided for the management of a list of dependent
 * buffers in mapping_metadata_bhs struct.
 *
 * The locking is a little subtle: The list of buffer heads is protected by
 * the lock in mapping_metadata_bhs so functions coming from bdev mapping
 * (such as try_to_free_buffers()) need to safely get to mapping_metadata_bhs
 * using RCU, grab the lock, verify we didn't race with somebody detaching the
 * bh / moving it to different inode and only then proceeding.
 */

void mmb_init(struct mapping_metadata_bhs *mmb, struct address_space *mapping)
{
	spin_lock_init(&mmb->lock);
	INIT_LIST_HEAD(&mmb->list);
	mmb->mapping = mapping;
}
EXPORT_SYMBOL(mmb_init);

static void __remove_assoc_queue(struct mapping_metadata_bhs *mmb,
			         struct buffer_head *bh)
{
	lockdep_assert_held(&mmb->lock);
	list_del_init(&bh->b_assoc_buffers);
	WARN_ON(!bh->b_mmb);
	bh->b_mmb = NULL;
}

static void remove_assoc_queue(struct buffer_head *bh)
{
	struct mapping_metadata_bhs *mmb;

	/*
	 * The locking dance is ugly here. We need to acquire the lock
	 * protecting the metadata bh list while possibly racing with bh
	 * being removed from the list or moved to a different one.  We
	 * use RCU to pin mapping_metadata_bhs in memory to
	 * opportunistically acquire the lock and then recheck the bh
	 * didn't move under us.
	 */
	while (bh->b_mmb) {
		rcu_read_lock();
		mmb = READ_ONCE(bh->b_mmb);
		if (mmb) {
			spin_lock(&mmb->lock);
			if (bh->b_mmb == mmb)
				__remove_assoc_queue(mmb, bh);
			spin_unlock(&mmb->lock);
		}
		rcu_read_unlock();
	}
}

bool mmb_has_buffers(struct mapping_metadata_bhs *mmb)
{
	return !list_empty(&mmb->list);
}
EXPORT_SYMBOL_GPL(mmb_has_buffers);

/**
 * mmb_sync - write out & wait upon all buffers in a list
 * @mmb: the list of buffers to write
 *
 * Starts I/O against the buffers in the given list and waits upon
 * that I/O. Basically, this is a convenience function for fsync().  @mmb is
 * for a file or directory which needs those buffers to be written for a
 * successful fsync().
 *
 * We have conflicting pressures: we want to make sure that all
 * initially dirty buffers get waited on, but that any subsequently
 * dirtied buffers don't.  After all, we don't want fsync to last
 * forever if somebody is actively writing to the file.
 *
 * Do this in two main stages: first we copy dirty buffers to a
 * temporary inode list, queueing the writes as we go. Then we clean
 * up, waiting for those writes to complete. mark_buffer_dirty_inode()
 * doesn't touch b_assoc_buffers list if b_mmb is not NULL so we are sure the
 * buffer stays on our list until IO completes (at which point it can be
 * reaped).
 */
int mmb_sync(struct mapping_metadata_bhs *mmb)
{
	struct buffer_head *bh;
	int err = 0;
	struct blk_plug plug;
	LIST_HEAD(tmp);

	if (!mmb_has_buffers(mmb))
		return 0;

	blk_start_plug(&plug);

	spin_lock(&mmb->lock);
	while (!list_empty(&mmb->list)) {
		bh = BH_ENTRY(mmb->list.next);
		WARN_ON_ONCE(bh->b_mmb != mmb);
		__remove_assoc_queue(mmb, bh);
		/* Avoid race with mark_buffer_dirty_inode() which does
		 * a lockless check and we rely on seeing the dirty bit */
		smp_mb();
		if (buffer_dirty(bh) || buffer_locked(bh)) {
			list_add(&bh->b_assoc_buffers, &tmp);
			bh->b_mmb = mmb;
			if (buffer_dirty(bh)) {
				get_bh(bh);
				spin_unlock(&mmb->lock);
				/*
				 * Ensure any pending I/O completes so that
				 * write_dirty_buffer() actually writes the
				 * current contents - it is a noop if I/O is
				 * still in flight on potentially older
				 * contents.
				 */
				write_dirty_buffer(bh, REQ_SYNC);

				/*
				 * Kick off IO for the previous mapping. Note
				 * that we will not run the very last mapping,
				 * wait_on_buffer() will do that for us
				 * through sync_buffer().
				 */
				brelse(bh);
				spin_lock(&mmb->lock);
			}
		}
	}

	spin_unlock(&mmb->lock);
	blk_finish_plug(&plug);
	spin_lock(&mmb->lock);

	while (!list_empty(&tmp)) {
		bh = BH_ENTRY(tmp.prev);
		get_bh(bh);
		__remove_assoc_queue(mmb, bh);
		/* Avoid race with mark_buffer_dirty_inode() which does
		 * a lockless check and we rely on seeing the dirty bit */
		smp_mb();
		if (buffer_dirty(bh)) {
			list_add(&bh->b_assoc_buffers, &mmb->list);
			bh->b_mmb = mmb;
		}
		spin_unlock(&mmb->lock);
		wait_on_buffer(bh);
		if (!buffer_uptodate(bh))
			err = -EIO;
		brelse(bh);
		spin_lock(&mmb->lock);
	}
	spin_unlock(&mmb->lock);
	return err;
}
EXPORT_SYMBOL(mmb_sync);

/**
 * mmb_fsync_noflush - fsync implementation for simple filesystems with
 * 		       metadata buffers list
 *
 * @file:	file to synchronize
 * @mmb:	list of metadata bhs to flush
 * @start:	start offset in bytes
 * @end:	end offset in bytes (inclusive)
 * @datasync:	only synchronize essential metadata if true
 *
 * This is an implementation of the fsync method for simple filesystems which
 * track all non-inode metadata in the buffers list hanging off the @mmb
 * structure.
 */
int mmb_fsync_noflush(struct file *file, struct mapping_metadata_bhs *mmb,
		      loff_t start, loff_t end, bool datasync)
{
	struct inode *inode = file->f_mapping->host;
	int err;
	int ret = 0;

	err = file_write_and_wait_range(file, start, end);
	if (err)
		return err;

	if (mmb)
		ret = mmb_sync(mmb);
	if (!(inode_state_read_once(inode) & I_DIRTY_ALL))
		goto out;
	if (datasync && !(inode_state_read_once(inode) & I_DIRTY_DATASYNC))
		goto out;

	err = sync_inode_metadata(inode, 1);
	if (ret == 0)
		ret = err;

out:
	/* check and advance again to catch errors after syncing out buffers */
	err = file_check_and_advance_wb_err(file);
	if (ret == 0)
		ret = err;
	return ret;
}
EXPORT_SYMBOL(mmb_fsync_noflush);

/**
 * mmb_fsync - fsync implementation for simple filesystems with metadata
 * 	       buffers list
 *
 * @file:	file to synchronize
 * @mmb:	list of metadata bhs to flush
 * @start:	start offset in bytes
 * @end:	end offset in bytes (inclusive)
 * @datasync:	only synchronize essential metadata if true
 *
 * This is an implementation of the fsync method for simple filesystems which
 * track all non-inode metadata in the buffers list hanging off the @mmb
 * structure. This also makes sure that a device cache flush operation is
 * called at the end.
 */
int mmb_fsync(struct file *file, struct mapping_metadata_bhs *mmb,
	      loff_t start, loff_t end, bool datasync)
{
	struct inode *inode = file->f_mapping->host;
	int ret;

	ret = mmb_fsync_noflush(file, mmb, start, end, datasync);
	if (!ret)
		ret = blkdev_issue_flush(inode->i_sb->s_bdev);
	return ret;
}
EXPORT_SYMBOL(mmb_fsync);

/*
 * Called when we've recently written block `bblock', and it is known that
 * `bblock' was for a buffer_boundary() buffer.  This means that the block at
 * `bblock + 1' is probably a dirty indirect block.  Hunt it down and, if it's
 * dirty, schedule it for IO.  So that indirects merge nicely with their data.
 */
void write_boundary_block(struct block_device *bdev,
			sector_t bblock, unsigned blocksize)
{
	struct buffer_head *bh;

	bh = __find_get_block_nonatomic(bdev, bblock + 1, blocksize);
	if (bh) {
		if (buffer_dirty(bh))
			write_dirty_buffer(bh, 0);
		put_bh(bh);
	}
}

void mmb_mark_buffer_dirty(struct buffer_head *bh,
			   struct mapping_metadata_bhs *mmb)
{
	mark_buffer_dirty(bh);
	if (!bh->b_mmb) {
		spin_lock(&mmb->lock);
		/*
		 * For a corrupted filesystem with multiply claimed blocks this
		 * can fail. Avoid corrupting the linked list in that case.
		 */
		if (cmpxchg(&bh->b_mmb, NULL, mmb) != NULL) {
			spin_unlock(&mmb->lock);
			return;
		}
		list_move_tail(&bh->b_assoc_buffers, &mmb->list);
		spin_unlock(&mmb->lock);
	}
}
EXPORT_SYMBOL(mmb_mark_buffer_dirty);

/**
 * block_dirty_folio - Mark a folio as dirty.
 * @mapping: The address space containing this folio.
 * @folio: The folio to mark dirty.
 *
 * Filesystems which use buffer_heads can use this function as their
 * ->dirty_folio implementation.  Some filesystems need to do a little
 * work before calling this function.  Filesystems which do not use
 * buffer_heads should call filemap_dirty_folio() instead.
 *
 * If the folio has buffers, the uptodate buffers are set dirty, to
 * preserve dirty-state coherency between the folio and the buffers.
 * Buffers added to a dirty folio are created dirty.
 *
 * The buffers are dirtied before the folio is dirtied.  There's a small
 * race window in which writeback may see the folio cleanness but not the
 * buffer dirtiness.  That's fine.  If this code were to set the folio
 * dirty before the buffers, writeback could clear the folio dirty flag,
 * see a bunch of clean buffers and we'd end up with dirty buffers/clean
 * folio on the dirty folio list.
 *
 * We use i_private_lock to lock against try_to_free_buffers() while
 * using the folio's buffer list.  This also prevents clean buffers
 * being added to the folio after it was set dirty.
 *
 * Context: May only be called from process context.  Does not sleep.
 * Caller must ensure that @folio cannot be truncated during this call,
 * typically by holding the folio lock or having a page in the folio
 * mapped and holding the page table lock.
 *
 * Return: True if the folio was dirtied; false if it was already dirtied.
 */
bool block_dirty_folio(struct address_space *mapping, struct folio *folio)
{
	struct buffer_head *head;
	bool newly_dirty;

	spin_lock(&mapping->i_private_lock);
	head = folio_buffers(folio);
	if (head) {
		struct buffer_head *bh = head;

		do {
			set_buffer_dirty(bh);
			bh = bh->b_this_page;
		} while (bh != head);
	}
	/*
	 * Lock out page's memcg migration to keep PageDirty
	 * synchronized with per-memcg dirty page counters.
	 */
	newly_dirty = !folio_test_set_dirty(folio);
	spin_unlock(&mapping->i_private_lock);

	if (newly_dirty)
		__folio_mark_dirty(folio, mapping, 1);

	if (newly_dirty)
		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);

	return newly_dirty;
}
EXPORT_SYMBOL(block_dirty_folio);

/*
 * Invalidate any and all dirty buffers on a given buffers list.  We are
 * probably unmounting the fs, but that doesn't mean we have already
 * done a sync().  Just drop the buffers from the inode list.
 */
void mmb_invalidate(struct mapping_metadata_bhs *mmb)
{
	if (mmb_has_buffers(mmb)) {
		spin_lock(&mmb->lock);
		while (!list_empty(&mmb->list))
			__remove_assoc_queue(mmb, BH_ENTRY(mmb->list.next));
		spin_unlock(&mmb->lock);
	}
}
EXPORT_SYMBOL(mmb_invalidate);

/*
 * Create the appropriate buffers when given a folio for data area and
 * the size of each buffer.. Use the bh->b_this_page linked list to
 * follow the buffers created.  Return NULL if unable to create more
 * buffers.
 *
 * The retry flag is used to differentiate async IO (paging, swapping)
 * which may not fail from ordinary buffer allocations.
 */
struct buffer_head *folio_alloc_buffers(struct folio *folio, unsigned long size,
					gfp_t gfp)
{
	struct buffer_head *bh, *head;
	long offset;
	struct mem_cgroup *memcg, *old_memcg;

	memcg = get_mem_cgroup_from_folio(folio);
	old_memcg = set_active_memcg(memcg);

	head = NULL;
	offset = folio_size(folio);
	while ((offset -= size) >= 0) {
		bh = alloc_buffer_head(gfp);
		if (!bh)
			goto no_grow;

		bh->b_this_page = head;
		bh->b_blocknr = -1;
		head = bh;

		bh->b_size = size;

		/* Link the buffer to its folio */
		folio_set_bh(bh, folio, offset);
	}
out:
	set_active_memcg(old_memcg);
	mem_cgroup_put(memcg);
	return head;
/*
 * In case anything failed, we just free everything we got.
 */
no_grow:
	if (head) {
		do {
			bh = head;
			head = head->b_this_page;
			free_buffer_head(bh);
		} while (head);
	}

	goto out;
}
EXPORT_SYMBOL_GPL(folio_alloc_buffers);

struct buffer_head *alloc_page_buffers(struct page *page, unsigned long size)
{
	gfp_t gfp = GFP_NOFS | __GFP_ACCOUNT;

	return folio_alloc_buffers(page_folio(page), size, gfp);
}
EXPORT_SYMBOL_GPL(alloc_page_buffers);

static inline void link_dev_buffers(struct folio *folio,
		struct buffer_head *head)
{
	struct buffer_head *bh, *tail;

	bh = head;
	do {
		tail = bh;
		bh = bh->b_this_page;
	} while (bh);
	tail->b_this_page = head;
	folio_attach_private(folio, head);
}

static sector_t blkdev_max_block(struct block_device *bdev, unsigned int size)
{
	sector_t retval = ~((sector_t)0);
	loff_t sz = bdev_nr_bytes(bdev);

	if (sz) {
		unsigned int sizebits = blksize_bits(size);
		retval = (sz >> sizebits);
	}
	return retval;
}

/*
 * Initialise the state of a blockdev folio's buffers.
 */
static sector_t folio_init_buffers(struct folio *folio,
		struct block_device *bdev, unsigned size)
{
	struct buffer_head *head = folio_buffers(folio);
	struct buffer_head *bh = head;
	bool uptodate = folio_test_uptodate(folio);
	sector_t block = div_u64(folio_pos(folio), size);
	sector_t end_block = blkdev_max_block(bdev, size);

	do {
		if (!buffer_mapped(bh)) {
			bh->b_private = NULL;
			bh->b_bdev = bdev;
			bh->b_blocknr = block;
			if (uptodate)
				set_buffer_uptodate(bh);
			if (block < end_block)
				set_buffer_mapped(bh);
		}
		block++;
		bh = bh->b_this_page;
	} while (bh != head);

	/*
	 * Caller needs to validate requested block against end of device.
	 */
	return end_block;
}

/*
 * Create the page-cache folio that contains the requested block.
 *
 * This is used purely for blockdev mappings.
 *
 * Returns false if we have a failure which cannot be cured by retrying
 * without sleeping.  Returns true if we succeeded, or the caller should retry.
 */
static bool grow_dev_folio(struct block_device *bdev, sector_t block,
		pgoff_t index, unsigned size, gfp_t gfp)
{
	struct address_space *mapping = bdev->bd_mapping;
	struct folio *folio;
	struct buffer_head *bh;
	sector_t end_block = 0;

	folio = __filemap_get_folio(mapping, index,
			FGP_LOCK | FGP_ACCESSED | FGP_CREAT, gfp);
	if (IS_ERR(folio))
		return false;

	bh = folio_buffers(folio);
	if (bh) {
		if (bh->b_size == size) {
			end_block = folio_init_buffers(folio, bdev, size);
			goto unlock;
		}

		/*
		 * Retrying may succeed; for example the folio may finish
		 * writeback, or buffers may be cleaned.  This should not
		 * happen very often; maybe we have old buffers attached to
		 * this blockdev's page cache and we're trying to change
		 * the block size?
		 */
		if (!try_to_free_buffers(folio)) {
			end_block = ~0ULL;
			goto unlock;
		}
	}

	bh = folio_alloc_buffers(folio, size, gfp | __GFP_ACCOUNT);
	if (!bh)
		goto unlock;

	/*
	 * Link the folio to the buffers and initialise them.  Take the
	 * lock to be atomic wrt __find_get_block(), which does not
	 * run under the folio lock.
	 */
	spin_lock(&mapping->i_private_lock);
	link_dev_buffers(folio, bh);
	end_block = folio_init_buffers(folio, bdev, size);
	spin_unlock(&mapping->i_private_lock);
unlock:
	folio_unlock(folio);
	folio_put(folio);
	return block < end_block;
}

/*
 * Create buffers for the specified block device block's folio.  If
 * that folio was dirty, the buffers are set dirty also.  Returns false
 * if we've hit a permanent error.
 */
static bool grow_buffers(struct block_device *bdev, sector_t block,
		unsigned size, gfp_t gfp)
{
	loff_t pos;

	/*
	 * Check for a block which lies outside our maximum possible
	 * pagecache index.
	 */
	if (check_mul_overflow(block, (sector_t)size, &pos) || pos > MAX_LFS_FILESIZE) {
		printk(KERN_ERR "%s: requested out-of-range block %llu for device %pg\n",
			__func__, (unsigned long long)block,
			bdev);
		return false;
	}

	/* Create a folio with the proper size buffers */
	return grow_dev_folio(bdev, block, pos / PAGE_SIZE, size, gfp);
}

static struct buffer_head *
__getblk_slow(struct block_device *bdev, sector_t block,
	     unsigned size, gfp_t gfp)
{
	bool blocking = gfpflags_allow_blocking(gfp);

	if (WARN_ON_ONCE(!IS_ALIGNED(size, bdev_logical_block_size(bdev)))) {
		printk(KERN_ERR "getblk(): block size %d not aligned to logical block size %d\n",
		       size, bdev_logical_block_size(bdev));
		return NULL;
	}

	for (;;) {
		struct buffer_head *bh;

		if (!grow_buffers(bdev, block, size, gfp))
			return NULL;

		if (blocking)
			bh = __find_get_block_nonatomic(bdev, block, size);
		else
			bh = __find_get_block(bdev, block, size);
		if (bh)
			return bh;
	}
}

/*
 * The relationship between dirty buffers and dirty pages:
 *
 * Whenever a page has any dirty buffers, the page's dirty bit is set, and
 * the page is tagged dirty in the page cache.
 *
 * At all times, the dirtiness of the buffers represents the dirtiness of
 * subsections of the page.  If the page has buffers, the page dirty bit is
 * merely a hint about the true dirty state.
 *
 * When a page is set dirty in its entirety, all its buffers are marked dirty
 * (if the page has buffers).
 *
 * When a buffer is marked dirty, its page is dirtied, but the page's other
 * buffers are not.
 *
 * Also.  When blockdev buffers are explicitly read with bread(), they
 * individually become uptodate.  But their backing page remains not
 * uptodate - even if all of its buffers are uptodate.  A subsequent
 * block_read_full_folio() against that folio will discover all the uptodate
 * buffers, will set the folio uptodate and will perform no I/O.
 */

/**
 * mark_buffer_dirty - mark a buffer_head as needing writeout
 * @bh: the buffer_head to mark dirty
 *
 * mark_buffer_dirty() will set the dirty bit against the buffer, then set
 * its backing page dirty, then tag the page as dirty in the page cache
 * and then attach the address_space's inode to its superblock's dirty
 * inode list.
 *
 * mark_buffer_dirty() is atomic.  It takes bh->b_folio->mapping->i_private_lock,
 * i_pages lock and mapping->host->i_lock.
 */
void mark_buffer_dirty(struct buffer_head *bh)
{
	WARN_ON_ONCE(!buffer_uptodate(bh));

	trace_block_dirty_buffer(bh);

	/*
	 * Very *carefully* optimize the it-is-already-dirty case.
	 *
	 * Don't let the final "is it dirty" escape to before we
	 * perhaps modified the buffer.
	 */
	if (buffer_dirty(bh)) {
		smp_mb();
		if (buffer_dirty(bh))
			return;
	}

	if (!test_set_buffer_dirty(bh)) {
		struct folio *folio = bh->b_folio;
		struct address_space *mapping = NULL;

		if (!folio_test_set_dirty(folio)) {
			mapping = folio->mapping;
			if (mapping)
				__folio_mark_dirty(folio, mapping, 0);
		}
		if (mapping)
			__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
	}
}
EXPORT_SYMBOL(mark_buffer_dirty);

void mark_buffer_write_io_error(struct buffer_head *bh)
{
	set_buffer_write_io_error(bh);
	/* FIXME: do we need to set this in both places? */
	if (bh->b_folio && bh->b_folio->mapping)
		mapping_set_error(bh->b_folio->mapping, -EIO);
	if (bh->b_mmb)
		mapping_set_error(bh->b_mmb->mapping, -EIO);
}
EXPORT_SYMBOL(mark_buffer_write_io_error);

/**
 * __brelse - Release a buffer.
 * @bh: The buffer to release.
 *
 * This variant of brelse() can be called if @bh is guaranteed to not be NULL.
 */
void __brelse(struct buffer_head *bh)
{
	if (atomic_read(&bh->b_count)) {
		put_bh(bh);
		return;
	}
	WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n");
}
EXPORT_SYMBOL(__brelse);

/**
 * __bforget - Discard any dirty data in a buffer.
 * @bh: The buffer to forget.
 *
 * This variant of bforget() can be called if @bh is guaranteed to not
 * be NULL.
 */
void __bforget(struct buffer_head *bh)
{
	clear_buffer_dirty(bh);
	remove_assoc_queue(bh);
	__brelse(bh);
}
EXPORT_SYMBOL(__bforget);

static void buffer_set_crypto_ctx(struct bio *bio, const struct buffer_head *bh,
				  gfp_t gfp_mask)
{
	const struct address_space *mapping = folio_mapping(bh->b_folio);

	/*
	 * The ext4 journal (jbd2) can submit a buffer_head it directly created
	 * for a non-pagecache page.  fscrypt doesn't care about these.
	 */
	if (!mapping)
		return;
	fscrypt_set_bio_crypt_ctx(bio, mapping->host,
			folio_pos(bh->b_folio) + bh_offset(bh), gfp_mask);
}

static void __bh_submit(struct buffer_head *bh, blk_opf_t opf,
		enum rw_hint write_hint, struct writeback_control *wbc,
		bio_end_io_t end_bio)
{
	const enum req_op op = opf & REQ_OP_MASK;
	struct bio *bio;

	BUG_ON(!buffer_locked(bh));
	BUG_ON(!buffer_mapped(bh));
	BUG_ON(buffer_delay(bh));
	BUG_ON(buffer_unwritten(bh));

	/*
	 * Only clear out a write error when rewriting
	 */
	if (test_set_buffer_req(bh) && (op == REQ_OP_WRITE))
		clear_buffer_write_io_error(bh);

	if (buffer_meta(bh))
		opf |= REQ_META;
	if (buffer_prio(bh))
		opf |= REQ_PRIO;

	bio = bio_alloc(bh->b_bdev, 1, opf, GFP_NOIO);

	if (IS_ENABLED(CONFIG_FS_ENCRYPTION))
		buffer_set_crypto_ctx(bio, bh, GFP_NOIO);

	bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
	bio->bi_write_hint = write_hint;

	bio_add_folio_nofail(bio, bh->b_folio, bh->b_size, bh_offset(bh));

	bio->bi_end_io = end_bio;
	bio->bi_private = bh;

	/* Take care of bh's that straddle the end of the device */
	guard_bio_eod(bio);

	if (wbc) {
		wbc_init_bio(wbc, bio);
		wbc_account_cgroup_owner(wbc, bh->b_folio, bh->b_size);
	}

	blk_crypto_submit_bio(bio);
}

/**
 * bh_submit - Start I/O against a buffer head
 * @bh: The buffer head to perform I/O on.
 * @opf: Operation and flags for bio.
 * @end_io: The routine to call when I/O has completed.
 *
 * If you need to do I/O on an individual bh (instead of allowing the
 * page cache to do I/O on the folio that it is in), call this function.
 */
void bh_submit(struct buffer_head *bh, blk_opf_t opf, bio_end_io_t end_io)
{
	__bh_submit(bh, opf, WRITE_LIFE_NOT_SET, NULL, end_io);
}
EXPORT_SYMBOL(bh_submit);

static struct buffer_head *__bread_slow(struct buffer_head *bh)
{
	lock_buffer(bh);
	if (buffer_uptodate(bh)) {
		unlock_buffer(bh);
		return bh;
	} else {
		bh_submit(bh, REQ_OP_READ, bh_end_read);
		wait_on_buffer(bh);
		if (buffer_uptodate(bh))
			return bh;
	}
	brelse(bh);
	return NULL;
}

/*
 * Per-cpu buffer LRU implementation.  To reduce the cost of __find_get_block().
 * The bhs[] array is sorted - newest buffer is at bhs[0].  Buffers have their
 * refcount elevated by one when they're in an LRU.  A buffer can only appear
 * once in a particular CPU's LRU.  A single buffer can be present in multiple
 * CPU's LRUs at the same time.
 *
 * This is a transparent caching front-end to sb_bread(), sb_getblk() and
 * sb_find_get_block().
 *
 * The LRUs themselves only need locking against invalidate_bh_lrus.  We use
 * a local interrupt disable for that.
 */

#define BH_LRU_SIZE	16

struct bh_lru {
	struct buffer_head *bhs[BH_LRU_SIZE];
};

static DEFINE_PER_CPU(struct bh_lru, bh_lrus) = {{ NULL }};

#ifdef CONFIG_SMP
#define bh_lru_lock()	local_irq_disable()
#define bh_lru_unlock()	local_irq_enable()
#else
#define bh_lru_lock()	preempt_disable()
#define bh_lru_unlock()	preempt_enable()
#endif

static inline void check_irqs_on(void)
{
#ifdef irqs_disabled
	BUG_ON(irqs_disabled());
#endif
}

/*
 * Install a buffer_head into this cpu's LRU.  If not already in the LRU, it is
 * inserted at the front, and the buffer_head at the back if any is evicted.
 * Or, if already in the LRU it is moved to the front.
 */
static void bh_lru_install(struct buffer_head *bh)
{
	struct buffer_head *evictee = bh;
	struct bh_lru *b;
	int i;

	check_irqs_on();
	bh_lru_lock();

	/*
	 * the refcount of buffer_head in bh_lru prevents dropping the
	 * attached page(i.e., try_to_free_buffers) so it could cause
	 * failing page migration.
	 * Skip putting upcoming bh into bh_lru until migration is done.
	 */
	if (lru_cache_disabled() || cpu_is_isolated(smp_processor_id())) {
		bh_lru_unlock();
		return;
	}

	b = this_cpu_ptr(&bh_lrus);
	for (i = 0; i < BH_LRU_SIZE; i++) {
		swap(evictee, b->bhs[i]);
		if (evictee == bh) {
			bh_lru_unlock();
			return;
		}
	}

	get_bh(bh);
	bh_lru_unlock();
	brelse(evictee);
}

/*
 * Look up the bh in this cpu's LRU.  If it's there, move it to the head.
 */
static struct buffer_head *
lookup_bh_lru(struct block_device *bdev, sector_t block, unsigned size)
{
	struct buffer_head *ret = NULL;
	unsigned int i;

	check_irqs_on();
	bh_lru_lock();
	if (cpu_is_isolated(smp_processor_id())) {
		bh_lru_unlock();
		return NULL;
	}
	for (i = 0; i < BH_LRU_SIZE; i++) {
		struct buffer_head *bh = __this_cpu_read(bh_lrus.bhs[i]);

		if (bh && bh->b_blocknr == block && bh->b_bdev == bdev &&
		    bh->b_size == size) {
			if (i) {
				while (i) {
					__this_cpu_write(bh_lrus.bhs[i],
						__this_cpu_read(bh_lrus.bhs[i - 1]));
					i--;
				}
				__this_cpu_write(bh_lrus.bhs[0], bh);
			}
			get_bh(bh);
			ret = bh;
			break;
		}
	}
	bh_lru_unlock();
	return ret;
}

/*
 * Perform a pagecache lookup for the matching buffer.  If it's there, refresh
 * it in the LRU and mark it as accessed.  If it is not present then return
 * NULL. Atomic context callers may also return NULL if the buffer is being
 * migrated; similarly the page is not marked accessed either.
 */
static struct buffer_head *
find_get_block_common(struct block_device *bdev, sector_t block,
			unsigned size, bool atomic)
{
	struct buffer_head *bh = lookup_bh_lru(bdev, block, size);

	if (bh == NULL) {
		/* __find_get_block_slow will mark the page accessed */
		bh = __find_get_block_slow(bdev, block, atomic);
		if (bh)
			bh_lru_install(bh);
	} else
		touch_buffer(bh);

	return bh;
}

struct buffer_head *
__find_get_block(struct block_device *bdev, sector_t block, unsigned size)
{
	return find_get_block_common(bdev, block, size, true);
}
EXPORT_SYMBOL(__find_get_block);

/* same as __find_get_block() but allows sleeping contexts */
struct buffer_head *
__find_get_block_nonatomic(struct block_device *bdev, sector_t block,
			   unsigned size)
{
	return find_get_block_common(bdev, block, size, false);
}
EXPORT_SYMBOL(__find_get_block_nonatomic);

/**
 * bdev_getblk - Get a buffer_head in a block device's buffer cache.
 * @bdev: The block device.
 * @block: The block number.
 * @size: The size of buffer_heads for this @bdev.
 * @gfp: The memory allocation flags to use.
 *
 * The returned buffer head has its reference count incremented, but is
 * not locked.  The caller should call brelse() when it has finished
 * with the buffer.  The buffer may not be uptodate.  If needed, the
 * caller can bring it uptodate either by reading it or overwriting it.
 *
 * Return: The buffer head, or NULL if memory could not be allocated.
 */
struct buffer_head *bdev_getblk(struct block_device *bdev, sector_t block,
		unsigned size, gfp_t gfp)
{
	struct buffer_head *bh;

	if (gfpflags_allow_blocking(gfp))
		bh = __find_get_block_nonatomic(bdev, block, size);
	else
		bh = __find_get_block(bdev, block, size);

	might_alloc(gfp);
	if (bh)
		return bh;

	return __getblk_slow(bdev, block, size, gfp);
}
EXPORT_SYMBOL(bdev_getblk);

/*
 * Do async read-ahead on a buffer..
 */
void __breadahead(struct block_device *bdev, sector_t block, unsigned size)
{
	struct buffer_head *bh = bdev_getblk(bdev, block, size,
			GFP_NOWAIT | __GFP_MOVABLE);

	if (likely(bh)) {
		bh_readahead(bh, REQ_RAHEAD);
		brelse(bh);
	}
}
EXPORT_SYMBOL(__breadahead);

/**
 * __bread_gfp() - Read a block.
 * @bdev: The block device to read from.
 * @block: Block number in units of block size.
 * @size: The block size of this device in bytes.
 * @gfp: Not page allocation flags; see below.
 *
 * You are not expected to call this function.  You should use one of
 * sb_bread(), sb_bread_unmovable() or __bread().
 *
 * Read a specified block, and return the buffer head that refers to it.
 * If @gfp is 0, the memory will be allocated using the block device's
 * default GFP flags.  If @gfp is __GFP_MOVABLE, the memory may be
 * allocated from a movable area.  Do not pass in a complete set of
 * GFP flags.
 *
 * The returned buffer head has its refcount increased.  The caller should
 * call brelse() when it has finished with the buffer.
 *
 * Context: May sleep waiting for I/O.
 * Return: NULL if the block was unreadable.
 */
struct buffer_head *__bread_gfp(struct block_device *bdev, sector_t block,
		unsigned size, gfp_t gfp)
{
	struct buffer_head *bh;

	gfp |= mapping_gfp_constraint(bdev->bd_mapping, ~__GFP_FS);

	/*
	 * Prefer looping in the allocator rather than here, at least that
	 * code knows what it's doing.
	 */
	gfp |= __GFP_NOFAIL;

	bh = bdev_getblk(bdev, block, size, gfp);

	if (likely(bh) && !buffer_uptodate(bh))
		bh = __bread_slow(bh);
	return bh;
}
EXPORT_SYMBOL(__bread_gfp);

static void __invalidate_bh_lrus(struct bh_lru *b)
{
	int i;

	for (i = 0; i < BH_LRU_SIZE; i++) {
		brelse(b->bhs[i]);
		b->bhs[i] = NULL;
	}
}
/*
 * invalidate_bh_lrus() is called rarely - but not only at unmount.
 * This doesn't race because it runs in each cpu either in irq
 * or with preempt disabled.
 */
static void invalidate_bh_lru(void *arg)
{
	struct bh_lru *b = &get_cpu_var(bh_lrus);

	__invalidate_bh_lrus(b);
	put_cpu_var(bh_lrus);
}

bool has_bh_in_lru(int cpu, void *dummy)
{
	struct bh_lru *b = per_cpu_ptr(&bh_lrus, cpu);
	int i;

	for (i = 0; i < BH_LRU_SIZE; i++) {
		if (b->bhs[i])
			return true;
	}

	return false;
}

void invalidate_bh_lrus(void)
{
	on_each_cpu_cond(has_bh_in_lru, invalidate_bh_lru, NULL, 1);
}
EXPORT_SYMBOL_GPL(invalidate_bh_lrus);

/*
 * It's called from workqueue context so we need a bh_lru_lock to close
 * the race with preemption/irq.
 */
void invalidate_bh_lrus_cpu(void)
{
	struct bh_lru *b;

	bh_lru_lock();
	b = this_cpu_ptr(&bh_lrus);
	__invalidate_bh_lrus(b);
	bh_lru_unlock();
}

void folio_set_bh(struct buffer_head *bh, struct folio *folio,
		  unsigned long offset)
{
	bh->b_folio = folio;
	BUG_ON(offset >= folio_size(folio));
	if (folio_test_highmem(folio))
		/*
		 * This catches illegal uses and preserves the offset:
		 */
		bh->b_data = (char *)(0 + offset);
	else
		bh->b_data = folio_address(folio) + offset;
}
EXPORT_SYMBOL(folio_set_bh);

/*
 * Called when truncating a buffer on a page completely.
 */

/* Bits that are cleared during an invalidate */
#define BUFFER_FLAGS_DISCARD \
	(1 << BH_Mapped | 1 << BH_New | 1 << BH_Req | \
	 1 << BH_Delay | 1 << BH_Unwritten)

static void discard_buffer(struct buffer_head * bh)
{
	unsigned long b_state;

	lock_buffer(bh);
	clear_buffer_dirty(bh);
	bh->b_bdev = NULL;
	b_state = READ_ONCE(bh->b_state);
	do {
	} while (!try_cmpxchg_relaxed(&bh->b_state, &b_state,
				      b_state & ~BUFFER_FLAGS_DISCARD));
	unlock_buffer(bh);
}

/**
 * block_invalidate_folio - Invalidate part or all of a buffer-backed folio.
 * @folio: The folio which is affected.
 * @offset: start of the range to invalidate
 * @length: length of the range to invalidate
 *
 * block_invalidate_folio() is called when all or part of the folio has been
 * invalidated by a truncate operation.
 *
 * block_invalidate_folio() does not have to release all buffers, but it must
 * ensure that no dirty buffer is left outside @offset and that no I/O
 * is underway against any of the blocks which are outside the truncation
 * point.  Because the caller is about to free (and possibly reuse) those
 * blocks on-disk.
 */
void block_invalidate_folio(struct folio *folio, size_t offset, size_t length)
{
	struct buffer_head *head, *bh, *next;
	size_t curr_off = 0;
	size_t stop = length + offset;

	BUG_ON(!folio_test_locked(folio));

	/*
	 * Check for overflow
	 */
	BUG_ON(stop > folio_size(folio) || stop < length);

	head = folio_buffers(folio);
	if (!head)
		return;

	bh = head;
	do {
		size_t next_off = curr_off + bh->b_size;
		next = bh->b_this_page;

		/*
		 * Are we still fully in range ?
		 */
		if (next_off > stop)
			goto out;

		/*
		 * is this block fully invalidated?
		 */
		if (offset <= curr_off)
			discard_buffer(bh);
		curr_off = next_off;
		bh = next;
	} while (bh != head);

	/*
	 * We release buffers only if the entire folio is being invalidated.
	 * The get_block cached value has been unconditionally invalidated,
	 * so real IO is not possible anymore.
	 */
	if (length == folio_size(folio))
		filemap_release_folio(folio, 0);
out:
	folio_clear_mappedtodisk(folio);
}
EXPORT_SYMBOL(block_invalidate_folio);

/*
 * We attach and possibly dirty the buffers atomically wrt
 * block_dirty_folio() via i_private_lock.  try_to_free_buffers
 * is already excluded via the folio lock.
 */
struct buffer_head *create_empty_buffers(struct folio *folio,
		unsigned long blocksize, unsigned long b_state)
{
	struct buffer_head *bh, *head, *tail;
	gfp_t gfp = GFP_NOFS | __GFP_ACCOUNT | __GFP_NOFAIL;

	head = folio_alloc_buffers(folio, blocksize, gfp);
	bh = head;
	do {
		bh->b_state |= b_state;
		tail = bh;
		bh = bh->b_this_page;
	} while (bh);
	tail->b_this_page = head;

	spin_lock(&folio->mapping->i_private_lock);
	if (folio_test_uptodate(folio) || folio_test_dirty(folio)) {
		bh = head;
		do {
			if (folio_test_dirty(folio))
				set_buffer_dirty(bh);
			if (folio_test_uptodate(folio))
				set_buffer_uptodate(bh);
			bh = bh->b_this_page;
		} while (bh != head);
	}
	folio_attach_private(folio, head);
	spin_unlock(&folio->mapping->i_private_lock);

	return head;
}
EXPORT_SYMBOL(create_empty_buffers);

/**
 * clean_bdev_aliases: clean a range of buffers in block device
 * @bdev: Block device to clean buffers in
 * @block: Start of a range of blocks to clean
 * @len: Number of blocks to clean
 *
 * We are taking a range of blocks for data and we don't want writeback of any
 * buffer-cache aliases starting from return from this function and until the
 * moment when something will explicitly mark the buffer dirty (hopefully that
 * will not happen until we will free that block ;-) We don't even need to mark
 * it not-uptodate - nobody can expect anything from a newly allocated buffer
 * anyway. We used to use unmap_buffer() for such invalidation, but that was
 * wrong. We definitely don't want to mark the alias unmapped, for example - it
 * would confuse anyone who might pick it with bread() afterwards...
 *
 * Also..  Note that bforget() doesn't lock the buffer.  So there can be
 * writeout I/O going on against recently-freed buffers.  We don't wait on that
 * I/O in bforget() - it's more efficient to wait on the I/O only if we really
 * need to.  That happens here.
 */
void clean_bdev_aliases(struct block_device *bdev, sector_t block, sector_t len)
{
	struct address_space *bd_mapping = bdev->bd_mapping;
	const int blkbits = bd_mapping->host->i_blkbits;
	struct folio_batch fbatch;
	pgoff_t index = ((loff_t)block << blkbits) / PAGE_SIZE;
	pgoff_t end;
	int i, count;
	struct buffer_head *bh;
	struct buffer_head *head;

	end = ((loff_t)(block + len - 1) << blkbits) / PAGE_SIZE;
	folio_batch_init(&fbatch);
	while (filemap_get_folios(bd_mapping, &index, end, &fbatch)) {
		count = folio_batch_count(&fbatch);
		for (i = 0; i < count; i++) {
			struct folio *folio = fbatch.folios[i];

			if (!folio_buffers(folio))
				continue;
			/*
			 * We use folio lock instead of bd_mapping->i_private_lock
			 * to pin buffers here since we can afford to sleep and
			 * it scales better than a global spinlock lock.
			 */
			folio_lock(folio);
			/* Recheck when the folio is locked which pins bhs */
			head = folio_buffers(folio);
			if (!head)
				goto unlock_page;
			bh = head;
			do {
				if (!buffer_mapped(bh) || (bh->b_blocknr < block))
					goto next;
				if (bh->b_blocknr >= block + len)
					break;
				clear_buffer_dirty(bh);
				wait_on_buffer(bh);
				clear_buffer_req(bh);
next:
				bh = bh->b_this_page;
			} while (bh != head);
unlock_page:
			folio_unlock(folio);
		}
		folio_batch_release(&fbatch);
		cond_resched();
		/* End of range already reached? */
		if (index > end || !index)
			break;
	}
}
EXPORT_SYMBOL(clean_bdev_aliases);

static struct buffer_head *folio_create_buffers(struct folio *folio,
						struct inode *inode,
						unsigned int b_state)
{
	struct buffer_head *bh;

	BUG_ON(!folio_test_locked(folio));

	bh = folio_buffers(folio);
	if (!bh)
		bh = create_empty_buffers(folio,
				1 << READ_ONCE(inode->i_blkbits), b_state);
	return bh;
}

/*
 * NOTE! All mapped/uptodate combinations are valid:
 *
 *	Mapped	Uptodate	Meaning
 *
 *	No	No		"unknown" - must do get_block()
 *	No	Yes		"hole" - zero-filled
 *	Yes	No		"allocated" - allocated on disk, not read in
 *	Yes	Yes		"valid" - allocated and up-to-date in memory.
 *
 * "Dirty" is valid only with the last case (mapped+uptodate).
 */

/*
 * While block_write_full_folio is writing back the dirty buffers under
 * the folio lock, whoever dirtied the buffers may decide to clean them
 * again at any time.  We handle that by only looking at the buffer
 * state inside lock_buffer().
 *
 * If block_write_full_folio() is called for regular writeback
 * (wbc->sync_mode == WB_SYNC_NONE) then it will redirty a folio which
 * has a locked buffer.   This only can happen if someone has written
 * the buffer directly, with bh_submit().  At the address_space level
 * the folio writeback flag prevents this contention from occurring.
 *
 * If block_write_full_folio() is called with wbc->sync_mode ==
 * WB_SYNC_ALL, the writes are posted using REQ_SYNC; this
 * causes the writes to be flagged as synchronous writes.
 */
int __block_write_full_folio(struct inode *inode, struct folio *folio,
			get_block_t *get_block, struct writeback_control *wbc)
{
	int err;
	sector_t block;
	sector_t last_block;
	struct buffer_head *bh, *head;
	size_t blocksize;
	int nr_underway = 0;
	blk_opf_t write_flags = wbc_to_write_flags(wbc);

	head = folio_create_buffers(folio, inode,
				    (1 << BH_Dirty) | (1 << BH_Uptodate));

	/*
	 * Be very careful.  We have no exclusion from block_dirty_folio
	 * here, and the (potentially unmapped) buffers may become dirty at
	 * any time.  If a buffer becomes dirty here after we've inspected it
	 * then we just miss that fact, and the folio stays dirty.
	 *
	 * Buffers outside i_size may be dirtied by block_dirty_folio;
	 * handle that here by just cleaning them.
	 */

	bh = head;
	blocksize = bh->b_size;

	block = div_u64(folio_pos(folio), blocksize);
	last_block = div_u64(i_size_read(inode) - 1, blocksize);

	/*
	 * Get all the dirty buffers mapped to disk addresses and
	 * handle any aliases from the underlying blockdev's mapping.
	 */
	do {
		if (block > last_block) {
			/*
			 * mapped buffers outside i_size will occur, because
			 * this folio can be outside i_size when there is a
			 * truncate in progress.
			 */
			/*
			 * The buffer was zeroed by block_write_full_folio()
			 */
			clear_buffer_dirty(bh);
			set_buffer_uptodate(bh);
		} else if ((!buffer_mapped(bh) || buffer_delay(bh)) &&
			   buffer_dirty(bh)) {
			WARN_ON(bh->b_size != blocksize);
			err = get_block(inode, block, bh, 1);
			if (err)
				goto recover;
			clear_buffer_delay(bh);
			if (buffer_new(bh)) {
				/* blockdev mappings never come here */
				clear_buffer_new(bh);
				clean_bdev_bh_alias(bh);
			}
		}
		bh = bh->b_this_page;
		block++;
	} while (bh != head);

	do {
		if (!buffer_mapped(bh))
			continue;
		/*
		 * If it's a fully non-blocking write attempt and we cannot
		 * lock the buffer then redirty the folio.  Note that this can
		 * potentially cause a busy-wait loop from writeback threads
		 * and kswapd activity, but those code paths have their own
		 * higher-level throttling.
		 */
		if (wbc->sync_mode != WB_SYNC_NONE) {
			lock_buffer(bh);
		} else if (!trylock_buffer(bh)) {
			folio_redirty_for_writepage(wbc, folio);
			continue;
		}
		if (test_clear_buffer_dirty(bh)) {
			set_buffer_async_write(bh);
		} else {
			unlock_buffer(bh);
		}
	} while ((bh = bh->b_this_page) != head);

	/*
	 * The folio and its buffers are protected by the writeback flag,
	 * so we can drop the bh refcounts early.
	 */
	BUG_ON(folio_test_writeback(folio));
	folio_start_writeback(folio);

	do {
		struct buffer_head *next = bh->b_this_page;
		if (buffer_async_write(bh)) {
			__bh_submit(bh, REQ_OP_WRITE | write_flags,
					inode->i_write_hint, wbc,
					bh_end_async_write);
			nr_underway++;
		}
		bh = next;
	} while (bh != head);
	folio_unlock(folio);

	err = 0;
done:
	if (nr_underway == 0) {
		/*
		 * The folio was marked dirty, but the buffers were
		 * clean.  Someone wrote them back by hand with
		 * write_dirty_buffer/bh_submit.  A rare case.
		 */
		folio_end_writeback(folio);

		/*
		 * The folio and buffer_heads can be released at any time from
		 * here on.
		 */
	}
	return err;

recover:
	/*
	 * ENOSPC, or some other error.  We may already have added some
	 * blocks to the file, so we need to write these out to avoid
	 * exposing stale data.
	 * The folio is currently locked and not marked for writeback
	 */
	bh = head;
	/* Recovery: lock and submit the mapped buffers */
	do {
		if (buffer_mapped(bh) && buffer_dirty(bh) &&
		    !buffer_delay(bh)) {
			lock_buffer(bh);
			set_buffer_async_write(bh);
		} else {
			/*
			 * The buffer may have been set dirty during
			 * attachment to a dirty folio.
			 */
			clear_buffer_dirty(bh);
		}
	} while ((bh = bh->b_this_page) != head);
	BUG_ON(folio_test_writeback(folio));
	mapping_set_error(folio->mapping, err);
	folio_start_writeback(folio);
	do {
		struct buffer_head *next = bh->b_this_page;
		if (buffer_async_write(bh)) {
			clear_buffer_dirty(bh);
			__bh_submit(bh, REQ_OP_WRITE | write_flags,
					inode->i_write_hint, wbc,
					bh_end_async_write);
			nr_underway++;
		}
		bh = next;
	} while (bh != head);
	folio_unlock(folio);
	goto done;
}
EXPORT_SYMBOL(__block_write_full_folio);

/*
 * If a folio has any new buffers, zero them out here, and mark them uptodate
 * and dirty so they'll be written out (in order to prevent uninitialised
 * block data from leaking). And clear the new bit.
 */
void folio_zero_new_buffers(struct folio *folio, size_t from, size_t to)
{
	size_t block_start, block_end;
	struct buffer_head *head, *bh;

	BUG_ON(!folio_test_locked(folio));
	head = folio_buffers(folio);
	if (!head)
		return;

	bh = head;
	block_start = 0;
	do {
		block_end = block_start + bh->b_size;

		if (buffer_new(bh)) {
			if (block_end > from && block_start < to) {
				if (!folio_test_uptodate(folio)) {
					size_t start, xend;

					start = max(from, block_start);
					xend = min(to, block_end);

					folio_zero_segment(folio, start, xend);
					set_buffer_uptodate(bh);
				}

				clear_buffer_new(bh);
				mark_buffer_dirty(bh);
			}
		}

		block_start = block_end;
		bh = bh->b_this_page;
	} while (bh != head);
}
EXPORT_SYMBOL(folio_zero_new_buffers);

static int
iomap_to_bh(struct inode *inode, sector_t block, struct buffer_head *bh,
		const struct iomap *iomap)
{
	loff_t offset = (loff_t)block << inode->i_blkbits;

	bh->b_bdev = iomap->bdev;

	/*
	 * Block points to offset in file we need to map, iomap contains
	 * the offset at which the map starts. If the map ends before the
	 * current block, then do not map the buffer and let the caller
	 * handle it.
	 */
	if (offset >= iomap->offset + iomap->length)
		return -EIO;

	switch (iomap->type) {
	case IOMAP_HOLE:
		/*
		 * If the buffer is not up to date or beyond the current EOF,
		 * we need to mark it as new to ensure sub-block zeroing is
		 * executed if necessary.
		 */
		if (!buffer_uptodate(bh) ||
		    (offset >= i_size_read(inode)))
			set_buffer_new(bh);
		return 0;
	case IOMAP_DELALLOC:
		if (!buffer_uptodate(bh) ||
		    (offset >= i_size_read(inode)))
			set_buffer_new(bh);
		set_buffer_uptodate(bh);
		set_buffer_mapped(bh);
		set_buffer_delay(bh);
		return 0;
	case IOMAP_UNWRITTEN:
		/*
		 * For unwritten regions, we always need to ensure that regions
		 * in the block we are not writing to are zeroed. Mark the
		 * buffer as new to ensure this.
		 */
		set_buffer_new(bh);
		set_buffer_unwritten(bh);
		fallthrough;
	case IOMAP_MAPPED:
		if ((iomap->flags & IOMAP_F_NEW) ||
		    offset >= i_size_read(inode)) {
			/*
			 * This can happen if truncating the block device races
			 * with the check in the caller as i_size updates on
			 * block devices aren't synchronized by i_rwsem for
			 * block devices.
			 */
			if (S_ISBLK(inode->i_mode))
				return -EIO;
			set_buffer_new(bh);
		}
		bh->b_blocknr = (iomap->addr + offset - iomap->offset) >>
				inode->i_blkbits;
		set_buffer_mapped(bh);
		return 0;
	default:
		WARN_ON_ONCE(1);
		return -EIO;
	}
}

int __block_write_begin_int(struct folio *folio, loff_t pos, unsigned len,
		get_block_t *get_block, const struct iomap *iomap)
{
	size_t from = offset_in_folio(folio, pos);
	size_t to = from + len;
	struct inode *inode = folio->mapping->host;
	size_t block_start, block_end;
	sector_t block;
	int err = 0;
	size_t blocksize;
	struct buffer_head *bh, *head, *wait[2], **wait_bh=wait;

	BUG_ON(!folio_test_locked(folio));
	BUG_ON(to > folio_size(folio));
	BUG_ON(from > to);

	head = folio_create_buffers(folio, inode, 0);
	blocksize = head->b_size;
	block = div_u64(folio_pos(folio), blocksize);

	for (bh = head, block_start = 0; bh != head || !block_start;
	    block++, block_start=block_end, bh = bh->b_this_page) {
		block_end = block_start + blocksize;
		if (block_end <= from || block_start >= to) {
			if (folio_test_uptodate(folio)) {
				if (!buffer_uptodate(bh))
					set_buffer_uptodate(bh);
			}
			continue;
		}
		if (buffer_new(bh))
			clear_buffer_new(bh);
		if (!buffer_mapped(bh)) {
			WARN_ON(bh->b_size != blocksize);
			if (get_block)
				err = get_block(inode, block, bh, 1);
			else
				err = iomap_to_bh(inode, block, bh, iomap);
			if (err)
				break;

			if (buffer_new(bh)) {
				clean_bdev_bh_alias(bh);
				if (folio_test_uptodate(folio)) {
					clear_buffer_new(bh);
					set_buffer_uptodate(bh);
					mark_buffer_dirty(bh);
					continue;
				}
				if (block_end > to || block_start < from)
					folio_zero_segments(folio,
						to, block_end,
						block_start, from);
				continue;
			}
		}
		if (folio_test_uptodate(folio)) {
			if (!buffer_uptodate(bh))
				set_buffer_uptodate(bh);
			continue;
		}
		if (!buffer_uptodate(bh) && !buffer_delay(bh) &&
		    !buffer_unwritten(bh) &&
		     (block_start < from || block_end > to)) {
			bh_read_nowait(bh, 0);
			*wait_bh++=bh;
		}
	}
	/*
	 * If we issued read requests - let them complete.
	 */
	while(wait_bh > wait) {
		wait_on_buffer(*--wait_bh);
		if (!buffer_uptodate(*wait_bh))
			err = -EIO;
	}
	if (unlikely(err))
		folio_zero_new_buffers(folio, from, to);
	return err;
}

int __block_write_begin(struct folio *folio, loff_t pos, unsigned len,
		get_block_t *get_block)
{
	return __block_write_begin_int(folio, pos, len, get_block, NULL);
}
EXPORT_SYMBOL(__block_write_begin);

void block_commit_write(struct folio *folio, size_t from, size_t to)
{
	size_t block_start, block_end;
	bool partial = false;
	unsigned blocksize;
	struct buffer_head *bh, *head;

	bh = head = folio_buffers(folio);
	if (!bh)
		return;
	blocksize = bh->b_size;

	block_start = 0;
	do {
		block_end = block_start + blocksize;
		if (block_end <= from || block_start >= to) {
			if (!buffer_uptodate(bh))
				partial = true;
		} else {
			set_buffer_uptodate(bh);
			mark_buffer_dirty(bh);
		}
		if (buffer_new(bh))
			clear_buffer_new(bh);

		block_start = block_end;
		bh = bh->b_this_page;
	} while (bh != head);

	/*
	 * If this is a partial write which happened to make all buffers
	 * uptodate then we can optimize away a bogus read_folio() for
	 * the next read(). Here we 'discover' whether the folio went
	 * uptodate as a result of this (potentially partial) write.
	 */
	if (!partial)
		folio_mark_uptodate(folio);
}
EXPORT_SYMBOL(block_commit_write);

/*
 * block_write_begin takes care of the basic task of block allocation and
 * bringing partial write blocks uptodate first.
 *
 * The filesystem needs to handle block truncation upon failure.
 */
int block_write_begin(struct address_space *mapping, loff_t pos, unsigned len,
		struct folio **foliop, get_block_t *get_block)
{
	pgoff_t index = pos >> PAGE_SHIFT;
	struct folio *folio;
	int status;

	folio = __filemap_get_folio(mapping, index, FGP_WRITEBEGIN,
			mapping_gfp_mask(mapping));
	if (IS_ERR(folio))
		return PTR_ERR(folio);

	status = __block_write_begin_int(folio, pos, len, get_block, NULL);
	if (unlikely(status)) {
		folio_unlock(folio);
		folio_put(folio);
		folio = NULL;
	}

	*foliop = folio;
	return status;
}
EXPORT_SYMBOL(block_write_begin);

int block_write_end(loff_t pos, unsigned len, unsigned copied,
		struct folio *folio)
{
	size_t start = pos - folio_pos(folio);

	if (unlikely(copied < len)) {
		/*
		 * The buffers that were written will now be uptodate, so
		 * we don't have to worry about a read_folio reading them
		 * and overwriting a partial write. However if we have
		 * encountered a short write and only partially written
		 * into a buffer, it will not be marked uptodate, so a
		 * read_folio might come in and destroy our partial write.
		 *
		 * Do the simplest thing, and just treat any short write to a
		 * non uptodate folio as a zero-length write, and force the
		 * caller to redo the whole thing.
		 */
		if (!folio_test_uptodate(folio))
			copied = 0;

		folio_zero_new_buffers(folio, start+copied, start+len);
	}
	flush_dcache_folio(folio);

	/* This could be a short (even 0-length) commit */
	block_commit_write(folio, start, start + copied);

	return copied;
}
EXPORT_SYMBOL(block_write_end);

int generic_write_end(const struct kiocb *iocb, struct address_space *mapping,
		      loff_t pos, unsigned len, unsigned copied,
		      struct folio *folio, void *fsdata)
{
	struct inode *inode = mapping->host;
	loff_t old_size = inode->i_size;
	bool i_size_changed = false;

	copied = block_write_end(pos, len, copied, folio);

	/*
	 * No need to use i_size_read() here, the i_size cannot change under us
	 * because we hold i_rwsem.
	 *
	 * But it's important to update i_size while still holding folio lock:
	 * page writeout could otherwise come in and zero beyond i_size.
	 */
	if (pos + copied > inode->i_size) {
		i_size_write(inode, pos + copied);
		i_size_changed = true;
	}

	folio_unlock(folio);
	folio_put(folio);

	if (old_size < pos)
		pagecache_isize_extended(inode, old_size, pos);
	/*
	 * Don't mark the inode dirty under page lock. First, it unnecessarily
	 * makes the holding time of page lock longer. Second, it forces lock
	 * ordering of page lock and transaction start for journaling
	 * filesystems.
	 */
	if (i_size_changed)
		mark_inode_dirty(inode);
	return copied;
}
EXPORT_SYMBOL(generic_write_end);

/*
 * block_is_partially_uptodate checks whether buffers within a folio are
 * uptodate or not.
 *
 * Returns true if all buffers which correspond to the specified part
 * of the folio are uptodate.
 */
bool block_is_partially_uptodate(struct folio *folio, size_t from, size_t count)
{
	unsigned block_start, block_end, blocksize;
	unsigned to;
	struct buffer_head *bh, *head;
	bool ret = true;

	head = folio_buffers(folio);
	if (!head)
		return false;
	blocksize = head->b_size;
	to = min(folio_size(folio) - from, count);
	to = from + to;
	if (from < blocksize && to > folio_size(folio) - blocksize)
		return false;

	bh = head;
	block_start = 0;
	do {
		block_end = block_start + blocksize;
		if (block_end > from && block_start < to) {
			if (!buffer_uptodate(bh)) {
				ret = false;
				break;
			}
			if (block_end >= to)
				break;
		}
		block_start = block_end;
		bh = bh->b_this_page;
	} while (bh != head);

	return ret;
}
EXPORT_SYMBOL(block_is_partially_uptodate);

/*
 * Generic "read_folio" function for block devices that have the normal
 * get_block functionality. This is most of the block device filesystems.
 * Reads the folio asynchronously --- the unlock_buffer() and
 * set/clear_buffer_uptodate() functions propagate buffer state into the
 * folio once IO has completed.
 */
int block_read_full_folio(struct folio *folio, get_block_t *get_block)
{
	struct inode *inode = folio->mapping->host;
	sector_t iblock, lblock;
	struct buffer_head *bh, *head, *prev = NULL;
	size_t blocksize;
	int fully_mapped = 1;
	bool page_error = false;
	loff_t limit = i_size_read(inode);

	/* This is needed for ext4. */
	if (IS_ENABLED(CONFIG_FS_VERITY) && IS_VERITY(inode))
		limit = inode->i_sb->s_maxbytes;

	head = folio_create_buffers(folio, inode, 0);
	blocksize = head->b_size;

	iblock = div_u64(folio_pos(folio), blocksize);
	lblock = div_u64(limit + blocksize - 1, blocksize);
	bh = head;

	do {
		if (buffer_uptodate(bh))
			continue;

		if (!buffer_mapped(bh)) {
			int err = 0;

			fully_mapped = 0;
			if (iblock < lblock) {
				WARN_ON(bh->b_size != blocksize);
				err = get_block(inode, iblock, bh, 0);
				if (err)
					page_error = true;
			}
			if (!buffer_mapped(bh)) {
				folio_zero_range(folio, bh_offset(bh),
						blocksize);
				if (!err)
					set_buffer_uptodate(bh);
				continue;
			}
			/*
			 * get_block() might have updated the buffer
			 * synchronously
			 */
			if (buffer_uptodate(bh))
				continue;
		}

		lock_buffer(bh);
		if (buffer_uptodate(bh)) {
			unlock_buffer(bh);
			continue;
		}

		/*
		 * If a folio's buffers are under async readin
		 * (end_buffer_async_read completion) then there is a
		 * possibility that another thread of control could lock
		 * one of the buffers after it has completed but while
		 * some of the other buffers have not completed.  This
		 * locked buffer would confuse end_buffer_async_read()
		 * into not unlocking the folio.  So the absence of
		 * BH_Async_Read tells end_buffer_async_read() that this
		 * buffer is not under async I/O.
		 *
		 * The folio comes unlocked when it has no locked
		 * buffer_async buffers left.
		 *
		 * The folio lock prevents anyone starting new async
		 * I/O reads into any of the buffers.
		 *
		 * The writeback flag is used to prevent simultaneous
		 * writeout of the same folio.
		 *
		 * The folio lock prevents anyone from starting writeback
		 * of a folio which is under read I/O (the writeback
		 * flag is only ever set on a locked folio).
		 */
		set_buffer_async_read(bh);
		if (prev)
			bh_submit(prev, REQ_OP_READ, bh_end_async_read);
		prev = bh;
	} while (iblock++, (bh = bh->b_this_page) != head);

	if (fully_mapped)
		folio_set_mappedtodisk(folio);

	/*
	 * All buffers are uptodate or get_block() returned an error
	 * when trying to map them - we must finish the read because
	 * end_buffer_async_read() will never be called on any buffer
	 * in this folio.
	 */
	if (prev)
		bh_submit(prev, REQ_OP_READ, bh_end_async_read);
	else
		folio_end_read(folio, !page_error);

	return 0;
}
EXPORT_SYMBOL(block_read_full_folio);

/* utility function for filesystems that need to do work on expanding
 * truncates.  Uses filesystem pagecache writes to allow the filesystem to
 * deal with the hole.
 */
int generic_cont_expand_simple(struct inode *inode, loff_t size)
{
	struct address_space *mapping = inode->i_mapping;
	const struct address_space_operations *aops = mapping->a_ops;
	struct folio *folio;
	void *fsdata = NULL;
	int err;

	err = inode_newsize_ok(inode, size);
	if (err)
		goto out;

	err = aops->write_begin(NULL, mapping, size, 0, &folio, &fsdata);
	if (err)
		goto out;

	err = aops->write_end(NULL, mapping, size, 0, 0, folio, fsdata);
	BUG_ON(err > 0);

out:
	return err;
}
EXPORT_SYMBOL(generic_cont_expand_simple);

static int cont_expand_zero(const struct kiocb *iocb,
			    struct address_space *mapping,
			    loff_t pos, loff_t *bytes)
{
	struct inode *inode = mapping->host;
	const struct address_space_operations *aops = mapping->a_ops;
	unsigned int blocksize = i_blocksize(inode);
	struct folio *folio;
	void *fsdata = NULL;
	pgoff_t index, curidx;
	loff_t curpos;
	unsigned zerofrom, offset, len;
	int err = 0;

	index = pos >> PAGE_SHIFT;
	offset = pos & ~PAGE_MASK;

	while (index > (curidx = (curpos = *bytes)>>PAGE_SHIFT)) {
		zerofrom = curpos & ~PAGE_MASK;
		if (zerofrom & (blocksize-1)) {
			*bytes |= (blocksize-1);
			(*bytes)++;
		}
		len = PAGE_SIZE - zerofrom;

		err = aops->write_begin(iocb, mapping, curpos, len,
					    &folio, &fsdata);
		if (err)
			goto out;
		folio_zero_range(folio, offset_in_folio(folio, curpos), len);
		err = aops->write_end(iocb, mapping, curpos, len, len,
						folio, fsdata);
		if (err < 0)
			goto out;
		BUG_ON(err != len);
		err = 0;

		balance_dirty_pages_ratelimited(mapping);

		if (fatal_signal_pending(current)) {
			err = -EINTR;
			goto out;
		}
	}

	/* page covers the boundary, find the boundary offset */
	if (index == curidx) {
		zerofrom = curpos & ~PAGE_MASK;
		/* if we will expand the thing last block will be filled */
		if (offset <= zerofrom) {
			goto out;
		}
		if (zerofrom & (blocksize-1)) {
			*bytes |= (blocksize-1);
			(*bytes)++;
		}
		len = offset - zerofrom;

		err = aops->write_begin(iocb, mapping, curpos, len,
					    &folio, &fsdata);
		if (err)
			goto out;
		folio_zero_range(folio, offset_in_folio(folio, curpos), len);
		err = aops->write_end(iocb, mapping, curpos, len, len,
						folio, fsdata);
		if (err < 0)
			goto out;
		BUG_ON(err != len);
		err = 0;
	}
out:
	return err;
}

/*
 * For moronic filesystems that do not allow holes in file.
 * We may have to extend the file.
 */
int cont_write_begin(const struct kiocb *iocb, struct address_space *mapping,
		     loff_t pos, unsigned len, struct folio **foliop,
		     void **fsdata, get_block_t *get_block, loff_t *bytes)
{
	struct inode *inode = mapping->host;
	unsigned int blocksize = i_blocksize(inode);
	unsigned int zerofrom;
	int err;

	err = cont_expand_zero(iocb, mapping, pos, bytes);
	if (err)
		return err;

	zerofrom = *bytes & ~PAGE_MASK;
	if (pos+len > *bytes && zerofrom & (blocksize-1)) {
		*bytes |= (blocksize-1);
		(*bytes)++;
	}

	return block_write_begin(mapping, pos, len, foliop, get_block);
}
EXPORT_SYMBOL(cont_write_begin);

/*
 * block_page_mkwrite() is not allowed to change the file size as it gets
 * called from a page fault handler when a page is first dirtied. Hence we must
 * be careful to check for EOF conditions here. We set the page up correctly
 * for a written page which means we get ENOSPC checking when writing into
 * holes and correct delalloc and unwritten extent mapping on filesystems that
 * support these features.
 *
 * We are not allowed to take the i_rwsem here so we have to play games to
 * protect against truncate races as the page could now be beyond EOF.  Because
 * truncate writes the inode size before removing pages, once we have the
 * page lock we can determine safely if the page is beyond EOF. If it is not
 * beyond EOF, then the page is guaranteed safe against truncation until we
 * unlock the page.
 *
 * Direct callers of this function should protect against filesystem freezing
 * using sb_start_pagefault() - sb_end_pagefault() functions.
 */
int block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf,
			 get_block_t get_block)
{
	struct folio *folio = page_folio(vmf->page);
	struct inode *inode = file_inode(vma->vm_file);
	unsigned long end;
	loff_t size;
	int ret;

	folio_lock(folio);
	size = i_size_read(inode);
	if ((folio->mapping != inode->i_mapping) ||
	    (folio_pos(folio) >= size)) {
		/* We overload EFAULT to mean page got truncated */
		ret = -EFAULT;
		goto out_unlock;
	}

	end = folio_size(folio);
	/* folio is wholly or partially inside EOF */
	if (folio_pos(folio) + end > size)
		end = size - folio_pos(folio);

	ret = __block_write_begin_int(folio, 0, end, get_block, NULL);
	if (unlikely(ret))
		goto out_unlock;

	block_commit_write(folio, 0, end);

	folio_mark_dirty(folio);
	folio_wait_stable(folio);
	return 0;
out_unlock:
	folio_unlock(folio);
	return ret;
}
EXPORT_SYMBOL(block_page_mkwrite);

int block_truncate_page(struct address_space *mapping,
			loff_t from, get_block_t *get_block)
{
	pgoff_t index = from >> PAGE_SHIFT;
	unsigned blocksize;
	sector_t iblock;
	size_t offset, length, pos;
	struct inode *inode = mapping->host;
	struct folio *folio;
	struct buffer_head *bh;
	int err = 0;

	blocksize = i_blocksize(inode);
	length = from & (blocksize - 1);

	/* Block boundary? Nothing to do */
	if (!length)
		return 0;

	length = blocksize - length;
	iblock = ((loff_t)index * PAGE_SIZE) >> inode->i_blkbits;

	folio = filemap_grab_folio(mapping, index);
	if (IS_ERR(folio))
		return PTR_ERR(folio);

	bh = folio_buffers(folio);
	if (!bh)
		bh = create_empty_buffers(folio, blocksize, 0);

	/* Find the buffer that contains "offset" */
	offset = offset_in_folio(folio, from);
	pos = blocksize;
	while (offset >= pos) {
		bh = bh->b_this_page;
		iblock++;
		pos += blocksize;
	}

	if (!buffer_mapped(bh)) {
		WARN_ON(bh->b_size != blocksize);
		err = get_block(inode, iblock, bh, 0);
		if (err)
			goto unlock;
		/* unmapped? It's a hole - nothing to do */
		if (!buffer_mapped(bh))
			goto unlock;
	}

	/* Ok, it's mapped. Make sure it's up-to-date */
	if (folio_test_uptodate(folio))
		set_buffer_uptodate(bh);

	if (!buffer_uptodate(bh) && !buffer_delay(bh) && !buffer_unwritten(bh)) {
		err = bh_read(bh, 0);
		/* Uhhuh. Read error. Complain and punt. */
		if (err < 0)
			goto unlock;
	}

	folio_zero_range(folio, offset, length);
	mark_buffer_dirty(bh);

unlock:
	folio_unlock(folio);
	folio_put(folio);

	return err;
}
EXPORT_SYMBOL(block_truncate_page);

/*
 * The generic write folio function for buffer-backed address_spaces
 */
int block_write_full_folio(struct folio *folio, struct writeback_control *wbc,
		void *get_block)
{
	struct inode * const inode = folio->mapping->host;
	loff_t i_size = i_size_read(inode);

	/* Is the folio fully inside i_size? */
	if (folio_next_pos(folio) <= i_size)
		return __block_write_full_folio(inode, folio, get_block, wbc);

	/* Is the folio fully outside i_size? (truncate in progress) */
	if (folio_pos(folio) >= i_size) {
		folio_unlock(folio);
		return 0; /* don't care */
	}

	/*
	 * The folio straddles i_size.  It must be zeroed out on each and every
	 * writeback invocation because it may be mmapped.  "A file is mapped
	 * in multiples of the page size.  For a file that is not a multiple of
	 * the page size, the remaining memory is zeroed when mapped, and
	 * writes to that region are not written out to the file."
	 */
	folio_zero_segment(folio, offset_in_folio(folio, i_size),
			folio_size(folio));
	return __block_write_full_folio(inode, folio, get_block, wbc);
}

sector_t generic_block_bmap(struct address_space *mapping, sector_t block,
			    get_block_t *get_block)
{
	struct inode *inode = mapping->host;
	struct buffer_head tmp = {
		.b_size = i_blocksize(inode),
	};

	get_block(inode, block, &tmp, 0);
	return tmp.b_blocknr;
}
EXPORT_SYMBOL(generic_block_bmap);

void write_dirty_buffer(struct buffer_head *bh, blk_opf_t op_flags)
{
	lock_buffer(bh);
	if (!test_clear_buffer_dirty(bh)) {
		unlock_buffer(bh);
		return;
	}
	bh_submit(bh, REQ_OP_WRITE | op_flags, bh_end_write);
}
EXPORT_SYMBOL(write_dirty_buffer);

/*
 * For a data-integrity writeout, we need to wait upon any in-progress I/O
 * and then start new I/O and then wait upon it.  The caller must have a ref on
 * the buffer_head.
 */
int __sync_dirty_buffer(struct buffer_head *bh, blk_opf_t op_flags)
{
	WARN_ON(atomic_read(&bh->b_count) < 1);
	lock_buffer(bh);
	if (test_clear_buffer_dirty(bh)) {
		/*
		 * The bh should be mapped, but it might not be if the
		 * device was hot-removed. Not much we can do but fail the I/O.
		 */
		if (!buffer_mapped(bh)) {
			unlock_buffer(bh);
			return -EIO;
		}

		bh_submit(bh, REQ_OP_WRITE | op_flags, bh_end_write);
		wait_on_buffer(bh);
		if (!buffer_uptodate(bh))
			return -EIO;
	} else {
		unlock_buffer(bh);
	}
	return 0;
}
EXPORT_SYMBOL(__sync_dirty_buffer);

int sync_dirty_buffer(struct buffer_head *bh)
{
	return __sync_dirty_buffer(bh, REQ_SYNC);
}
EXPORT_SYMBOL(sync_dirty_buffer);

static inline int buffer_busy(struct buffer_head *bh)
{
	return atomic_read(&bh->b_count) |
		(bh->b_state & ((1 << BH_Dirty) | (1 << BH_Lock)));
}

static bool
drop_buffers(struct folio *folio, struct buffer_head **buffers_to_free)
{
	struct buffer_head *head = folio_buffers(folio);
	struct buffer_head *bh;

	bh = head;
	do {
		if (buffer_busy(bh))
			goto failed;
		bh = bh->b_this_page;
	} while (bh != head);

	do {
		struct buffer_head *next = bh->b_this_page;

		remove_assoc_queue(bh);
		bh = next;
	} while (bh != head);
	*buffers_to_free = head;
	folio_detach_private(folio);
	return true;
failed:
	return false;
}

/**
 * try_to_free_buffers - Release buffers attached to this folio.
 * @folio: The folio.
 *
 * If any buffers are in use (dirty, under writeback, elevated refcount),
 * no buffers will be freed.
 *
 * If the folio is dirty but all the buffers are clean then we need to
 * be sure to mark the folio clean as well.  This is because the folio
 * may be against a block device, and a later reattachment of buffers
 * to a dirty folio will set *all* buffers dirty.  Which would corrupt
 * filesystem data on the same device.
 *
 * The same applies to regular filesystem folios: if all the buffers are
 * clean then we set the folio clean and proceed.  To do that, we require
 * total exclusion from block_dirty_folio().  That is obtained with
 * i_private_lock.
 *
 * Exclusion against try_to_free_buffers may be obtained by either
 * locking the folio or by holding its mapping's i_private_lock.
 *
 * Context: Process context.  @folio must be locked.  Will not sleep.
 * Return: true if all buffers attached to this folio were freed.
 */
bool try_to_free_buffers(struct folio *folio)
{
	struct address_space * const mapping = folio->mapping;
	struct buffer_head *buffers_to_free = NULL;
	bool ret = 0;

	BUG_ON(!folio_test_locked(folio));
	if (folio_test_writeback(folio))
		return false;

	/* Misconfigured folio check */
	if (WARN_ON_ONCE(!folio_buffers(folio)))
		return true;

	if (mapping == NULL) {		/* can this still happen? */
		ret = drop_buffers(folio, &buffers_to_free);
		goto out;
	}

	spin_lock(&mapping->i_private_lock);
	ret = drop_buffers(folio, &buffers_to_free);

	/*
	 * If the filesystem writes its buffers by hand (eg ext3)
	 * then we can have clean buffers against a dirty folio.  We
	 * clean the folio here; otherwise the VM will never notice
	 * that the filesystem did any IO at all.
	 *
	 * Also, during truncate, discard_buffer will have marked all
	 * the folio's buffers clean.  We discover that here and clean
	 * the folio also.
	 *
	 * i_private_lock must be held over this entire operation in order
	 * to synchronise against block_dirty_folio and prevent the
	 * dirty bit from being lost.
	 */
	if (ret)
		folio_cancel_dirty(folio);
	spin_unlock(&mapping->i_private_lock);
out:
	if (buffers_to_free) {
		struct buffer_head *bh = buffers_to_free;

		do {
			struct buffer_head *next = bh->b_this_page;
			free_buffer_head(bh);
			bh = next;
		} while (bh != buffers_to_free);
	}
	return ret;
}
EXPORT_SYMBOL(try_to_free_buffers);

/*
 * Buffer-head allocation
 */
static struct kmem_cache *bh_cachep __ro_after_init;

/*
 * Once the number of bh's in the machine exceeds this level, we start
 * stripping them in writeback.
 */
static unsigned long max_buffer_heads __ro_after_init;

int buffer_heads_over_limit;

struct bh_accounting {
	int nr;			/* Number of live bh's */
	int ratelimit;		/* Limit cacheline bouncing */
};

static DEFINE_PER_CPU(struct bh_accounting, bh_accounting) = {0, 0};

static void recalc_bh_state(void)
{
	int i;
	int tot = 0;

	if (__this_cpu_inc_return(bh_accounting.ratelimit) - 1 < 4096)
		return;
	__this_cpu_write(bh_accounting.ratelimit, 0);
	for_each_online_cpu(i)
		tot += per_cpu(bh_accounting, i).nr;
	buffer_heads_over_limit = (tot > max_buffer_heads);
}

struct buffer_head *alloc_buffer_head(gfp_t gfp_flags)
{
	struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags);
	if (ret) {
		INIT_LIST_HEAD(&ret->b_assoc_buffers);
		spin_lock_init(&ret->b_uptodate_lock);
		preempt_disable();
		__this_cpu_inc(bh_accounting.nr);
		recalc_bh_state();
		preempt_enable();
	}
	return ret;
}
EXPORT_SYMBOL(alloc_buffer_head);

void free_buffer_head(struct buffer_head *bh)
{
	BUG_ON(!list_empty(&bh->b_assoc_buffers));
	kmem_cache_free(bh_cachep, bh);
	preempt_disable();
	__this_cpu_dec(bh_accounting.nr);
	recalc_bh_state();
	preempt_enable();
}
EXPORT_SYMBOL(free_buffer_head);

static int buffer_exit_cpu_dead(unsigned int cpu)
{
	int i;
	struct bh_lru *b = &per_cpu(bh_lrus, cpu);

	for (i = 0; i < BH_LRU_SIZE; i++) {
		brelse(b->bhs[i]);
		b->bhs[i] = NULL;
	}
	this_cpu_add(bh_accounting.nr, per_cpu(bh_accounting, cpu).nr);
	per_cpu(bh_accounting, cpu).nr = 0;
	return 0;
}

/**
 * bh_uptodate_or_lock - Test whether the buffer is uptodate
 * @bh: struct buffer_head
 *
 * Return true if the buffer is up-to-date and false,
 * with the buffer locked, if not.
 */
int bh_uptodate_or_lock(struct buffer_head *bh)
{
	if (!buffer_uptodate(bh)) {
		lock_buffer(bh);
		if (!buffer_uptodate(bh))
			return 0;
		unlock_buffer(bh);
	}
	return 1;
}
EXPORT_SYMBOL(bh_uptodate_or_lock);

/**
 * __bh_read - Submit read for a locked buffer
 * @bh: struct buffer_head
 * @op_flags: appending REQ_OP_* flags besides REQ_OP_READ
 * @wait: wait until reading finish
 *
 * Returns zero on success or don't wait, and -EIO on error.
 */
int __bh_read(struct buffer_head *bh, blk_opf_t op_flags, bool wait)
{
	int ret = 0;

	BUG_ON(!buffer_locked(bh));

	bh_submit(bh, REQ_OP_READ | op_flags, bh_end_read);
	if (wait) {
		wait_on_buffer(bh);
		if (!buffer_uptodate(bh))
			ret = -EIO;
	}
	return ret;
}
EXPORT_SYMBOL(__bh_read);

/**
 * __bh_read_batch - Submit read for a batch of unlocked buffers
 * @nr: entry number of the buffer batch
 * @bhs: a batch of struct buffer_head
 * @op_flags: appending REQ_OP_* flags besides REQ_OP_READ
 * @force_lock: force to get a lock on the buffer if set, otherwise drops any
 *              buffer that cannot lock.
 *
 * Returns zero on success or don't wait, and -EIO on error.
 */
void __bh_read_batch(int nr, struct buffer_head *bhs[],
		     blk_opf_t op_flags, bool force_lock)
{
	int i;

	for (i = 0; i < nr; i++) {
		struct buffer_head *bh = bhs[i];

		if (buffer_uptodate(bh))
			continue;

		if (force_lock)
			lock_buffer(bh);
		else
			if (!trylock_buffer(bh))
				continue;

		if (buffer_uptodate(bh)) {
			unlock_buffer(bh);
			continue;
		}

		bh_submit(bh, REQ_OP_READ | op_flags, bh_end_read);
	}
}
EXPORT_SYMBOL(__bh_read_batch);

void __init buffer_init(void)
{
	unsigned long nrpages;
	int ret;

	bh_cachep = KMEM_CACHE(buffer_head,
				SLAB_RECLAIM_ACCOUNT|SLAB_PANIC);
	/*
	 * Limit the bh occupancy to 10% of ZONE_NORMAL
	 */
	nrpages = (nr_free_buffer_pages() * 10) / 100;
	max_buffer_heads = nrpages * (PAGE_SIZE / sizeof(struct buffer_head));
	ret = cpuhp_setup_state_nocalls(CPUHP_FS_BUFF_DEAD, "fs/buffer:dead",
					NULL, buffer_exit_cpu_dead);
	WARN_ON(ret < 0);
}