Adding upstream version 6.6.15.upstream/6.6.15

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

21 files changed, 9067 insertions, 0 deletions

diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig
new file mode 100644
index 0000000000..f6dc961e6c
--- /dev/null
+++ b/fs/erofs/Kconfig
@@ -0,0 +1,145 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+config EROFS_FS
+	tristate "EROFS filesystem support"
+	depends on BLOCK
+	select FS_IOMAP
+	select LIBCRC32C
+	help
+	  EROFS (Enhanced Read-Only File System) is a lightweight read-only
+	  file system with modern designs (e.g. no buffer heads, inline
+	  xattrs/data, chunk-based deduplication, multiple devices, etc.) for
+	  scenarios which need high-performance read-only solutions, e.g.
+	  smartphones with Android OS, LiveCDs and high-density hosts with
+	  numerous containers;
+
+	  It also provides fixed-sized output compression support in order to
+	  improve storage density as well as keep relatively higher compression
+	  ratios and implements in-place decompression to reuse the file page
+	  for compressed data temporarily with proper strategies, which is
+	  quite useful to ensure guaranteed end-to-end runtime decompression
+	  performance under extremely memory pressure without extra cost.
+
+	  See the documentation at <file:Documentation/filesystems/erofs.rst>
+	  for more details.
+
+	  If unsure, say N.
+
+config EROFS_FS_DEBUG
+	bool "EROFS debugging feature"
+	depends on EROFS_FS
+	help
+	  Print debugging messages and enable more BUG_ONs which check
+	  filesystem consistency and find potential issues aggressively,
+	  which can be used for Android eng build, for example.
+
+	  For daily use, say N.
+
+config EROFS_FS_XATTR
+	bool "EROFS extended attributes"
+	depends on EROFS_FS
+	select XXHASH
+	default y
+	help
+	  Extended attributes are name:value pairs associated with inodes by
+	  the kernel or by users (see the attr(5) manual page, or visit
+	  <http://acl.bestbits.at/> for details).
+
+	  If unsure, say N.
+
+config EROFS_FS_POSIX_ACL
+	bool "EROFS Access Control Lists"
+	depends on EROFS_FS_XATTR
+	select FS_POSIX_ACL
+	default y
+	help
+	  Posix Access Control Lists (ACLs) support permissions for users and
+	  groups beyond the owner/group/world scheme.
+
+	  To learn more about Access Control Lists, visit the POSIX ACLs for
+	  Linux website <http://acl.bestbits.at/>.
+
+	  If you don't know what Access Control Lists are, say N.
+
+config EROFS_FS_SECURITY
+	bool "EROFS Security Labels"
+	depends on EROFS_FS_XATTR
+	default y
+	help
+	  Security labels provide an access control facility to support Linux
+	  Security Models (LSMs) accepted by AppArmor, SELinux, Smack and TOMOYO
+	  Linux. This option enables an extended attribute handler for file
+	  security labels in the erofs filesystem, so that it requires enabling
+	  the extended attribute support in advance.
+
+	  If you are not using a security module, say N.
+
+config EROFS_FS_ZIP
+	bool "EROFS Data Compression Support"
+	depends on EROFS_FS
+	select LZ4_DECOMPRESS
+	default y
+	help
+	  Enable fixed-sized output compression for EROFS.
+
+	  If you don't want to enable compression feature, say N.
+
+config EROFS_FS_ZIP_LZMA
+	bool "EROFS LZMA compressed data support"
+	depends on EROFS_FS_ZIP
+	select XZ_DEC
+	select XZ_DEC_MICROLZMA
+	help
+	  Saying Y here includes support for reading EROFS file systems
+	  containing LZMA compressed data, specifically called microLZMA. it
+	  gives better compression ratios than the LZ4 algorithm, at the
+	  expense of more CPU overhead.
+
+	  LZMA support is an experimental feature for now and so most file
+	  systems will be readable without selecting this option.
+
+	  If unsure, say N.
+
+config EROFS_FS_ZIP_DEFLATE
+	bool "EROFS DEFLATE compressed data support"
+	depends on EROFS_FS_ZIP
+	select ZLIB_INFLATE
+	help
+	  Saying Y here includes support for reading EROFS file systems
+	  containing DEFLATE compressed data.  It gives better compression
+	  ratios than the default LZ4 format, while it costs more CPU
+	  overhead.
+
+	  DEFLATE support is an experimental feature for now and so most
+	  file systems will be readable without selecting this option.
+
+	  If unsure, say N.
+
+config EROFS_FS_ONDEMAND
+	bool "EROFS fscache-based on-demand read support"
+	depends on CACHEFILES_ONDEMAND && (EROFS_FS=m && FSCACHE || EROFS_FS=y && FSCACHE=y)
+	default n
+	help
+	  This permits EROFS to use fscache-backed data blobs with on-demand
+	  read support.
+
+	  If unsure, say N.
+
+config EROFS_FS_PCPU_KTHREAD
+	bool "EROFS per-cpu decompression kthread workers"
+	depends on EROFS_FS_ZIP
+	help
+	  Saying Y here enables per-CPU kthread workers pool to carry out
+	  async decompression for low latencies on some architectures.
+
+	  If unsure, say N.
+
+config EROFS_FS_PCPU_KTHREAD_HIPRI
+	bool "EROFS high priority per-CPU kthread workers"
+	depends on EROFS_FS_ZIP && EROFS_FS_PCPU_KTHREAD
+	default y
+	help
+	  This permits EROFS to configure per-CPU kthread workers to run
+	  at higher priority.
+
+	  If unsure, say N.
diff --git a/fs/erofs/Makefile b/fs/erofs/Makefile
new file mode 100644
index 0000000000..994d0b9ded
--- /dev/null
+++ b/fs/erofs/Makefile
@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+obj-$(CONFIG_EROFS_FS) += erofs.o
+erofs-objs := super.o inode.o data.o namei.o dir.o utils.o sysfs.o
+erofs-$(CONFIG_EROFS_FS_XATTR) += xattr.o
+erofs-$(CONFIG_EROFS_FS_ZIP) += decompressor.o zmap.o zdata.o pcpubuf.o
+erofs-$(CONFIG_EROFS_FS_ZIP_LZMA) += decompressor_lzma.o
+erofs-$(CONFIG_EROFS_FS_ZIP_DEFLATE) += decompressor_deflate.o
+erofs-$(CONFIG_EROFS_FS_ONDEMAND) += fscache.o
diff --git a/fs/erofs/compress.h b/fs/erofs/compress.h
new file mode 100644
index 0000000000..279933e007
--- /dev/null
+++ b/fs/erofs/compress.h
@@ -0,0 +1,105 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2019 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ */
+#ifndef __EROFS_FS_COMPRESS_H
+#define __EROFS_FS_COMPRESS_H
+
+#include "internal.h"
+
+struct z_erofs_decompress_req {
+	struct super_block *sb;
+	struct page **in, **out;
+
+	unsigned short pageofs_in, pageofs_out;
+	unsigned int inputsize, outputsize;
+
+	/* indicate the algorithm will be used for decompression */
+	unsigned int alg;
+	bool inplace_io, partial_decoding, fillgaps;
+};
+
+struct z_erofs_decompressor {
+	int (*config)(struct super_block *sb, struct erofs_super_block *dsb,
+		      void *data, int size);
+	int (*decompress)(struct z_erofs_decompress_req *rq,
+			  struct page **pagepool);
+	char *name;
+};
+
+/* some special page->private (unsigned long, see below) */
+#define Z_EROFS_SHORTLIVED_PAGE		(-1UL << 2)
+#define Z_EROFS_PREALLOCATED_PAGE	(-2UL << 2)
+
+/*
+ * For all pages in a pcluster, page->private should be one of
+ * Type                         Last 2bits      page->private
+ * short-lived page             00              Z_EROFS_SHORTLIVED_PAGE
+ * preallocated page (tryalloc) 00              Z_EROFS_PREALLOCATED_PAGE
+ * cached/managed page          00              pointer to z_erofs_pcluster
+ * online page (file-backed,    01/10/11        sub-index << 2 | count
+ *              some pages can be used for inplace I/O)
+ *
+ * page->mapping should be one of
+ * Type                 page->mapping
+ * short-lived page     NULL
+ * preallocated page    NULL
+ * cached/managed page  non-NULL or NULL (invalidated/truncated page)
+ * online page          non-NULL
+ *
+ * For all managed pages, PG_private should be set with 1 extra refcount,
+ * which is used for page reclaim / migration.
+ */
+
+/*
+ * short-lived pages are pages directly from buddy system with specific
+ * page->private (no need to set PagePrivate since these are non-LRU /
+ * non-movable pages and bypass reclaim / migration code).
+ */
+static inline bool z_erofs_is_shortlived_page(struct page *page)
+{
+	if (page->private != Z_EROFS_SHORTLIVED_PAGE)
+		return false;
+
+	DBG_BUGON(page->mapping);
+	return true;
+}
+
+static inline bool z_erofs_put_shortlivedpage(struct page **pagepool,
+					      struct page *page)
+{
+	if (!z_erofs_is_shortlived_page(page))
+		return false;
+
+	/* short-lived pages should not be used by others at the same time */
+	if (page_ref_count(page) > 1) {
+		put_page(page);
+	} else {
+		/* follow the pcluster rule above. */
+		erofs_pagepool_add(pagepool, page);
+	}
+	return true;
+}
+
+#define MNGD_MAPPING(sbi)	((sbi)->managed_cache->i_mapping)
+static inline bool erofs_page_is_managed(const struct erofs_sb_info *sbi,
+					 struct page *page)
+{
+	return page->mapping == MNGD_MAPPING(sbi);
+}
+
+int z_erofs_fixup_insize(struct z_erofs_decompress_req *rq, const char *padbuf,
+			 unsigned int padbufsize);
+extern const struct z_erofs_decompressor erofs_decompressors[];
+
+/* prototypes for specific algorithms */
+int z_erofs_load_lzma_config(struct super_block *sb,
+			struct erofs_super_block *dsb, void *data, int size);
+int z_erofs_load_deflate_config(struct super_block *sb,
+			struct erofs_super_block *dsb, void *data, int size);
+int z_erofs_lzma_decompress(struct z_erofs_decompress_req *rq,
+			    struct page **pagepool);
+int z_erofs_deflate_decompress(struct z_erofs_decompress_req *rq,
+			       struct page **pagepool);
+#endif
diff --git a/fs/erofs/data.c b/fs/erofs/data.c
new file mode 100644
index 0000000000..0c2c99c58b
--- /dev/null
+++ b/fs/erofs/data.c
@@ -0,0 +1,452 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2021, Alibaba Cloud
+ */
+#include "internal.h"
+#include <linux/prefetch.h>
+#include <linux/sched/mm.h>
+#include <linux/dax.h>
+#include <trace/events/erofs.h>
+
+void erofs_unmap_metabuf(struct erofs_buf *buf)
+{
+	if (buf->kmap_type == EROFS_KMAP)
+		kunmap_local(buf->base);
+	buf->base = NULL;
+	buf->kmap_type = EROFS_NO_KMAP;
+}
+
+void erofs_put_metabuf(struct erofs_buf *buf)
+{
+	if (!buf->page)
+		return;
+	erofs_unmap_metabuf(buf);
+	put_page(buf->page);
+	buf->page = NULL;
+}
+
+/*
+ * Derive the block size from inode->i_blkbits to make compatible with
+ * anonymous inode in fscache mode.
+ */
+void *erofs_bread(struct erofs_buf *buf, erofs_blk_t blkaddr,
+		  enum erofs_kmap_type type)
+{
+	struct inode *inode = buf->inode;
+	erofs_off_t offset = (erofs_off_t)blkaddr << inode->i_blkbits;
+	pgoff_t index = offset >> PAGE_SHIFT;
+	struct page *page = buf->page;
+	struct folio *folio;
+	unsigned int nofs_flag;
+
+	if (!page || page->index != index) {
+		erofs_put_metabuf(buf);
+
+		nofs_flag = memalloc_nofs_save();
+		folio = read_cache_folio(inode->i_mapping, index, NULL, NULL);
+		memalloc_nofs_restore(nofs_flag);
+		if (IS_ERR(folio))
+			return folio;
+
+		/* should already be PageUptodate, no need to lock page */
+		page = folio_file_page(folio, index);
+		buf->page = page;
+	}
+	if (buf->kmap_type == EROFS_NO_KMAP) {
+		if (type == EROFS_KMAP)
+			buf->base = kmap_local_page(page);
+		buf->kmap_type = type;
+	} else if (buf->kmap_type != type) {
+		DBG_BUGON(1);
+		return ERR_PTR(-EFAULT);
+	}
+	if (type == EROFS_NO_KMAP)
+		return NULL;
+	return buf->base + (offset & ~PAGE_MASK);
+}
+
+void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb)
+{
+	if (erofs_is_fscache_mode(sb))
+		buf->inode = EROFS_SB(sb)->s_fscache->inode;
+	else
+		buf->inode = sb->s_bdev->bd_inode;
+}
+
+void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb,
+			 erofs_blk_t blkaddr, enum erofs_kmap_type type)
+{
+	erofs_init_metabuf(buf, sb);
+	return erofs_bread(buf, blkaddr, type);
+}
+
+static int erofs_map_blocks_flatmode(struct inode *inode,
+				     struct erofs_map_blocks *map)
+{
+	erofs_blk_t nblocks, lastblk;
+	u64 offset = map->m_la;
+	struct erofs_inode *vi = EROFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	bool tailendpacking = (vi->datalayout == EROFS_INODE_FLAT_INLINE);
+
+	nblocks = erofs_iblks(inode);
+	lastblk = nblocks - tailendpacking;
+
+	/* there is no hole in flatmode */
+	map->m_flags = EROFS_MAP_MAPPED;
+	if (offset < erofs_pos(sb, lastblk)) {
+		map->m_pa = erofs_pos(sb, vi->raw_blkaddr) + map->m_la;
+		map->m_plen = erofs_pos(sb, lastblk) - offset;
+	} else if (tailendpacking) {
+		map->m_pa = erofs_iloc(inode) + vi->inode_isize +
+			vi->xattr_isize + erofs_blkoff(sb, offset);
+		map->m_plen = inode->i_size - offset;
+
+		/* inline data should be located in the same meta block */
+		if (erofs_blkoff(sb, map->m_pa) + map->m_plen > sb->s_blocksize) {
+			erofs_err(sb, "inline data cross block boundary @ nid %llu",
+				  vi->nid);
+			DBG_BUGON(1);
+			return -EFSCORRUPTED;
+		}
+		map->m_flags |= EROFS_MAP_META;
+	} else {
+		erofs_err(sb, "internal error @ nid: %llu (size %llu), m_la 0x%llx",
+			  vi->nid, inode->i_size, map->m_la);
+		DBG_BUGON(1);
+		return -EIO;
+	}
+	return 0;
+}
+
+int erofs_map_blocks(struct inode *inode, struct erofs_map_blocks *map)
+{
+	struct super_block *sb = inode->i_sb;
+	struct erofs_inode *vi = EROFS_I(inode);
+	struct erofs_inode_chunk_index *idx;
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	u64 chunknr;
+	unsigned int unit;
+	erofs_off_t pos;
+	void *kaddr;
+	int err = 0;
+
+	trace_erofs_map_blocks_enter(inode, map, 0);
+	map->m_deviceid = 0;
+	if (map->m_la >= inode->i_size) {
+		/* leave out-of-bound access unmapped */
+		map->m_flags = 0;
+		map->m_plen = 0;
+		goto out;
+	}
+
+	if (vi->datalayout != EROFS_INODE_CHUNK_BASED) {
+		err = erofs_map_blocks_flatmode(inode, map);
+		goto out;
+	}
+
+	if (vi->chunkformat & EROFS_CHUNK_FORMAT_INDEXES)
+		unit = sizeof(*idx);			/* chunk index */
+	else
+		unit = EROFS_BLOCK_MAP_ENTRY_SIZE;	/* block map */
+
+	chunknr = map->m_la >> vi->chunkbits;
+	pos = ALIGN(erofs_iloc(inode) + vi->inode_isize +
+		    vi->xattr_isize, unit) + unit * chunknr;
+
+	kaddr = erofs_read_metabuf(&buf, sb, erofs_blknr(sb, pos), EROFS_KMAP);
+	if (IS_ERR(kaddr)) {
+		err = PTR_ERR(kaddr);
+		goto out;
+	}
+	map->m_la = chunknr << vi->chunkbits;
+	map->m_plen = min_t(erofs_off_t, 1UL << vi->chunkbits,
+			round_up(inode->i_size - map->m_la, sb->s_blocksize));
+
+	/* handle block map */
+	if (!(vi->chunkformat & EROFS_CHUNK_FORMAT_INDEXES)) {
+		__le32 *blkaddr = kaddr + erofs_blkoff(sb, pos);
+
+		if (le32_to_cpu(*blkaddr) == EROFS_NULL_ADDR) {
+			map->m_flags = 0;
+		} else {
+			map->m_pa = erofs_pos(sb, le32_to_cpu(*blkaddr));
+			map->m_flags = EROFS_MAP_MAPPED;
+		}
+		goto out_unlock;
+	}
+	/* parse chunk indexes */
+	idx = kaddr + erofs_blkoff(sb, pos);
+	switch (le32_to_cpu(idx->blkaddr)) {
+	case EROFS_NULL_ADDR:
+		map->m_flags = 0;
+		break;
+	default:
+		map->m_deviceid = le16_to_cpu(idx->device_id) &
+			EROFS_SB(sb)->device_id_mask;
+		map->m_pa = erofs_pos(sb, le32_to_cpu(idx->blkaddr));
+		map->m_flags = EROFS_MAP_MAPPED;
+		break;
+	}
+out_unlock:
+	erofs_put_metabuf(&buf);
+out:
+	if (!err)
+		map->m_llen = map->m_plen;
+	trace_erofs_map_blocks_exit(inode, map, 0, err);
+	return err;
+}
+
+int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *map)
+{
+	struct erofs_dev_context *devs = EROFS_SB(sb)->devs;
+	struct erofs_device_info *dif;
+	int id;
+
+	map->m_bdev = sb->s_bdev;
+	map->m_daxdev = EROFS_SB(sb)->dax_dev;
+	map->m_dax_part_off = EROFS_SB(sb)->dax_part_off;
+	map->m_fscache = EROFS_SB(sb)->s_fscache;
+
+	if (map->m_deviceid) {
+		down_read(&devs->rwsem);
+		dif = idr_find(&devs->tree, map->m_deviceid - 1);
+		if (!dif) {
+			up_read(&devs->rwsem);
+			return -ENODEV;
+		}
+		if (devs->flatdev) {
+			map->m_pa += erofs_pos(sb, dif->mapped_blkaddr);
+			up_read(&devs->rwsem);
+			return 0;
+		}
+		map->m_bdev = dif->bdev;
+		map->m_daxdev = dif->dax_dev;
+		map->m_dax_part_off = dif->dax_part_off;
+		map->m_fscache = dif->fscache;
+		up_read(&devs->rwsem);
+	} else if (devs->extra_devices && !devs->flatdev) {
+		down_read(&devs->rwsem);
+		idr_for_each_entry(&devs->tree, dif, id) {
+			erofs_off_t startoff, length;
+
+			if (!dif->mapped_blkaddr)
+				continue;
+			startoff = erofs_pos(sb, dif->mapped_blkaddr);
+			length = erofs_pos(sb, dif->blocks);
+
+			if (map->m_pa >= startoff &&
+			    map->m_pa < startoff + length) {
+				map->m_pa -= startoff;
+				map->m_bdev = dif->bdev;
+				map->m_daxdev = dif->dax_dev;
+				map->m_dax_part_off = dif->dax_part_off;
+				map->m_fscache = dif->fscache;
+				break;
+			}
+		}
+		up_read(&devs->rwsem);
+	}
+	return 0;
+}
+
+static int erofs_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
+		unsigned int flags, struct iomap *iomap, struct iomap *srcmap)
+{
+	int ret;
+	struct super_block *sb = inode->i_sb;
+	struct erofs_map_blocks map;
+	struct erofs_map_dev mdev;
+
+	map.m_la = offset;
+	map.m_llen = length;
+
+	ret = erofs_map_blocks(inode, &map);
+	if (ret < 0)
+		return ret;
+
+	mdev = (struct erofs_map_dev) {
+		.m_deviceid = map.m_deviceid,
+		.m_pa = map.m_pa,
+	};
+	ret = erofs_map_dev(sb, &mdev);
+	if (ret)
+		return ret;
+
+	iomap->offset = map.m_la;
+	if (flags & IOMAP_DAX)
+		iomap->dax_dev = mdev.m_daxdev;
+	else
+		iomap->bdev = mdev.m_bdev;
+	iomap->length = map.m_llen;
+	iomap->flags = 0;
+	iomap->private = NULL;
+
+	if (!(map.m_flags & EROFS_MAP_MAPPED)) {
+		iomap->type = IOMAP_HOLE;
+		iomap->addr = IOMAP_NULL_ADDR;
+		if (!iomap->length)
+			iomap->length = length;
+		return 0;
+	}
+
+	if (map.m_flags & EROFS_MAP_META) {
+		void *ptr;
+		struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+
+		iomap->type = IOMAP_INLINE;
+		ptr = erofs_read_metabuf(&buf, sb,
+				erofs_blknr(sb, mdev.m_pa), EROFS_KMAP);
+		if (IS_ERR(ptr))
+			return PTR_ERR(ptr);
+		iomap->inline_data = ptr + erofs_blkoff(sb, mdev.m_pa);
+		iomap->private = buf.base;
+	} else {
+		iomap->type = IOMAP_MAPPED;
+		iomap->addr = mdev.m_pa;
+		if (flags & IOMAP_DAX)
+			iomap->addr += mdev.m_dax_part_off;
+	}
+	return 0;
+}
+
+static int erofs_iomap_end(struct inode *inode, loff_t pos, loff_t length,
+		ssize_t written, unsigned int flags, struct iomap *iomap)
+{
+	void *ptr = iomap->private;
+
+	if (ptr) {
+		struct erofs_buf buf = {
+			.page = kmap_to_page(ptr),
+			.base = ptr,
+			.kmap_type = EROFS_KMAP,
+		};
+
+		DBG_BUGON(iomap->type != IOMAP_INLINE);
+		erofs_put_metabuf(&buf);
+	} else {
+		DBG_BUGON(iomap->type == IOMAP_INLINE);
+	}
+	return written;
+}
+
+static const struct iomap_ops erofs_iomap_ops = {
+	.iomap_begin = erofs_iomap_begin,
+	.iomap_end = erofs_iomap_end,
+};
+
+int erofs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		 u64 start, u64 len)
+{
+	if (erofs_inode_is_data_compressed(EROFS_I(inode)->datalayout)) {
+#ifdef CONFIG_EROFS_FS_ZIP
+		return iomap_fiemap(inode, fieinfo, start, len,
+				    &z_erofs_iomap_report_ops);
+#else
+		return -EOPNOTSUPP;
+#endif
+	}
+	return iomap_fiemap(inode, fieinfo, start, len, &erofs_iomap_ops);
+}
+
+/*
+ * since we dont have write or truncate flows, so no inode
+ * locking needs to be held at the moment.
+ */
+static int erofs_read_folio(struct file *file, struct folio *folio)
+{
+	return iomap_read_folio(folio, &erofs_iomap_ops);
+}
+
+static void erofs_readahead(struct readahead_control *rac)
+{
+	return iomap_readahead(rac, &erofs_iomap_ops);
+}
+
+static sector_t erofs_bmap(struct address_space *mapping, sector_t block)
+{
+	return iomap_bmap(mapping, block, &erofs_iomap_ops);
+}
+
+static ssize_t erofs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	/* no need taking (shared) inode lock since it's a ro filesystem */
+	if (!iov_iter_count(to))
+		return 0;
+
+#ifdef CONFIG_FS_DAX
+	if (IS_DAX(inode))
+		return dax_iomap_rw(iocb, to, &erofs_iomap_ops);
+#endif
+	if (iocb->ki_flags & IOCB_DIRECT) {
+		struct block_device *bdev = inode->i_sb->s_bdev;
+		unsigned int blksize_mask;
+
+		if (bdev)
+			blksize_mask = bdev_logical_block_size(bdev) - 1;
+		else
+			blksize_mask = i_blocksize(inode) - 1;
+
+		if ((iocb->ki_pos | iov_iter_count(to) |
+		     iov_iter_alignment(to)) & blksize_mask)
+			return -EINVAL;
+
+		return iomap_dio_rw(iocb, to, &erofs_iomap_ops,
+				    NULL, 0, NULL, 0);
+	}
+	return filemap_read(iocb, to, 0);
+}
+
+/* for uncompressed (aligned) files and raw access for other files */
+const struct address_space_operations erofs_raw_access_aops = {
+	.read_folio = erofs_read_folio,
+	.readahead = erofs_readahead,
+	.bmap = erofs_bmap,
+	.direct_IO = noop_direct_IO,
+	.release_folio = iomap_release_folio,
+	.invalidate_folio = iomap_invalidate_folio,
+};
+
+#ifdef CONFIG_FS_DAX
+static vm_fault_t erofs_dax_huge_fault(struct vm_fault *vmf,
+		unsigned int order)
+{
+	return dax_iomap_fault(vmf, order, NULL, NULL, &erofs_iomap_ops);
+}
+
+static vm_fault_t erofs_dax_fault(struct vm_fault *vmf)
+{
+	return erofs_dax_huge_fault(vmf, 0);
+}
+
+static const struct vm_operations_struct erofs_dax_vm_ops = {
+	.fault		= erofs_dax_fault,
+	.huge_fault	= erofs_dax_huge_fault,
+};
+
+static int erofs_file_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	if (!IS_DAX(file_inode(file)))
+		return generic_file_readonly_mmap(file, vma);
+
+	if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
+		return -EINVAL;
+
+	vma->vm_ops = &erofs_dax_vm_ops;
+	vm_flags_set(vma, VM_HUGEPAGE);
+	return 0;
+}
+#else
+#define erofs_file_mmap	generic_file_readonly_mmap
+#endif
+
+const struct file_operations erofs_file_fops = {
+	.llseek		= generic_file_llseek,
+	.read_iter	= erofs_file_read_iter,
+	.mmap		= erofs_file_mmap,
+	.splice_read	= filemap_splice_read,
+};
diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c
new file mode 100644
index 0000000000..d36b3963c0
--- /dev/null
+++ b/fs/erofs/decompressor.c
@@ -0,0 +1,445 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2019 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ */
+#include "compress.h"
+#include <linux/module.h>
+#include <linux/lz4.h>
+
+#ifndef LZ4_DISTANCE_MAX	/* history window size */
+#define LZ4_DISTANCE_MAX 65535	/* set to maximum value by default */
+#endif
+
+#define LZ4_MAX_DISTANCE_PAGES	(DIV_ROUND_UP(LZ4_DISTANCE_MAX, PAGE_SIZE) + 1)
+#ifndef LZ4_DECOMPRESS_INPLACE_MARGIN
+#define LZ4_DECOMPRESS_INPLACE_MARGIN(srcsize)  (((srcsize) >> 8) + 32)
+#endif
+
+struct z_erofs_lz4_decompress_ctx {
+	struct z_erofs_decompress_req *rq;
+	/* # of encoded, decoded pages */
+	unsigned int inpages, outpages;
+	/* decoded block total length (used for in-place decompression) */
+	unsigned int oend;
+};
+
+static int z_erofs_load_lz4_config(struct super_block *sb,
+			    struct erofs_super_block *dsb, void *data, int size)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct z_erofs_lz4_cfgs *lz4 = data;
+	u16 distance;
+
+	if (lz4) {
+		if (size < sizeof(struct z_erofs_lz4_cfgs)) {
+			erofs_err(sb, "invalid lz4 cfgs, size=%u", size);
+			return -EINVAL;
+		}
+		distance = le16_to_cpu(lz4->max_distance);
+
+		sbi->lz4.max_pclusterblks = le16_to_cpu(lz4->max_pclusterblks);
+		if (!sbi->lz4.max_pclusterblks) {
+			sbi->lz4.max_pclusterblks = 1;	/* reserved case */
+		} else if (sbi->lz4.max_pclusterblks >
+			   erofs_blknr(sb, Z_EROFS_PCLUSTER_MAX_SIZE)) {
+			erofs_err(sb, "too large lz4 pclusterblks %u",
+				  sbi->lz4.max_pclusterblks);
+			return -EINVAL;
+		}
+	} else {
+		distance = le16_to_cpu(dsb->u1.lz4_max_distance);
+		sbi->lz4.max_pclusterblks = 1;
+	}
+
+	sbi->lz4.max_distance_pages = distance ?
+					DIV_ROUND_UP(distance, PAGE_SIZE) + 1 :
+					LZ4_MAX_DISTANCE_PAGES;
+	return erofs_pcpubuf_growsize(sbi->lz4.max_pclusterblks);
+}
+
+/*
+ * Fill all gaps with bounce pages if it's a sparse page list. Also check if
+ * all physical pages are consecutive, which can be seen for moderate CR.
+ */
+static int z_erofs_lz4_prepare_dstpages(struct z_erofs_lz4_decompress_ctx *ctx,
+					struct page **pagepool)
+{
+	struct z_erofs_decompress_req *rq = ctx->rq;
+	struct page *availables[LZ4_MAX_DISTANCE_PAGES] = { NULL };
+	unsigned long bounced[DIV_ROUND_UP(LZ4_MAX_DISTANCE_PAGES,
+					   BITS_PER_LONG)] = { 0 };
+	unsigned int lz4_max_distance_pages =
+				EROFS_SB(rq->sb)->lz4.max_distance_pages;
+	void *kaddr = NULL;
+	unsigned int i, j, top;
+
+	top = 0;
+	for (i = j = 0; i < ctx->outpages; ++i, ++j) {
+		struct page *const page = rq->out[i];
+		struct page *victim;
+
+		if (j >= lz4_max_distance_pages)
+			j = 0;
+
+		/* 'valid' bounced can only be tested after a complete round */
+		if (!rq->fillgaps && test_bit(j, bounced)) {
+			DBG_BUGON(i < lz4_max_distance_pages);
+			DBG_BUGON(top >= lz4_max_distance_pages);
+			availables[top++] = rq->out[i - lz4_max_distance_pages];
+		}
+
+		if (page) {
+			__clear_bit(j, bounced);
+			if (!PageHighMem(page)) {
+				if (!i) {
+					kaddr = page_address(page);
+					continue;
+				}
+				if (kaddr &&
+				    kaddr + PAGE_SIZE == page_address(page)) {
+					kaddr += PAGE_SIZE;
+					continue;
+				}
+			}
+			kaddr = NULL;
+			continue;
+		}
+		kaddr = NULL;
+		__set_bit(j, bounced);
+
+		if (top) {
+			victim = availables[--top];
+			get_page(victim);
+		} else {
+			victim = erofs_allocpage(pagepool,
+						 GFP_KERNEL | __GFP_NOFAIL);
+			set_page_private(victim, Z_EROFS_SHORTLIVED_PAGE);
+		}
+		rq->out[i] = victim;
+	}
+	return kaddr ? 1 : 0;
+}
+
+static void *z_erofs_lz4_handle_overlap(struct z_erofs_lz4_decompress_ctx *ctx,
+			void *inpage, void *out, unsigned int *inputmargin,
+			int *maptype, bool may_inplace)
+{
+	struct z_erofs_decompress_req *rq = ctx->rq;
+	unsigned int omargin, total, i;
+	struct page **in;
+	void *src, *tmp;
+
+	if (rq->inplace_io) {
+		omargin = PAGE_ALIGN(ctx->oend) - ctx->oend;
+		if (rq->partial_decoding || !may_inplace ||
+		    omargin < LZ4_DECOMPRESS_INPLACE_MARGIN(rq->inputsize))
+			goto docopy;
+
+		for (i = 0; i < ctx->inpages; ++i)
+			if (rq->out[ctx->outpages - ctx->inpages + i] !=
+			    rq->in[i])
+				goto docopy;
+		kunmap_local(inpage);
+		*maptype = 3;
+		return out + ((ctx->outpages - ctx->inpages) << PAGE_SHIFT);
+	}
+
+	if (ctx->inpages <= 1) {
+		*maptype = 0;
+		return inpage;
+	}
+	kunmap_local(inpage);
+	src = erofs_vm_map_ram(rq->in, ctx->inpages);
+	if (!src)
+		return ERR_PTR(-ENOMEM);
+	*maptype = 1;
+	return src;
+
+docopy:
+	/* Or copy compressed data which can be overlapped to per-CPU buffer */
+	in = rq->in;
+	src = erofs_get_pcpubuf(ctx->inpages);
+	if (!src) {
+		DBG_BUGON(1);
+		kunmap_local(inpage);
+		return ERR_PTR(-EFAULT);
+	}
+
+	tmp = src;
+	total = rq->inputsize;
+	while (total) {
+		unsigned int page_copycnt =
+			min_t(unsigned int, total, PAGE_SIZE - *inputmargin);
+
+		if (!inpage)
+			inpage = kmap_local_page(*in);
+		memcpy(tmp, inpage + *inputmargin, page_copycnt);
+		kunmap_local(inpage);
+		inpage = NULL;
+		tmp += page_copycnt;
+		total -= page_copycnt;
+		++in;
+		*inputmargin = 0;
+	}
+	*maptype = 2;
+	return src;
+}
+
+/*
+ * Get the exact inputsize with zero_padding feature.
+ *  - For LZ4, it should work if zero_padding feature is on (5.3+);
+ *  - For MicroLZMA, it'd be enabled all the time.
+ */
+int z_erofs_fixup_insize(struct z_erofs_decompress_req *rq, const char *padbuf,
+			 unsigned int padbufsize)
+{
+	const char *padend;
+
+	padend = memchr_inv(padbuf, 0, padbufsize);
+	if (!padend)
+		return -EFSCORRUPTED;
+	rq->inputsize -= padend - padbuf;
+	rq->pageofs_in += padend - padbuf;
+	return 0;
+}
+
+static int z_erofs_lz4_decompress_mem(struct z_erofs_lz4_decompress_ctx *ctx,
+				      u8 *dst)
+{
+	struct z_erofs_decompress_req *rq = ctx->rq;
+	bool support_0padding = false, may_inplace = false;
+	unsigned int inputmargin;
+	u8 *out, *headpage, *src;
+	int ret, maptype;
+
+	DBG_BUGON(*rq->in == NULL);
+	headpage = kmap_local_page(*rq->in);
+
+	/* LZ4 decompression inplace is only safe if zero_padding is enabled */
+	if (erofs_sb_has_zero_padding(EROFS_SB(rq->sb))) {
+		support_0padding = true;
+		ret = z_erofs_fixup_insize(rq, headpage + rq->pageofs_in,
+				min_t(unsigned int, rq->inputsize,
+				      rq->sb->s_blocksize - rq->pageofs_in));
+		if (ret) {
+			kunmap_local(headpage);
+			return ret;
+		}
+		may_inplace = !((rq->pageofs_in + rq->inputsize) &
+				(rq->sb->s_blocksize - 1));
+	}
+
+	inputmargin = rq->pageofs_in;
+	src = z_erofs_lz4_handle_overlap(ctx, headpage, dst, &inputmargin,
+					 &maptype, may_inplace);
+	if (IS_ERR(src))
+		return PTR_ERR(src);
+
+	out = dst + rq->pageofs_out;
+	/* legacy format could compress extra data in a pcluster. */
+	if (rq->partial_decoding || !support_0padding)
+		ret = LZ4_decompress_safe_partial(src + inputmargin, out,
+				rq->inputsize, rq->outputsize, rq->outputsize);
+	else
+		ret = LZ4_decompress_safe(src + inputmargin, out,
+					  rq->inputsize, rq->outputsize);
+
+	if (ret != rq->outputsize) {
+		erofs_err(rq->sb, "failed to decompress %d in[%u, %u] out[%u]",
+			  ret, rq->inputsize, inputmargin, rq->outputsize);
+
+		print_hex_dump(KERN_DEBUG, "[ in]: ", DUMP_PREFIX_OFFSET,
+			       16, 1, src + inputmargin, rq->inputsize, true);
+		print_hex_dump(KERN_DEBUG, "[out]: ", DUMP_PREFIX_OFFSET,
+			       16, 1, out, rq->outputsize, true);
+
+		if (ret >= 0)
+			memset(out + ret, 0, rq->outputsize - ret);
+		ret = -EIO;
+	} else {
+		ret = 0;
+	}
+
+	if (maptype == 0) {
+		kunmap_local(headpage);
+	} else if (maptype == 1) {
+		vm_unmap_ram(src, ctx->inpages);
+	} else if (maptype == 2) {
+		erofs_put_pcpubuf(src);
+	} else if (maptype != 3) {
+		DBG_BUGON(1);
+		return -EFAULT;
+	}
+	return ret;
+}
+
+static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq,
+				  struct page **pagepool)
+{
+	struct z_erofs_lz4_decompress_ctx ctx;
+	unsigned int dst_maptype;
+	void *dst;
+	int ret;
+
+	ctx.rq = rq;
+	ctx.oend = rq->pageofs_out + rq->outputsize;
+	ctx.outpages = PAGE_ALIGN(ctx.oend) >> PAGE_SHIFT;
+	ctx.inpages = PAGE_ALIGN(rq->inputsize) >> PAGE_SHIFT;
+
+	/* one optimized fast path only for non bigpcluster cases yet */
+	if (ctx.inpages == 1 && ctx.outpages == 1 && !rq->inplace_io) {
+		DBG_BUGON(!*rq->out);
+		dst = kmap_local_page(*rq->out);
+		dst_maptype = 0;
+		goto dstmap_out;
+	}
+
+	/* general decoding path which can be used for all cases */
+	ret = z_erofs_lz4_prepare_dstpages(&ctx, pagepool);
+	if (ret < 0) {
+		return ret;
+	} else if (ret > 0) {
+		dst = page_address(*rq->out);
+		dst_maptype = 1;
+	} else {
+		dst = erofs_vm_map_ram(rq->out, ctx.outpages);
+		if (!dst)
+			return -ENOMEM;
+		dst_maptype = 2;
+	}
+
+dstmap_out:
+	ret = z_erofs_lz4_decompress_mem(&ctx, dst);
+	if (!dst_maptype)
+		kunmap_local(dst);
+	else if (dst_maptype == 2)
+		vm_unmap_ram(dst, ctx.outpages);
+	return ret;
+}
+
+static int z_erofs_transform_plain(struct z_erofs_decompress_req *rq,
+				   struct page **pagepool)
+{
+	const unsigned int inpages = PAGE_ALIGN(rq->inputsize) >> PAGE_SHIFT;
+	const unsigned int outpages =
+		PAGE_ALIGN(rq->pageofs_out + rq->outputsize) >> PAGE_SHIFT;
+	const unsigned int righthalf = min_t(unsigned int, rq->outputsize,
+					     PAGE_SIZE - rq->pageofs_out);
+	const unsigned int lefthalf = rq->outputsize - righthalf;
+	const unsigned int interlaced_offset =
+		rq->alg == Z_EROFS_COMPRESSION_SHIFTED ? 0 : rq->pageofs_out;
+	u8 *src;
+
+	if (outpages > 2 && rq->alg == Z_EROFS_COMPRESSION_SHIFTED) {
+		DBG_BUGON(1);
+		return -EFSCORRUPTED;
+	}
+
+	if (rq->out[0] == *rq->in) {
+		DBG_BUGON(rq->pageofs_out);
+		return 0;
+	}
+
+	src = kmap_local_page(rq->in[inpages - 1]) + rq->pageofs_in;
+	if (rq->out[0])
+		memcpy_to_page(rq->out[0], rq->pageofs_out,
+			       src + interlaced_offset, righthalf);
+
+	if (outpages > inpages) {
+		DBG_BUGON(!rq->out[outpages - 1]);
+		if (rq->out[outpages - 1] != rq->in[inpages - 1]) {
+			memcpy_to_page(rq->out[outpages - 1], 0, src +
+					(interlaced_offset ? 0 : righthalf),
+				       lefthalf);
+		} else if (!interlaced_offset) {
+			memmove(src, src + righthalf, lefthalf);
+			flush_dcache_page(rq->in[inpages - 1]);
+		}
+	}
+	kunmap_local(src);
+	return 0;
+}
+
+const struct z_erofs_decompressor erofs_decompressors[] = {
+	[Z_EROFS_COMPRESSION_SHIFTED] = {
+		.decompress = z_erofs_transform_plain,
+		.name = "shifted"
+	},
+	[Z_EROFS_COMPRESSION_INTERLACED] = {
+		.decompress = z_erofs_transform_plain,
+		.name = "interlaced"
+	},
+	[Z_EROFS_COMPRESSION_LZ4] = {
+		.config = z_erofs_load_lz4_config,
+		.decompress = z_erofs_lz4_decompress,
+		.name = "lz4"
+	},
+#ifdef CONFIG_EROFS_FS_ZIP_LZMA
+	[Z_EROFS_COMPRESSION_LZMA] = {
+		.config = z_erofs_load_lzma_config,
+		.decompress = z_erofs_lzma_decompress,
+		.name = "lzma"
+	},
+#endif
+#ifdef CONFIG_EROFS_FS_ZIP_DEFLATE
+	[Z_EROFS_COMPRESSION_DEFLATE] = {
+		.config = z_erofs_load_deflate_config,
+		.decompress = z_erofs_deflate_decompress,
+		.name = "deflate"
+	},
+#endif
+};
+
+int z_erofs_parse_cfgs(struct super_block *sb, struct erofs_super_block *dsb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	unsigned int algs, alg;
+	erofs_off_t offset;
+	int size, ret = 0;
+
+	if (!erofs_sb_has_compr_cfgs(sbi)) {
+		sbi->available_compr_algs = 1 << Z_EROFS_COMPRESSION_LZ4;
+		return z_erofs_load_lz4_config(sb, dsb, NULL, 0);
+	}
+
+	sbi->available_compr_algs = le16_to_cpu(dsb->u1.available_compr_algs);
+	if (sbi->available_compr_algs & ~Z_EROFS_ALL_COMPR_ALGS) {
+		erofs_err(sb, "unidentified algorithms %x, please upgrade kernel",
+			  sbi->available_compr_algs & ~Z_EROFS_ALL_COMPR_ALGS);
+		return -EOPNOTSUPP;
+	}
+
+	erofs_init_metabuf(&buf, sb);
+	offset = EROFS_SUPER_OFFSET + sbi->sb_size;
+	alg = 0;
+	for (algs = sbi->available_compr_algs; algs; algs >>= 1, ++alg) {
+		void *data;
+
+		if (!(algs & 1))
+			continue;
+
+		data = erofs_read_metadata(sb, &buf, &offset, &size);
+		if (IS_ERR(data)) {
+			ret = PTR_ERR(data);
+			break;
+		}
+
+		if (alg >= ARRAY_SIZE(erofs_decompressors) ||
+		    !erofs_decompressors[alg].config) {
+			erofs_err(sb, "algorithm %d isn't enabled on this kernel",
+				  alg);
+			ret = -EOPNOTSUPP;
+		} else {
+			ret = erofs_decompressors[alg].config(sb,
+					dsb, data, size);
+		}
+
+		kfree(data);
+		if (ret)
+			break;
+	}
+	erofs_put_metabuf(&buf);
+	return ret;
+}
diff --git a/fs/erofs/decompressor_deflate.c b/fs/erofs/decompressor_deflate.c
new file mode 100644
index 0000000000..0e1946a6bd
--- /dev/null
+++ b/fs/erofs/decompressor_deflate.c
@@ -0,0 +1,248 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+#include <linux/module.h>
+#include <linux/zlib.h>
+#include "compress.h"
+
+struct z_erofs_deflate {
+	struct z_erofs_deflate *next;
+	struct z_stream_s z;
+	u8 bounce[PAGE_SIZE];
+};
+
+static DEFINE_SPINLOCK(z_erofs_deflate_lock);
+static unsigned int z_erofs_deflate_nstrms, z_erofs_deflate_avail_strms;
+static struct z_erofs_deflate *z_erofs_deflate_head;
+static DECLARE_WAIT_QUEUE_HEAD(z_erofs_deflate_wq);
+
+module_param_named(deflate_streams, z_erofs_deflate_nstrms, uint, 0444);
+
+void z_erofs_deflate_exit(void)
+{
+	/* there should be no running fs instance */
+	while (z_erofs_deflate_avail_strms) {
+		struct z_erofs_deflate *strm;
+
+		spin_lock(&z_erofs_deflate_lock);
+		strm = z_erofs_deflate_head;
+		if (!strm) {
+			spin_unlock(&z_erofs_deflate_lock);
+			continue;
+		}
+		z_erofs_deflate_head = NULL;
+		spin_unlock(&z_erofs_deflate_lock);
+
+		while (strm) {
+			struct z_erofs_deflate *n = strm->next;
+
+			vfree(strm->z.workspace);
+			kfree(strm);
+			--z_erofs_deflate_avail_strms;
+			strm = n;
+		}
+	}
+}
+
+int __init z_erofs_deflate_init(void)
+{
+	/* by default, use # of possible CPUs instead */
+	if (!z_erofs_deflate_nstrms)
+		z_erofs_deflate_nstrms = num_possible_cpus();
+
+	for (; z_erofs_deflate_avail_strms < z_erofs_deflate_nstrms;
+	     ++z_erofs_deflate_avail_strms) {
+		struct z_erofs_deflate *strm;
+
+		strm = kzalloc(sizeof(*strm), GFP_KERNEL);
+		if (!strm)
+			goto out_failed;
+
+		/* XXX: in-kernel zlib cannot shrink windowbits currently */
+		strm->z.workspace = vmalloc(zlib_inflate_workspacesize());
+		if (!strm->z.workspace) {
+			kfree(strm);
+			goto out_failed;
+		}
+
+		spin_lock(&z_erofs_deflate_lock);
+		strm->next = z_erofs_deflate_head;
+		z_erofs_deflate_head = strm;
+		spin_unlock(&z_erofs_deflate_lock);
+	}
+	return 0;
+
+out_failed:
+	pr_err("failed to allocate zlib workspace\n");
+	z_erofs_deflate_exit();
+	return -ENOMEM;
+}
+
+int z_erofs_load_deflate_config(struct super_block *sb,
+			struct erofs_super_block *dsb, void *data, int size)
+{
+	struct z_erofs_deflate_cfgs *dfl = data;
+
+	if (!dfl || size < sizeof(struct z_erofs_deflate_cfgs)) {
+		erofs_err(sb, "invalid deflate cfgs, size=%u", size);
+		return -EINVAL;
+	}
+
+	if (dfl->windowbits > MAX_WBITS) {
+		erofs_err(sb, "unsupported windowbits %u", dfl->windowbits);
+		return -EOPNOTSUPP;
+	}
+
+	erofs_info(sb, "EXPERIMENTAL DEFLATE feature in use. Use at your own risk!");
+	return 0;
+}
+
+int z_erofs_deflate_decompress(struct z_erofs_decompress_req *rq,
+			       struct page **pagepool)
+{
+	const unsigned int nrpages_out =
+		PAGE_ALIGN(rq->pageofs_out + rq->outputsize) >> PAGE_SHIFT;
+	const unsigned int nrpages_in =
+		PAGE_ALIGN(rq->inputsize) >> PAGE_SHIFT;
+	struct super_block *sb = rq->sb;
+	unsigned int insz, outsz, pofs;
+	struct z_erofs_deflate *strm;
+	u8 *kin, *kout = NULL;
+	bool bounced = false;
+	int no = -1, ni = 0, j = 0, zerr, err;
+
+	/* 1. get the exact DEFLATE compressed size */
+	kin = kmap_local_page(*rq->in);
+	err = z_erofs_fixup_insize(rq, kin + rq->pageofs_in,
+			min_t(unsigned int, rq->inputsize,
+			      sb->s_blocksize - rq->pageofs_in));
+	if (err) {
+		kunmap_local(kin);
+		return err;
+	}
+
+	/* 2. get an available DEFLATE context */
+again:
+	spin_lock(&z_erofs_deflate_lock);
+	strm = z_erofs_deflate_head;
+	if (!strm) {
+		spin_unlock(&z_erofs_deflate_lock);
+		wait_event(z_erofs_deflate_wq, READ_ONCE(z_erofs_deflate_head));
+		goto again;
+	}
+	z_erofs_deflate_head = strm->next;
+	spin_unlock(&z_erofs_deflate_lock);
+
+	/* 3. multi-call decompress */
+	insz = rq->inputsize;
+	outsz = rq->outputsize;
+	zerr = zlib_inflateInit2(&strm->z, -MAX_WBITS);
+	if (zerr != Z_OK) {
+		err = -EIO;
+		goto failed_zinit;
+	}
+
+	pofs = rq->pageofs_out;
+	strm->z.avail_in = min_t(u32, insz, PAGE_SIZE - rq->pageofs_in);
+	insz -= strm->z.avail_in;
+	strm->z.next_in = kin + rq->pageofs_in;
+	strm->z.avail_out = 0;
+
+	while (1) {
+		if (!strm->z.avail_out) {
+			if (++no >= nrpages_out || !outsz) {
+				erofs_err(sb, "insufficient space for decompressed data");
+				err = -EFSCORRUPTED;
+				break;
+			}
+
+			if (kout)
+				kunmap_local(kout);
+			strm->z.avail_out = min_t(u32, outsz, PAGE_SIZE - pofs);
+			outsz -= strm->z.avail_out;
+			if (!rq->out[no]) {
+				rq->out[no] = erofs_allocpage(pagepool,
+						GFP_KERNEL | __GFP_NOFAIL);
+				set_page_private(rq->out[no],
+						 Z_EROFS_SHORTLIVED_PAGE);
+			}
+			kout = kmap_local_page(rq->out[no]);
+			strm->z.next_out = kout + pofs;
+			pofs = 0;
+		}
+
+		if (!strm->z.avail_in && insz) {
+			if (++ni >= nrpages_in) {
+				erofs_err(sb, "invalid compressed data");
+				err = -EFSCORRUPTED;
+				break;
+			}
+
+			if (kout) { /* unlike kmap(), take care of the orders */
+				j = strm->z.next_out - kout;
+				kunmap_local(kout);
+			}
+			kunmap_local(kin);
+			strm->z.avail_in = min_t(u32, insz, PAGE_SIZE);
+			insz -= strm->z.avail_in;
+			kin = kmap_local_page(rq->in[ni]);
+			strm->z.next_in = kin;
+			bounced = false;
+			if (kout) {
+				kout = kmap_local_page(rq->out[no]);
+				strm->z.next_out = kout + j;
+			}
+		}
+
+		/*
+		 * Handle overlapping: Use bounced buffer if the compressed
+		 * data is under processing; Or use short-lived pages from the
+		 * on-stack pagepool where pages share among the same request
+		 * and not _all_ inplace I/O pages are needed to be doubled.
+		 */
+		if (!bounced && rq->out[no] == rq->in[ni]) {
+			memcpy(strm->bounce, strm->z.next_in, strm->z.avail_in);
+			strm->z.next_in = strm->bounce;
+			bounced = true;
+		}
+
+		for (j = ni + 1; j < nrpages_in; ++j) {
+			struct page *tmppage;
+
+			if (rq->out[no] != rq->in[j])
+				continue;
+
+			DBG_BUGON(erofs_page_is_managed(EROFS_SB(sb),
+							rq->in[j]));
+			tmppage = erofs_allocpage(pagepool,
+						  GFP_KERNEL | __GFP_NOFAIL);
+			set_page_private(tmppage, Z_EROFS_SHORTLIVED_PAGE);
+			copy_highpage(tmppage, rq->in[j]);
+			rq->in[j] = tmppage;
+		}
+
+		zerr = zlib_inflate(&strm->z, Z_SYNC_FLUSH);
+		if (zerr != Z_OK || !(outsz + strm->z.avail_out)) {
+			if (zerr == Z_OK && rq->partial_decoding)
+				break;
+			if (zerr == Z_STREAM_END && !outsz)
+				break;
+			erofs_err(sb, "failed to decompress %d in[%u] out[%u]",
+				  zerr, rq->inputsize, rq->outputsize);
+			err = -EFSCORRUPTED;
+			break;
+		}
+	}
+
+	if (zlib_inflateEnd(&strm->z) != Z_OK && !err)
+		err = -EIO;
+	if (kout)
+		kunmap_local(kout);
+failed_zinit:
+	kunmap_local(kin);
+	/* 4. push back DEFLATE stream context to the global list */
+	spin_lock(&z_erofs_deflate_lock);
+	strm->next = z_erofs_deflate_head;
+	z_erofs_deflate_head = strm;
+	spin_unlock(&z_erofs_deflate_lock);
+	wake_up(&z_erofs_deflate_wq);
+	return err;
+}
diff --git a/fs/erofs/decompressor_lzma.c b/fs/erofs/decompressor_lzma.c
new file mode 100644
index 0000000000..ba4ec73f4a
--- /dev/null
+++ b/fs/erofs/decompressor_lzma.c
@@ -0,0 +1,294 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+#include <linux/xz.h>
+#include <linux/module.h>
+#include "compress.h"
+
+struct z_erofs_lzma {
+	struct z_erofs_lzma *next;
+	struct xz_dec_microlzma *state;
+	struct xz_buf buf;
+	u8 bounce[PAGE_SIZE];
+};
+
+/* considering the LZMA performance, no need to use a lockless list for now */
+static DEFINE_SPINLOCK(z_erofs_lzma_lock);
+static unsigned int z_erofs_lzma_max_dictsize;
+static unsigned int z_erofs_lzma_nstrms, z_erofs_lzma_avail_strms;
+static struct z_erofs_lzma *z_erofs_lzma_head;
+static DECLARE_WAIT_QUEUE_HEAD(z_erofs_lzma_wq);
+
+module_param_named(lzma_streams, z_erofs_lzma_nstrms, uint, 0444);
+
+void z_erofs_lzma_exit(void)
+{
+	/* there should be no running fs instance */
+	while (z_erofs_lzma_avail_strms) {
+		struct z_erofs_lzma *strm;
+
+		spin_lock(&z_erofs_lzma_lock);
+		strm = z_erofs_lzma_head;
+		if (!strm) {
+			spin_unlock(&z_erofs_lzma_lock);
+			DBG_BUGON(1);
+			return;
+		}
+		z_erofs_lzma_head = NULL;
+		spin_unlock(&z_erofs_lzma_lock);
+
+		while (strm) {
+			struct z_erofs_lzma *n = strm->next;
+
+			if (strm->state)
+				xz_dec_microlzma_end(strm->state);
+			kfree(strm);
+			--z_erofs_lzma_avail_strms;
+			strm = n;
+		}
+	}
+}
+
+int __init z_erofs_lzma_init(void)
+{
+	unsigned int i;
+
+	/* by default, use # of possible CPUs instead */
+	if (!z_erofs_lzma_nstrms)
+		z_erofs_lzma_nstrms = num_possible_cpus();
+
+	for (i = 0; i < z_erofs_lzma_nstrms; ++i) {
+		struct z_erofs_lzma *strm = kzalloc(sizeof(*strm), GFP_KERNEL);
+
+		if (!strm) {
+			z_erofs_lzma_exit();
+			return -ENOMEM;
+		}
+		spin_lock(&z_erofs_lzma_lock);
+		strm->next = z_erofs_lzma_head;
+		z_erofs_lzma_head = strm;
+		spin_unlock(&z_erofs_lzma_lock);
+		++z_erofs_lzma_avail_strms;
+	}
+	return 0;
+}
+
+int z_erofs_load_lzma_config(struct super_block *sb,
+			struct erofs_super_block *dsb, void *data, int size)
+{
+	static DEFINE_MUTEX(lzma_resize_mutex);
+	struct z_erofs_lzma_cfgs *lzma = data;
+	unsigned int dict_size, i;
+	struct z_erofs_lzma *strm, *head = NULL;
+	int err;
+
+	if (!lzma || size < sizeof(struct z_erofs_lzma_cfgs)) {
+		erofs_err(sb, "invalid lzma cfgs, size=%u", size);
+		return -EINVAL;
+	}
+	if (lzma->format) {
+		erofs_err(sb, "unidentified lzma format %x, please check kernel version",
+			  le16_to_cpu(lzma->format));
+		return -EINVAL;
+	}
+	dict_size = le32_to_cpu(lzma->dict_size);
+	if (dict_size > Z_EROFS_LZMA_MAX_DICT_SIZE || dict_size < 4096) {
+		erofs_err(sb, "unsupported lzma dictionary size %u",
+			  dict_size);
+		return -EINVAL;
+	}
+
+	erofs_info(sb, "EXPERIMENTAL MicroLZMA in use. Use at your own risk!");
+
+	/* in case 2 z_erofs_load_lzma_config() race to avoid deadlock */
+	mutex_lock(&lzma_resize_mutex);
+
+	if (z_erofs_lzma_max_dictsize >= dict_size) {
+		mutex_unlock(&lzma_resize_mutex);
+		return 0;
+	}
+
+	/* 1. collect/isolate all streams for the following check */
+	for (i = 0; i < z_erofs_lzma_avail_strms; ++i) {
+		struct z_erofs_lzma *last;
+
+again:
+		spin_lock(&z_erofs_lzma_lock);
+		strm = z_erofs_lzma_head;
+		if (!strm) {
+			spin_unlock(&z_erofs_lzma_lock);
+			wait_event(z_erofs_lzma_wq,
+				   READ_ONCE(z_erofs_lzma_head));
+			goto again;
+		}
+		z_erofs_lzma_head = NULL;
+		spin_unlock(&z_erofs_lzma_lock);
+
+		for (last = strm; last->next; last = last->next)
+			++i;
+		last->next = head;
+		head = strm;
+	}
+
+	err = 0;
+	/* 2. walk each isolated stream and grow max dict_size if needed */
+	for (strm = head; strm; strm = strm->next) {
+		if (strm->state)
+			xz_dec_microlzma_end(strm->state);
+		strm->state = xz_dec_microlzma_alloc(XZ_PREALLOC, dict_size);
+		if (!strm->state)
+			err = -ENOMEM;
+	}
+
+	/* 3. push back all to the global list and update max dict_size */
+	spin_lock(&z_erofs_lzma_lock);
+	DBG_BUGON(z_erofs_lzma_head);
+	z_erofs_lzma_head = head;
+	spin_unlock(&z_erofs_lzma_lock);
+	wake_up_all(&z_erofs_lzma_wq);
+
+	z_erofs_lzma_max_dictsize = dict_size;
+	mutex_unlock(&lzma_resize_mutex);
+	return err;
+}
+
+int z_erofs_lzma_decompress(struct z_erofs_decompress_req *rq,
+			    struct page **pagepool)
+{
+	const unsigned int nrpages_out =
+		PAGE_ALIGN(rq->pageofs_out + rq->outputsize) >> PAGE_SHIFT;
+	const unsigned int nrpages_in =
+		PAGE_ALIGN(rq->inputsize) >> PAGE_SHIFT;
+	unsigned int inlen, outlen, pageofs;
+	struct z_erofs_lzma *strm;
+	u8 *kin;
+	bool bounced = false;
+	int no, ni, j, err = 0;
+
+	/* 1. get the exact LZMA compressed size */
+	kin = kmap(*rq->in);
+	err = z_erofs_fixup_insize(rq, kin + rq->pageofs_in,
+			min_t(unsigned int, rq->inputsize,
+			      rq->sb->s_blocksize - rq->pageofs_in));
+	if (err) {
+		kunmap(*rq->in);
+		return err;
+	}
+
+	/* 2. get an available lzma context */
+again:
+	spin_lock(&z_erofs_lzma_lock);
+	strm = z_erofs_lzma_head;
+	if (!strm) {
+		spin_unlock(&z_erofs_lzma_lock);
+		wait_event(z_erofs_lzma_wq, READ_ONCE(z_erofs_lzma_head));
+		goto again;
+	}
+	z_erofs_lzma_head = strm->next;
+	spin_unlock(&z_erofs_lzma_lock);
+
+	/* 3. multi-call decompress */
+	inlen = rq->inputsize;
+	outlen = rq->outputsize;
+	xz_dec_microlzma_reset(strm->state, inlen, outlen,
+			       !rq->partial_decoding);
+	pageofs = rq->pageofs_out;
+	strm->buf.in = kin + rq->pageofs_in;
+	strm->buf.in_pos = 0;
+	strm->buf.in_size = min_t(u32, inlen, PAGE_SIZE - rq->pageofs_in);
+	inlen -= strm->buf.in_size;
+	strm->buf.out = NULL;
+	strm->buf.out_pos = 0;
+	strm->buf.out_size = 0;
+
+	for (ni = 0, no = -1;;) {
+		enum xz_ret xz_err;
+
+		if (strm->buf.out_pos == strm->buf.out_size) {
+			if (strm->buf.out) {
+				kunmap(rq->out[no]);
+				strm->buf.out = NULL;
+			}
+
+			if (++no >= nrpages_out || !outlen) {
+				erofs_err(rq->sb, "decompressed buf out of bound");
+				err = -EFSCORRUPTED;
+				break;
+			}
+			strm->buf.out_pos = 0;
+			strm->buf.out_size = min_t(u32, outlen,
+						   PAGE_SIZE - pageofs);
+			outlen -= strm->buf.out_size;
+			if (!rq->out[no] && rq->fillgaps) {	/* deduped */
+				rq->out[no] = erofs_allocpage(pagepool,
+						GFP_KERNEL | __GFP_NOFAIL);
+				set_page_private(rq->out[no],
+						 Z_EROFS_SHORTLIVED_PAGE);
+			}
+			if (rq->out[no])
+				strm->buf.out = kmap(rq->out[no]) + pageofs;
+			pageofs = 0;
+		} else if (strm->buf.in_pos == strm->buf.in_size) {
+			kunmap(rq->in[ni]);
+
+			if (++ni >= nrpages_in || !inlen) {
+				erofs_err(rq->sb, "compressed buf out of bound");
+				err = -EFSCORRUPTED;
+				break;
+			}
+			strm->buf.in_pos = 0;
+			strm->buf.in_size = min_t(u32, inlen, PAGE_SIZE);
+			inlen -= strm->buf.in_size;
+			kin = kmap(rq->in[ni]);
+			strm->buf.in = kin;
+			bounced = false;
+		}
+
+		/*
+		 * Handle overlapping: Use bounced buffer if the compressed
+		 * data is under processing; Otherwise, Use short-lived pages
+		 * from the on-stack pagepool where pages share with the same
+		 * request.
+		 */
+		if (!bounced && rq->out[no] == rq->in[ni]) {
+			memcpy(strm->bounce, strm->buf.in, strm->buf.in_size);
+			strm->buf.in = strm->bounce;
+			bounced = true;
+		}
+		for (j = ni + 1; j < nrpages_in; ++j) {
+			struct page *tmppage;
+
+			if (rq->out[no] != rq->in[j])
+				continue;
+
+			DBG_BUGON(erofs_page_is_managed(EROFS_SB(rq->sb),
+							rq->in[j]));
+			tmppage = erofs_allocpage(pagepool,
+						  GFP_KERNEL | __GFP_NOFAIL);
+			set_page_private(tmppage, Z_EROFS_SHORTLIVED_PAGE);
+			copy_highpage(tmppage, rq->in[j]);
+			rq->in[j] = tmppage;
+		}
+		xz_err = xz_dec_microlzma_run(strm->state, &strm->buf);
+		DBG_BUGON(strm->buf.out_pos > strm->buf.out_size);
+		DBG_BUGON(strm->buf.in_pos > strm->buf.in_size);
+
+		if (xz_err != XZ_OK) {
+			if (xz_err == XZ_STREAM_END && !outlen)
+				break;
+			erofs_err(rq->sb, "failed to decompress %d in[%u] out[%u]",
+				  xz_err, rq->inputsize, rq->outputsize);
+			err = -EFSCORRUPTED;
+			break;
+		}
+	}
+	if (no < nrpages_out && strm->buf.out)
+		kunmap(rq->out[no]);
+	if (ni < nrpages_in)
+		kunmap(rq->in[ni]);
+	/* 4. push back LZMA stream context to the global list */
+	spin_lock(&z_erofs_lzma_lock);
+	strm->next = z_erofs_lzma_head;
+	z_erofs_lzma_head = strm;
+	spin_unlock(&z_erofs_lzma_lock);
+	wake_up(&z_erofs_lzma_wq);
+	return err;
+}
diff --git a/fs/erofs/dir.c b/fs/erofs/dir.c
new file mode 100644
index 0000000000..b80abec053
--- /dev/null
+++ b/fs/erofs/dir.c
@@ -0,0 +1,111 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2022, Alibaba Cloud
+ */
+#include "internal.h"
+
+static int erofs_fill_dentries(struct inode *dir, struct dir_context *ctx,
+			       void *dentry_blk, struct erofs_dirent *de,
+			       unsigned int nameoff, unsigned int maxsize)
+{
+	const struct erofs_dirent *end = dentry_blk + nameoff;
+
+	while (de < end) {
+		const char *de_name;
+		unsigned int de_namelen;
+		unsigned char d_type;
+
+		d_type = fs_ftype_to_dtype(de->file_type);
+
+		nameoff = le16_to_cpu(de->nameoff);
+		de_name = (char *)dentry_blk + nameoff;
+
+		/* the last dirent in the block? */
+		if (de + 1 >= end)
+			de_namelen = strnlen(de_name, maxsize - nameoff);
+		else
+			de_namelen = le16_to_cpu(de[1].nameoff) - nameoff;
+
+		/* a corrupted entry is found */
+		if (nameoff + de_namelen > maxsize ||
+		    de_namelen > EROFS_NAME_LEN) {
+			erofs_err(dir->i_sb, "bogus dirent @ nid %llu",
+				  EROFS_I(dir)->nid);
+			DBG_BUGON(1);
+			return -EFSCORRUPTED;
+		}
+
+		if (!dir_emit(ctx, de_name, de_namelen,
+			      le64_to_cpu(de->nid), d_type))
+			return 1;
+		++de;
+		ctx->pos += sizeof(struct erofs_dirent);
+	}
+	return 0;
+}
+
+static int erofs_readdir(struct file *f, struct dir_context *ctx)
+{
+	struct inode *dir = file_inode(f);
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	struct super_block *sb = dir->i_sb;
+	unsigned long bsz = sb->s_blocksize;
+	const size_t dirsize = i_size_read(dir);
+	unsigned int i = erofs_blknr(sb, ctx->pos);
+	unsigned int ofs = erofs_blkoff(sb, ctx->pos);
+	int err = 0;
+	bool initial = true;
+
+	buf.inode = dir;
+	while (ctx->pos < dirsize) {
+		struct erofs_dirent *de;
+		unsigned int nameoff, maxsize;
+
+		de = erofs_bread(&buf, i, EROFS_KMAP);
+		if (IS_ERR(de)) {
+			erofs_err(sb, "fail to readdir of logical block %u of nid %llu",
+				  i, EROFS_I(dir)->nid);
+			err = PTR_ERR(de);
+			break;
+		}
+
+		nameoff = le16_to_cpu(de->nameoff);
+		if (nameoff < sizeof(struct erofs_dirent) || nameoff >= bsz) {
+			erofs_err(sb, "invalid de[0].nameoff %u @ nid %llu",
+				  nameoff, EROFS_I(dir)->nid);
+			err = -EFSCORRUPTED;
+			break;
+		}
+
+		maxsize = min_t(unsigned int, dirsize - ctx->pos + ofs, bsz);
+
+		/* search dirents at the arbitrary position */
+		if (initial) {
+			initial = false;
+
+			ofs = roundup(ofs, sizeof(struct erofs_dirent));
+			ctx->pos = erofs_pos(sb, i) + ofs;
+			if (ofs >= nameoff)
+				goto skip_this;
+		}
+
+		err = erofs_fill_dentries(dir, ctx, de, (void *)de + ofs,
+					  nameoff, maxsize);
+		if (err)
+			break;
+skip_this:
+		ctx->pos = erofs_pos(sb, i) + maxsize;
+		++i;
+		ofs = 0;
+	}
+	erofs_put_metabuf(&buf);
+	return err < 0 ? err : 0;
+}
+
+const struct file_operations erofs_dir_fops = {
+	.llseek		= generic_file_llseek,
+	.read		= generic_read_dir,
+	.iterate_shared	= erofs_readdir,
+};
diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h
new file mode 100644
index 0000000000..a03ec70ba6
--- /dev/null
+++ b/fs/erofs/erofs_fs.h
@@ -0,0 +1,461 @@
+/* SPDX-License-Identifier: GPL-2.0-only OR Apache-2.0 */
+/*
+ * EROFS (Enhanced ROM File System) on-disk format definition
+ *
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2021, Alibaba Cloud
+ */
+#ifndef __EROFS_FS_H
+#define __EROFS_FS_H
+
+#define EROFS_SUPER_OFFSET      1024
+
+#define EROFS_FEATURE_COMPAT_SB_CHKSUM          0x00000001
+#define EROFS_FEATURE_COMPAT_MTIME              0x00000002
+#define EROFS_FEATURE_COMPAT_XATTR_FILTER	0x00000004
+
+/*
+ * Any bits that aren't in EROFS_ALL_FEATURE_INCOMPAT should
+ * be incompatible with this kernel version.
+ */
+#define EROFS_FEATURE_INCOMPAT_ZERO_PADDING	0x00000001
+#define EROFS_FEATURE_INCOMPAT_COMPR_CFGS	0x00000002
+#define EROFS_FEATURE_INCOMPAT_BIG_PCLUSTER	0x00000002
+#define EROFS_FEATURE_INCOMPAT_CHUNKED_FILE	0x00000004
+#define EROFS_FEATURE_INCOMPAT_DEVICE_TABLE	0x00000008
+#define EROFS_FEATURE_INCOMPAT_COMPR_HEAD2	0x00000008
+#define EROFS_FEATURE_INCOMPAT_ZTAILPACKING	0x00000010
+#define EROFS_FEATURE_INCOMPAT_FRAGMENTS	0x00000020
+#define EROFS_FEATURE_INCOMPAT_DEDUPE		0x00000020
+#define EROFS_FEATURE_INCOMPAT_XATTR_PREFIXES	0x00000040
+#define EROFS_ALL_FEATURE_INCOMPAT		\
+	(EROFS_FEATURE_INCOMPAT_ZERO_PADDING | \
+	 EROFS_FEATURE_INCOMPAT_COMPR_CFGS | \
+	 EROFS_FEATURE_INCOMPAT_BIG_PCLUSTER | \
+	 EROFS_FEATURE_INCOMPAT_CHUNKED_FILE | \
+	 EROFS_FEATURE_INCOMPAT_DEVICE_TABLE | \
+	 EROFS_FEATURE_INCOMPAT_COMPR_HEAD2 | \
+	 EROFS_FEATURE_INCOMPAT_ZTAILPACKING | \
+	 EROFS_FEATURE_INCOMPAT_FRAGMENTS | \
+	 EROFS_FEATURE_INCOMPAT_DEDUPE | \
+	 EROFS_FEATURE_INCOMPAT_XATTR_PREFIXES)
+
+#define EROFS_SB_EXTSLOT_SIZE	16
+
+struct erofs_deviceslot {
+	u8 tag[64];		/* digest(sha256), etc. */
+	__le32 blocks;		/* total fs blocks of this device */
+	__le32 mapped_blkaddr;	/* map starting at mapped_blkaddr */
+	u8 reserved[56];
+};
+#define EROFS_DEVT_SLOT_SIZE	sizeof(struct erofs_deviceslot)
+
+/* erofs on-disk super block (currently 128 bytes) */
+struct erofs_super_block {
+	__le32 magic;           /* file system magic number */
+	__le32 checksum;        /* crc32c(super_block) */
+	__le32 feature_compat;
+	__u8 blkszbits;         /* filesystem block size in bit shift */
+	__u8 sb_extslots;	/* superblock size = 128 + sb_extslots * 16 */
+
+	__le16 root_nid;	/* nid of root directory */
+	__le64 inos;            /* total valid ino # (== f_files - f_favail) */
+
+	__le64 build_time;      /* compact inode time derivation */
+	__le32 build_time_nsec;	/* compact inode time derivation in ns scale */
+	__le32 blocks;          /* used for statfs */
+	__le32 meta_blkaddr;	/* start block address of metadata area */
+	__le32 xattr_blkaddr;	/* start block address of shared xattr area */
+	__u8 uuid[16];          /* 128-bit uuid for volume */
+	__u8 volume_name[16];   /* volume name */
+	__le32 feature_incompat;
+	union {
+		/* bitmap for available compression algorithms */
+		__le16 available_compr_algs;
+		/* customized sliding window size instead of 64k by default */
+		__le16 lz4_max_distance;
+	} __packed u1;
+	__le16 extra_devices;	/* # of devices besides the primary device */
+	__le16 devt_slotoff;	/* startoff = devt_slotoff * devt_slotsize */
+	__u8 dirblkbits;	/* directory block size in bit shift */
+	__u8 xattr_prefix_count;	/* # of long xattr name prefixes */
+	__le32 xattr_prefix_start;	/* start of long xattr prefixes */
+	__le64 packed_nid;	/* nid of the special packed inode */
+	__u8 xattr_filter_reserved; /* reserved for xattr name filter */
+	__u8 reserved2[23];
+};
+
+/*
+ * EROFS inode datalayout (i_format in on-disk inode):
+ * 0 - uncompressed flat inode without tail-packing inline data:
+ * 1 - compressed inode with non-compact indexes:
+ * 2 - uncompressed flat inode with tail-packing inline data:
+ * 3 - compressed inode with compact indexes:
+ * 4 - chunk-based inode with (optional) multi-device support:
+ * 5~7 - reserved
+ */
+enum {
+	EROFS_INODE_FLAT_PLAIN			= 0,
+	EROFS_INODE_COMPRESSED_FULL		= 1,
+	EROFS_INODE_FLAT_INLINE			= 2,
+	EROFS_INODE_COMPRESSED_COMPACT		= 3,
+	EROFS_INODE_CHUNK_BASED			= 4,
+	EROFS_INODE_DATALAYOUT_MAX
+};
+
+static inline bool erofs_inode_is_data_compressed(unsigned int datamode)
+{
+	return datamode == EROFS_INODE_COMPRESSED_COMPACT ||
+		datamode == EROFS_INODE_COMPRESSED_FULL;
+}
+
+/* bit definitions of inode i_format */
+#define EROFS_I_VERSION_MASK            0x01
+#define EROFS_I_DATALAYOUT_MASK         0x07
+
+#define EROFS_I_VERSION_BIT             0
+#define EROFS_I_DATALAYOUT_BIT          1
+#define EROFS_I_ALL_BIT			4
+
+#define EROFS_I_ALL	((1 << EROFS_I_ALL_BIT) - 1)
+
+/* indicate chunk blkbits, thus 'chunksize = blocksize << chunk blkbits' */
+#define EROFS_CHUNK_FORMAT_BLKBITS_MASK		0x001F
+/* with chunk indexes or just a 4-byte blkaddr array */
+#define EROFS_CHUNK_FORMAT_INDEXES		0x0020
+
+#define EROFS_CHUNK_FORMAT_ALL	\
+	(EROFS_CHUNK_FORMAT_BLKBITS_MASK | EROFS_CHUNK_FORMAT_INDEXES)
+
+/* 32-byte on-disk inode */
+#define EROFS_INODE_LAYOUT_COMPACT	0
+/* 64-byte on-disk inode */
+#define EROFS_INODE_LAYOUT_EXTENDED	1
+
+struct erofs_inode_chunk_info {
+	__le16 format;		/* chunk blkbits, etc. */
+	__le16 reserved;
+};
+
+union erofs_inode_i_u {
+	/* total compressed blocks for compressed inodes */
+	__le32 compressed_blocks;
+
+	/* block address for uncompressed flat inodes */
+	__le32 raw_blkaddr;
+
+	/* for device files, used to indicate old/new device # */
+	__le32 rdev;
+
+	/* for chunk-based files, it contains the summary info */
+	struct erofs_inode_chunk_info c;
+};
+
+/* 32-byte reduced form of an ondisk inode */
+struct erofs_inode_compact {
+	__le16 i_format;	/* inode format hints */
+
+/* 1 header + n-1 * 4 bytes inline xattr to keep continuity */
+	__le16 i_xattr_icount;
+	__le16 i_mode;
+	__le16 i_nlink;
+	__le32 i_size;
+	__le32 i_reserved;
+	union erofs_inode_i_u i_u;
+
+	__le32 i_ino;		/* only used for 32-bit stat compatibility */
+	__le16 i_uid;
+	__le16 i_gid;
+	__le32 i_reserved2;
+};
+
+/* 64-byte complete form of an ondisk inode */
+struct erofs_inode_extended {
+	__le16 i_format;	/* inode format hints */
+
+/* 1 header + n-1 * 4 bytes inline xattr to keep continuity */
+	__le16 i_xattr_icount;
+	__le16 i_mode;
+	__le16 i_reserved;
+	__le64 i_size;
+	union erofs_inode_i_u i_u;
+
+	__le32 i_ino;		/* only used for 32-bit stat compatibility */
+	__le32 i_uid;
+	__le32 i_gid;
+	__le64 i_mtime;
+	__le32 i_mtime_nsec;
+	__le32 i_nlink;
+	__u8   i_reserved2[16];
+};
+
+/*
+ * inline xattrs (n == i_xattr_icount):
+ * erofs_xattr_ibody_header(1) + (n - 1) * 4 bytes
+ *          12 bytes           /                   \
+ *                            /                     \
+ *                           /-----------------------\
+ *                           |  erofs_xattr_entries+ |
+ *                           +-----------------------+
+ * inline xattrs must starts in erofs_xattr_ibody_header,
+ * for read-only fs, no need to introduce h_refcount
+ */
+struct erofs_xattr_ibody_header {
+	__le32 h_name_filter;		/* bit value 1 indicates not-present */
+	__u8   h_shared_count;
+	__u8   h_reserved2[7];
+	__le32 h_shared_xattrs[];       /* shared xattr id array */
+};
+
+/* Name indexes */
+#define EROFS_XATTR_INDEX_USER              1
+#define EROFS_XATTR_INDEX_POSIX_ACL_ACCESS  2
+#define EROFS_XATTR_INDEX_POSIX_ACL_DEFAULT 3
+#define EROFS_XATTR_INDEX_TRUSTED           4
+#define EROFS_XATTR_INDEX_LUSTRE            5
+#define EROFS_XATTR_INDEX_SECURITY          6
+
+/*
+ * bit 7 of e_name_index is set when it refers to a long xattr name prefix,
+ * while the remained lower bits represent the index of the prefix.
+ */
+#define EROFS_XATTR_LONG_PREFIX		0x80
+#define EROFS_XATTR_LONG_PREFIX_MASK	0x7f
+
+#define EROFS_XATTR_FILTER_BITS		32
+#define EROFS_XATTR_FILTER_DEFAULT	UINT32_MAX
+#define EROFS_XATTR_FILTER_SEED		0x25BBE08F
+
+/* xattr entry (for both inline & shared xattrs) */
+struct erofs_xattr_entry {
+	__u8   e_name_len;      /* length of name */
+	__u8   e_name_index;    /* attribute name index */
+	__le16 e_value_size;    /* size of attribute value */
+	/* followed by e_name and e_value */
+	char   e_name[];        /* attribute name */
+};
+
+/* long xattr name prefix */
+struct erofs_xattr_long_prefix {
+	__u8   base_index;	/* short xattr name prefix index */
+	char   infix[];		/* infix apart from short prefix */
+};
+
+static inline unsigned int erofs_xattr_ibody_size(__le16 i_xattr_icount)
+{
+	if (!i_xattr_icount)
+		return 0;
+
+	return sizeof(struct erofs_xattr_ibody_header) +
+		sizeof(__u32) * (le16_to_cpu(i_xattr_icount) - 1);
+}
+
+#define EROFS_XATTR_ALIGN(size) round_up(size, sizeof(struct erofs_xattr_entry))
+
+static inline unsigned int erofs_xattr_entry_size(struct erofs_xattr_entry *e)
+{
+	return EROFS_XATTR_ALIGN(sizeof(struct erofs_xattr_entry) +
+				 e->e_name_len + le16_to_cpu(e->e_value_size));
+}
+
+/* represent a zeroed chunk (hole) */
+#define EROFS_NULL_ADDR			-1
+
+/* 4-byte block address array */
+#define EROFS_BLOCK_MAP_ENTRY_SIZE	sizeof(__le32)
+
+/* 8-byte inode chunk indexes */
+struct erofs_inode_chunk_index {
+	__le16 advise;		/* always 0, don't care for now */
+	__le16 device_id;	/* back-end storage id (with bits masked) */
+	__le32 blkaddr;		/* start block address of this inode chunk */
+};
+
+/* dirent sorts in alphabet order, thus we can do binary search */
+struct erofs_dirent {
+	__le64 nid;     /* node number */
+	__le16 nameoff; /* start offset of file name */
+	__u8 file_type; /* file type */
+	__u8 reserved;  /* reserved */
+} __packed;
+
+/*
+ * EROFS file types should match generic FT_* types and
+ * it seems no need to add BUILD_BUG_ONs since potential
+ * unmatchness will break other fses as well...
+ */
+
+#define EROFS_NAME_LEN      255
+
+/* maximum supported size of a physical compression cluster */
+#define Z_EROFS_PCLUSTER_MAX_SIZE	(1024 * 1024)
+
+/* available compression algorithm types (for h_algorithmtype) */
+enum {
+	Z_EROFS_COMPRESSION_LZ4		= 0,
+	Z_EROFS_COMPRESSION_LZMA	= 1,
+	Z_EROFS_COMPRESSION_DEFLATE	= 2,
+	Z_EROFS_COMPRESSION_MAX
+};
+#define Z_EROFS_ALL_COMPR_ALGS		((1 << Z_EROFS_COMPRESSION_MAX) - 1)
+
+/* 14 bytes (+ length field = 16 bytes) */
+struct z_erofs_lz4_cfgs {
+	__le16 max_distance;
+	__le16 max_pclusterblks;
+	u8 reserved[10];
+} __packed;
+
+/* 14 bytes (+ length field = 16 bytes) */
+struct z_erofs_lzma_cfgs {
+	__le32 dict_size;
+	__le16 format;
+	u8 reserved[8];
+} __packed;
+
+#define Z_EROFS_LZMA_MAX_DICT_SIZE	(8 * Z_EROFS_PCLUSTER_MAX_SIZE)
+
+/* 6 bytes (+ length field = 8 bytes) */
+struct z_erofs_deflate_cfgs {
+	u8 windowbits;			/* 8..15 for DEFLATE */
+	u8 reserved[5];
+} __packed;
+
+/*
+ * bit 0 : COMPACTED_2B indexes (0 - off; 1 - on)
+ *  e.g. for 4k logical cluster size,      4B        if compacted 2B is off;
+ *                                  (4B) + 2B + (4B) if compacted 2B is on.
+ * bit 1 : HEAD1 big pcluster (0 - off; 1 - on)
+ * bit 2 : HEAD2 big pcluster (0 - off; 1 - on)
+ * bit 3 : tailpacking inline pcluster (0 - off; 1 - on)
+ * bit 4 : interlaced plain pcluster (0 - off; 1 - on)
+ * bit 5 : fragment pcluster (0 - off; 1 - on)
+ */
+#define Z_EROFS_ADVISE_COMPACTED_2B		0x0001
+#define Z_EROFS_ADVISE_BIG_PCLUSTER_1		0x0002
+#define Z_EROFS_ADVISE_BIG_PCLUSTER_2		0x0004
+#define Z_EROFS_ADVISE_INLINE_PCLUSTER		0x0008
+#define Z_EROFS_ADVISE_INTERLACED_PCLUSTER	0x0010
+#define Z_EROFS_ADVISE_FRAGMENT_PCLUSTER	0x0020
+
+#define Z_EROFS_FRAGMENT_INODE_BIT              7
+struct z_erofs_map_header {
+	union {
+		/* fragment data offset in the packed inode */
+		__le32  h_fragmentoff;
+		struct {
+			__le16  h_reserved1;
+			/* indicates the encoded size of tailpacking data */
+			__le16  h_idata_size;
+		};
+	};
+	__le16	h_advise;
+	/*
+	 * bit 0-3 : algorithm type of head 1 (logical cluster type 01);
+	 * bit 4-7 : algorithm type of head 2 (logical cluster type 11).
+	 */
+	__u8	h_algorithmtype;
+	/*
+	 * bit 0-2 : logical cluster bits - 12, e.g. 0 for 4096;
+	 * bit 3-6 : reserved;
+	 * bit 7   : move the whole file into packed inode or not.
+	 */
+	__u8	h_clusterbits;
+};
+
+/*
+ * On-disk logical cluster type:
+ *    0   - literal (uncompressed) lcluster
+ *    1,3 - compressed lcluster (for HEAD lclusters)
+ *    2   - compressed lcluster (for NONHEAD lclusters)
+ *
+ * In detail,
+ *    0 - literal (uncompressed) lcluster,
+ *        di_advise = 0
+ *        di_clusterofs = the literal data offset of the lcluster
+ *        di_blkaddr = the blkaddr of the literal pcluster
+ *
+ *    1,3 - compressed lcluster (for HEAD lclusters)
+ *        di_advise = 1 or 3
+ *        di_clusterofs = the decompressed data offset of the lcluster
+ *        di_blkaddr = the blkaddr of the compressed pcluster
+ *
+ *    2 - compressed lcluster (for NONHEAD lclusters)
+ *        di_advise = 2
+ *        di_clusterofs =
+ *           the decompressed data offset in its own HEAD lcluster
+ *        di_u.delta[0] = distance to this HEAD lcluster
+ *        di_u.delta[1] = distance to the next HEAD lcluster
+ */
+enum {
+	Z_EROFS_LCLUSTER_TYPE_PLAIN	= 0,
+	Z_EROFS_LCLUSTER_TYPE_HEAD1	= 1,
+	Z_EROFS_LCLUSTER_TYPE_NONHEAD	= 2,
+	Z_EROFS_LCLUSTER_TYPE_HEAD2	= 3,
+	Z_EROFS_LCLUSTER_TYPE_MAX
+};
+
+#define Z_EROFS_LI_LCLUSTER_TYPE_BITS        2
+#define Z_EROFS_LI_LCLUSTER_TYPE_BIT         0
+
+/* (noncompact only, HEAD) This pcluster refers to partial decompressed data */
+#define Z_EROFS_LI_PARTIAL_REF		(1 << 15)
+
+/*
+ * D0_CBLKCNT will be marked _only_ at the 1st non-head lcluster to store the
+ * compressed block count of a compressed extent (in logical clusters, aka.
+ * block count of a pcluster).
+ */
+#define Z_EROFS_LI_D0_CBLKCNT		(1 << 11)
+
+struct z_erofs_lcluster_index {
+	__le16 di_advise;
+	/* where to decompress in the head lcluster */
+	__le16 di_clusterofs;
+
+	union {
+		/* for the HEAD lclusters */
+		__le32 blkaddr;
+		/*
+		 * for the NONHEAD lclusters
+		 * [0] - distance to its HEAD lcluster
+		 * [1] - distance to the next HEAD lcluster
+		 */
+		__le16 delta[2];
+	} di_u;
+};
+
+#define Z_EROFS_FULL_INDEX_ALIGN(end)	\
+	(ALIGN(end, 8) + sizeof(struct z_erofs_map_header) + 8)
+
+/* check the EROFS on-disk layout strictly at compile time */
+static inline void erofs_check_ondisk_layout_definitions(void)
+{
+	const __le64 fmh = *(__le64 *)&(struct z_erofs_map_header) {
+		.h_clusterbits = 1 << Z_EROFS_FRAGMENT_INODE_BIT
+	};
+
+	BUILD_BUG_ON(sizeof(struct erofs_super_block) != 128);
+	BUILD_BUG_ON(sizeof(struct erofs_inode_compact) != 32);
+	BUILD_BUG_ON(sizeof(struct erofs_inode_extended) != 64);
+	BUILD_BUG_ON(sizeof(struct erofs_xattr_ibody_header) != 12);
+	BUILD_BUG_ON(sizeof(struct erofs_xattr_entry) != 4);
+	BUILD_BUG_ON(sizeof(struct erofs_inode_chunk_info) != 4);
+	BUILD_BUG_ON(sizeof(struct erofs_inode_chunk_index) != 8);
+	BUILD_BUG_ON(sizeof(struct z_erofs_map_header) != 8);
+	BUILD_BUG_ON(sizeof(struct z_erofs_lcluster_index) != 8);
+	BUILD_BUG_ON(sizeof(struct erofs_dirent) != 12);
+	/* keep in sync between 2 index structures for better extendibility */
+	BUILD_BUG_ON(sizeof(struct erofs_inode_chunk_index) !=
+		     sizeof(struct z_erofs_lcluster_index));
+	BUILD_BUG_ON(sizeof(struct erofs_deviceslot) != 128);
+
+	BUILD_BUG_ON(BIT(Z_EROFS_LI_LCLUSTER_TYPE_BITS) <
+		     Z_EROFS_LCLUSTER_TYPE_MAX - 1);
+	/* exclude old compiler versions like gcc 7.5.0 */
+	BUILD_BUG_ON(__builtin_constant_p(fmh) ?
+		     fmh != cpu_to_le64(1ULL << 63) : 0);
+}
+
+#endif
diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c
new file mode 100644
index 0000000000..87ff35bff8
--- /dev/null
+++ b/fs/erofs/fscache.c
@@ -0,0 +1,612 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2022, Alibaba Cloud
+ * Copyright (C) 2022, Bytedance Inc. All rights reserved.
+ */
+#include <linux/fscache.h>
+#include "internal.h"
+
+static DEFINE_MUTEX(erofs_domain_list_lock);
+static DEFINE_MUTEX(erofs_domain_cookies_lock);
+static LIST_HEAD(erofs_domain_list);
+static LIST_HEAD(erofs_domain_cookies_list);
+static struct vfsmount *erofs_pseudo_mnt;
+
+struct erofs_fscache_request {
+	struct erofs_fscache_request *primary;
+	struct netfs_cache_resources cache_resources;
+	struct address_space	*mapping;	/* The mapping being accessed */
+	loff_t			start;		/* Start position */
+	size_t			len;		/* Length of the request */
+	size_t			submitted;	/* Length of submitted */
+	short			error;		/* 0 or error that occurred */
+	refcount_t		ref;
+};
+
+static struct erofs_fscache_request *erofs_fscache_req_alloc(struct address_space *mapping,
+					     loff_t start, size_t len)
+{
+	struct erofs_fscache_request *req;
+
+	req = kzalloc(sizeof(struct erofs_fscache_request), GFP_KERNEL);
+	if (!req)
+		return ERR_PTR(-ENOMEM);
+
+	req->mapping = mapping;
+	req->start   = start;
+	req->len     = len;
+	refcount_set(&req->ref, 1);
+
+	return req;
+}
+
+static struct erofs_fscache_request *erofs_fscache_req_chain(struct erofs_fscache_request *primary,
+					     size_t len)
+{
+	struct erofs_fscache_request *req;
+
+	/* use primary request for the first submission */
+	if (!primary->submitted) {
+		refcount_inc(&primary->ref);
+		return primary;
+	}
+
+	req = erofs_fscache_req_alloc(primary->mapping,
+			primary->start + primary->submitted, len);
+	if (!IS_ERR(req)) {
+		req->primary = primary;
+		refcount_inc(&primary->ref);
+	}
+	return req;
+}
+
+static void erofs_fscache_req_complete(struct erofs_fscache_request *req)
+{
+	struct folio *folio;
+	bool failed = req->error;
+	pgoff_t start_page = req->start / PAGE_SIZE;
+	pgoff_t last_page = ((req->start + req->len) / PAGE_SIZE) - 1;
+
+	XA_STATE(xas, &req->mapping->i_pages, start_page);
+
+	rcu_read_lock();
+	xas_for_each(&xas, folio, last_page) {
+		if (xas_retry(&xas, folio))
+			continue;
+		if (!failed)
+			folio_mark_uptodate(folio);
+		folio_unlock(folio);
+	}
+	rcu_read_unlock();
+}
+
+static void erofs_fscache_req_put(struct erofs_fscache_request *req)
+{
+	if (refcount_dec_and_test(&req->ref)) {
+		if (req->cache_resources.ops)
+			req->cache_resources.ops->end_operation(&req->cache_resources);
+		if (!req->primary)
+			erofs_fscache_req_complete(req);
+		else
+			erofs_fscache_req_put(req->primary);
+		kfree(req);
+	}
+}
+
+static void erofs_fscache_subreq_complete(void *priv,
+		ssize_t transferred_or_error, bool was_async)
+{
+	struct erofs_fscache_request *req = priv;
+
+	if (IS_ERR_VALUE(transferred_or_error)) {
+		if (req->primary)
+			req->primary->error = transferred_or_error;
+		else
+			req->error = transferred_or_error;
+	}
+	erofs_fscache_req_put(req);
+}
+
+/*
+ * Read data from fscache (cookie, pstart, len), and fill the read data into
+ * page cache described by (req->mapping, lstart, len). @pstart describeis the
+ * start physical address in the cache file.
+ */
+static int erofs_fscache_read_folios_async(struct fscache_cookie *cookie,
+		struct erofs_fscache_request *req, loff_t pstart, size_t len)
+{
+	enum netfs_io_source source;
+	struct super_block *sb = req->mapping->host->i_sb;
+	struct netfs_cache_resources *cres = &req->cache_resources;
+	struct iov_iter iter;
+	loff_t lstart = req->start + req->submitted;
+	size_t done = 0;
+	int ret;
+
+	DBG_BUGON(len > req->len - req->submitted);
+
+	ret = fscache_begin_read_operation(cres, cookie);
+	if (ret)
+		return ret;
+
+	while (done < len) {
+		loff_t sstart = pstart + done;
+		size_t slen = len - done;
+		unsigned long flags = 1 << NETFS_SREQ_ONDEMAND;
+
+		source = cres->ops->prepare_ondemand_read(cres,
+				sstart, &slen, LLONG_MAX, &flags, 0);
+		if (WARN_ON(slen == 0))
+			source = NETFS_INVALID_READ;
+		if (source != NETFS_READ_FROM_CACHE) {
+			erofs_err(sb, "failed to fscache prepare_read (source %d)", source);
+			return -EIO;
+		}
+
+		refcount_inc(&req->ref);
+		iov_iter_xarray(&iter, ITER_DEST, &req->mapping->i_pages,
+				lstart + done, slen);
+
+		ret = fscache_read(cres, sstart, &iter, NETFS_READ_HOLE_FAIL,
+				   erofs_fscache_subreq_complete, req);
+		if (ret == -EIOCBQUEUED)
+			ret = 0;
+		if (ret) {
+			erofs_err(sb, "failed to fscache_read (ret %d)", ret);
+			return ret;
+		}
+
+		done += slen;
+	}
+	DBG_BUGON(done != len);
+	return 0;
+}
+
+static int erofs_fscache_meta_read_folio(struct file *data, struct folio *folio)
+{
+	int ret;
+	struct erofs_fscache *ctx = folio_mapping(folio)->host->i_private;
+	struct erofs_fscache_request *req;
+
+	req = erofs_fscache_req_alloc(folio_mapping(folio),
+				folio_pos(folio), folio_size(folio));
+	if (IS_ERR(req)) {
+		folio_unlock(folio);
+		return PTR_ERR(req);
+	}
+
+	ret = erofs_fscache_read_folios_async(ctx->cookie, req,
+				folio_pos(folio), folio_size(folio));
+	if (ret)
+		req->error = ret;
+
+	erofs_fscache_req_put(req);
+	return ret;
+}
+
+static int erofs_fscache_data_read_slice(struct erofs_fscache_request *primary)
+{
+	struct address_space *mapping = primary->mapping;
+	struct inode *inode = mapping->host;
+	struct super_block *sb = inode->i_sb;
+	struct erofs_fscache_request *req;
+	struct erofs_map_blocks map;
+	struct erofs_map_dev mdev;
+	struct iov_iter iter;
+	loff_t pos = primary->start + primary->submitted;
+	size_t count;
+	int ret;
+
+	map.m_la = pos;
+	ret = erofs_map_blocks(inode, &map);
+	if (ret)
+		return ret;
+
+	if (map.m_flags & EROFS_MAP_META) {
+		struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+		erofs_blk_t blknr;
+		size_t offset, size;
+		void *src;
+
+		/* For tail packing layout, the offset may be non-zero. */
+		offset = erofs_blkoff(sb, map.m_pa);
+		blknr = erofs_blknr(sb, map.m_pa);
+		size = map.m_llen;
+
+		src = erofs_read_metabuf(&buf, sb, blknr, EROFS_KMAP);
+		if (IS_ERR(src))
+			return PTR_ERR(src);
+
+		iov_iter_xarray(&iter, ITER_DEST, &mapping->i_pages, pos, PAGE_SIZE);
+		if (copy_to_iter(src + offset, size, &iter) != size) {
+			erofs_put_metabuf(&buf);
+			return -EFAULT;
+		}
+		iov_iter_zero(PAGE_SIZE - size, &iter);
+		erofs_put_metabuf(&buf);
+		primary->submitted += PAGE_SIZE;
+		return 0;
+	}
+
+	count = primary->len - primary->submitted;
+	if (!(map.m_flags & EROFS_MAP_MAPPED)) {
+		iov_iter_xarray(&iter, ITER_DEST, &mapping->i_pages, pos, count);
+		iov_iter_zero(count, &iter);
+		primary->submitted += count;
+		return 0;
+	}
+
+	count = min_t(size_t, map.m_llen - (pos - map.m_la), count);
+	DBG_BUGON(!count || count % PAGE_SIZE);
+
+	mdev = (struct erofs_map_dev) {
+		.m_deviceid = map.m_deviceid,
+		.m_pa = map.m_pa,
+	};
+	ret = erofs_map_dev(sb, &mdev);
+	if (ret)
+		return ret;
+
+	req = erofs_fscache_req_chain(primary, count);
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+
+	ret = erofs_fscache_read_folios_async(mdev.m_fscache->cookie,
+			req, mdev.m_pa + (pos - map.m_la), count);
+	erofs_fscache_req_put(req);
+	primary->submitted += count;
+	return ret;
+}
+
+static int erofs_fscache_data_read(struct erofs_fscache_request *req)
+{
+	int ret;
+
+	do {
+		ret = erofs_fscache_data_read_slice(req);
+		if (ret)
+			req->error = ret;
+	} while (!ret && req->submitted < req->len);
+
+	return ret;
+}
+
+static int erofs_fscache_read_folio(struct file *file, struct folio *folio)
+{
+	struct erofs_fscache_request *req;
+	int ret;
+
+	req = erofs_fscache_req_alloc(folio_mapping(folio),
+			folio_pos(folio), folio_size(folio));
+	if (IS_ERR(req)) {
+		folio_unlock(folio);
+		return PTR_ERR(req);
+	}
+
+	ret = erofs_fscache_data_read(req);
+	erofs_fscache_req_put(req);
+	return ret;
+}
+
+static void erofs_fscache_readahead(struct readahead_control *rac)
+{
+	struct erofs_fscache_request *req;
+
+	if (!readahead_count(rac))
+		return;
+
+	req = erofs_fscache_req_alloc(rac->mapping,
+			readahead_pos(rac), readahead_length(rac));
+	if (IS_ERR(req))
+		return;
+
+	/* The request completion will drop refs on the folios. */
+	while (readahead_folio(rac))
+		;
+
+	erofs_fscache_data_read(req);
+	erofs_fscache_req_put(req);
+}
+
+static const struct address_space_operations erofs_fscache_meta_aops = {
+	.read_folio = erofs_fscache_meta_read_folio,
+};
+
+const struct address_space_operations erofs_fscache_access_aops = {
+	.read_folio = erofs_fscache_read_folio,
+	.readahead = erofs_fscache_readahead,
+};
+
+static void erofs_fscache_domain_put(struct erofs_domain *domain)
+{
+	mutex_lock(&erofs_domain_list_lock);
+	if (refcount_dec_and_test(&domain->ref)) {
+		list_del(&domain->list);
+		if (list_empty(&erofs_domain_list)) {
+			kern_unmount(erofs_pseudo_mnt);
+			erofs_pseudo_mnt = NULL;
+		}
+		fscache_relinquish_volume(domain->volume, NULL, false);
+		mutex_unlock(&erofs_domain_list_lock);
+		kfree(domain->domain_id);
+		kfree(domain);
+		return;
+	}
+	mutex_unlock(&erofs_domain_list_lock);
+}
+
+static int erofs_fscache_register_volume(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	char *domain_id = sbi->domain_id;
+	struct fscache_volume *volume;
+	char *name;
+	int ret = 0;
+
+	name = kasprintf(GFP_KERNEL, "erofs,%s",
+			 domain_id ? domain_id : sbi->fsid);
+	if (!name)
+		return -ENOMEM;
+
+	volume = fscache_acquire_volume(name, NULL, NULL, 0);
+	if (IS_ERR_OR_NULL(volume)) {
+		erofs_err(sb, "failed to register volume for %s", name);
+		ret = volume ? PTR_ERR(volume) : -EOPNOTSUPP;
+		volume = NULL;
+	}
+
+	sbi->volume = volume;
+	kfree(name);
+	return ret;
+}
+
+static int erofs_fscache_init_domain(struct super_block *sb)
+{
+	int err;
+	struct erofs_domain *domain;
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+
+	domain = kzalloc(sizeof(struct erofs_domain), GFP_KERNEL);
+	if (!domain)
+		return -ENOMEM;
+
+	domain->domain_id = kstrdup(sbi->domain_id, GFP_KERNEL);
+	if (!domain->domain_id) {
+		kfree(domain);
+		return -ENOMEM;
+	}
+
+	err = erofs_fscache_register_volume(sb);
+	if (err)
+		goto out;
+
+	if (!erofs_pseudo_mnt) {
+		erofs_pseudo_mnt = kern_mount(&erofs_fs_type);
+		if (IS_ERR(erofs_pseudo_mnt)) {
+			err = PTR_ERR(erofs_pseudo_mnt);
+			goto out;
+		}
+	}
+
+	domain->volume = sbi->volume;
+	refcount_set(&domain->ref, 1);
+	list_add(&domain->list, &erofs_domain_list);
+	sbi->domain = domain;
+	return 0;
+out:
+	kfree(domain->domain_id);
+	kfree(domain);
+	return err;
+}
+
+static int erofs_fscache_register_domain(struct super_block *sb)
+{
+	int err;
+	struct erofs_domain *domain;
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+
+	mutex_lock(&erofs_domain_list_lock);
+	list_for_each_entry(domain, &erofs_domain_list, list) {
+		if (!strcmp(domain->domain_id, sbi->domain_id)) {
+			sbi->domain = domain;
+			sbi->volume = domain->volume;
+			refcount_inc(&domain->ref);
+			mutex_unlock(&erofs_domain_list_lock);
+			return 0;
+		}
+	}
+	err = erofs_fscache_init_domain(sb);
+	mutex_unlock(&erofs_domain_list_lock);
+	return err;
+}
+
+static struct erofs_fscache *erofs_fscache_acquire_cookie(struct super_block *sb,
+						char *name, unsigned int flags)
+{
+	struct fscache_volume *volume = EROFS_SB(sb)->volume;
+	struct erofs_fscache *ctx;
+	struct fscache_cookie *cookie;
+	struct super_block *isb;
+	struct inode *inode;
+	int ret;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return ERR_PTR(-ENOMEM);
+	INIT_LIST_HEAD(&ctx->node);
+	refcount_set(&ctx->ref, 1);
+
+	cookie = fscache_acquire_cookie(volume, FSCACHE_ADV_WANT_CACHE_SIZE,
+					name, strlen(name), NULL, 0, 0);
+	if (!cookie) {
+		erofs_err(sb, "failed to get cookie for %s", name);
+		ret = -EINVAL;
+		goto err;
+	}
+	fscache_use_cookie(cookie, false);
+
+	/*
+	 * Allocate anonymous inode in global pseudo mount for shareable blobs,
+	 * so that they are accessible among erofs fs instances.
+	 */
+	isb = flags & EROFS_REG_COOKIE_SHARE ? erofs_pseudo_mnt->mnt_sb : sb;
+	inode = new_inode(isb);
+	if (!inode) {
+		erofs_err(sb, "failed to get anon inode for %s", name);
+		ret = -ENOMEM;
+		goto err_cookie;
+	}
+
+	inode->i_size = OFFSET_MAX;
+	inode->i_mapping->a_ops = &erofs_fscache_meta_aops;
+	mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);
+	inode->i_blkbits = EROFS_SB(sb)->blkszbits;
+	inode->i_private = ctx;
+
+	ctx->cookie = cookie;
+	ctx->inode = inode;
+	return ctx;
+
+err_cookie:
+	fscache_unuse_cookie(cookie, NULL, NULL);
+	fscache_relinquish_cookie(cookie, false);
+err:
+	kfree(ctx);
+	return ERR_PTR(ret);
+}
+
+static void erofs_fscache_relinquish_cookie(struct erofs_fscache *ctx)
+{
+	fscache_unuse_cookie(ctx->cookie, NULL, NULL);
+	fscache_relinquish_cookie(ctx->cookie, false);
+	iput(ctx->inode);
+	kfree(ctx->name);
+	kfree(ctx);
+}
+
+static struct erofs_fscache *erofs_domain_init_cookie(struct super_block *sb,
+						char *name, unsigned int flags)
+{
+	struct erofs_fscache *ctx;
+	struct erofs_domain *domain = EROFS_SB(sb)->domain;
+
+	ctx = erofs_fscache_acquire_cookie(sb, name, flags);
+	if (IS_ERR(ctx))
+		return ctx;
+
+	ctx->name = kstrdup(name, GFP_KERNEL);
+	if (!ctx->name) {
+		erofs_fscache_relinquish_cookie(ctx);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	refcount_inc(&domain->ref);
+	ctx->domain = domain;
+	list_add(&ctx->node, &erofs_domain_cookies_list);
+	return ctx;
+}
+
+static struct erofs_fscache *erofs_domain_register_cookie(struct super_block *sb,
+						char *name, unsigned int flags)
+{
+	struct erofs_fscache *ctx;
+	struct erofs_domain *domain = EROFS_SB(sb)->domain;
+
+	flags |= EROFS_REG_COOKIE_SHARE;
+	mutex_lock(&erofs_domain_cookies_lock);
+	list_for_each_entry(ctx, &erofs_domain_cookies_list, node) {
+		if (ctx->domain != domain || strcmp(ctx->name, name))
+			continue;
+		if (!(flags & EROFS_REG_COOKIE_NEED_NOEXIST)) {
+			refcount_inc(&ctx->ref);
+		} else {
+			erofs_err(sb, "%s already exists in domain %s", name,
+				  domain->domain_id);
+			ctx = ERR_PTR(-EEXIST);
+		}
+		mutex_unlock(&erofs_domain_cookies_lock);
+		return ctx;
+	}
+	ctx = erofs_domain_init_cookie(sb, name, flags);
+	mutex_unlock(&erofs_domain_cookies_lock);
+	return ctx;
+}
+
+struct erofs_fscache *erofs_fscache_register_cookie(struct super_block *sb,
+						    char *name,
+						    unsigned int flags)
+{
+	if (EROFS_SB(sb)->domain_id)
+		return erofs_domain_register_cookie(sb, name, flags);
+	return erofs_fscache_acquire_cookie(sb, name, flags);
+}
+
+void erofs_fscache_unregister_cookie(struct erofs_fscache *ctx)
+{
+	struct erofs_domain *domain = NULL;
+
+	if (!ctx)
+		return;
+	if (!ctx->domain)
+		return erofs_fscache_relinquish_cookie(ctx);
+
+	mutex_lock(&erofs_domain_cookies_lock);
+	if (refcount_dec_and_test(&ctx->ref)) {
+		domain = ctx->domain;
+		list_del(&ctx->node);
+		erofs_fscache_relinquish_cookie(ctx);
+	}
+	mutex_unlock(&erofs_domain_cookies_lock);
+	if (domain)
+		erofs_fscache_domain_put(domain);
+}
+
+int erofs_fscache_register_fs(struct super_block *sb)
+{
+	int ret;
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct erofs_fscache *fscache;
+	unsigned int flags = 0;
+
+	if (sbi->domain_id)
+		ret = erofs_fscache_register_domain(sb);
+	else
+		ret = erofs_fscache_register_volume(sb);
+	if (ret)
+		return ret;
+
+	/*
+	 * When shared domain is enabled, using NEED_NOEXIST to guarantee
+	 * the primary data blob (aka fsid) is unique in the shared domain.
+	 *
+	 * For non-shared-domain case, fscache_acquire_volume() invoked by
+	 * erofs_fscache_register_volume() has already guaranteed
+	 * the uniqueness of primary data blob.
+	 *
+	 * Acquired domain/volume will be relinquished in kill_sb() on error.
+	 */
+	if (sbi->domain_id)
+		flags |= EROFS_REG_COOKIE_NEED_NOEXIST;
+	fscache = erofs_fscache_register_cookie(sb, sbi->fsid, flags);
+	if (IS_ERR(fscache))
+		return PTR_ERR(fscache);
+
+	sbi->s_fscache = fscache;
+	return 0;
+}
+
+void erofs_fscache_unregister_fs(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+
+	erofs_fscache_unregister_cookie(sbi->s_fscache);
+
+	if (sbi->domain)
+		erofs_fscache_domain_put(sbi->domain);
+	else
+		fscache_relinquish_volume(sbi->volume, NULL, false);
+
+	sbi->s_fscache = NULL;
+	sbi->volume = NULL;
+	sbi->domain = NULL;
+}
diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c
new file mode 100644
index 0000000000..edc8ec7581
--- /dev/null
+++ b/fs/erofs/inode.c
@@ -0,0 +1,395 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2021, Alibaba Cloud
+ */
+#include "xattr.h"
+
+#include <trace/events/erofs.h>
+
+static void *erofs_read_inode(struct erofs_buf *buf,
+			      struct inode *inode, unsigned int *ofs)
+{
+	struct super_block *sb = inode->i_sb;
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct erofs_inode *vi = EROFS_I(inode);
+	const erofs_off_t inode_loc = erofs_iloc(inode);
+
+	erofs_blk_t blkaddr, nblks = 0;
+	void *kaddr;
+	struct erofs_inode_compact *dic;
+	struct erofs_inode_extended *die, *copied = NULL;
+	unsigned int ifmt;
+	int err;
+
+	blkaddr = erofs_blknr(sb, inode_loc);
+	*ofs = erofs_blkoff(sb, inode_loc);
+
+	kaddr = erofs_read_metabuf(buf, sb, blkaddr, EROFS_KMAP);
+	if (IS_ERR(kaddr)) {
+		erofs_err(sb, "failed to get inode (nid: %llu) page, err %ld",
+			  vi->nid, PTR_ERR(kaddr));
+		return kaddr;
+	}
+
+	dic = kaddr + *ofs;
+	ifmt = le16_to_cpu(dic->i_format);
+
+	if (ifmt & ~EROFS_I_ALL) {
+		erofs_err(inode->i_sb, "unsupported i_format %u of nid %llu",
+			  ifmt, vi->nid);
+		err = -EOPNOTSUPP;
+		goto err_out;
+	}
+
+	vi->datalayout = erofs_inode_datalayout(ifmt);
+	if (vi->datalayout >= EROFS_INODE_DATALAYOUT_MAX) {
+		erofs_err(inode->i_sb, "unsupported datalayout %u of nid %llu",
+			  vi->datalayout, vi->nid);
+		err = -EOPNOTSUPP;
+		goto err_out;
+	}
+
+	switch (erofs_inode_version(ifmt)) {
+	case EROFS_INODE_LAYOUT_EXTENDED:
+		vi->inode_isize = sizeof(struct erofs_inode_extended);
+		/* check if the extended inode acrosses block boundary */
+		if (*ofs + vi->inode_isize <= sb->s_blocksize) {
+			*ofs += vi->inode_isize;
+			die = (struct erofs_inode_extended *)dic;
+		} else {
+			const unsigned int gotten = sb->s_blocksize - *ofs;
+
+			copied = kmalloc(vi->inode_isize, GFP_NOFS);
+			if (!copied) {
+				err = -ENOMEM;
+				goto err_out;
+			}
+			memcpy(copied, dic, gotten);
+			kaddr = erofs_read_metabuf(buf, sb, blkaddr + 1,
+						   EROFS_KMAP);
+			if (IS_ERR(kaddr)) {
+				erofs_err(sb, "failed to get inode payload block (nid: %llu), err %ld",
+					  vi->nid, PTR_ERR(kaddr));
+				kfree(copied);
+				return kaddr;
+			}
+			*ofs = vi->inode_isize - gotten;
+			memcpy((u8 *)copied + gotten, kaddr, *ofs);
+			die = copied;
+		}
+		vi->xattr_isize = erofs_xattr_ibody_size(die->i_xattr_icount);
+
+		inode->i_mode = le16_to_cpu(die->i_mode);
+		switch (inode->i_mode & S_IFMT) {
+		case S_IFREG:
+		case S_IFDIR:
+		case S_IFLNK:
+			vi->raw_blkaddr = le32_to_cpu(die->i_u.raw_blkaddr);
+			break;
+		case S_IFCHR:
+		case S_IFBLK:
+			inode->i_rdev =
+				new_decode_dev(le32_to_cpu(die->i_u.rdev));
+			break;
+		case S_IFIFO:
+		case S_IFSOCK:
+			inode->i_rdev = 0;
+			break;
+		default:
+			goto bogusimode;
+		}
+		i_uid_write(inode, le32_to_cpu(die->i_uid));
+		i_gid_write(inode, le32_to_cpu(die->i_gid));
+		set_nlink(inode, le32_to_cpu(die->i_nlink));
+
+		/* extended inode has its own timestamp */
+		inode_set_ctime(inode, le64_to_cpu(die->i_mtime),
+				le32_to_cpu(die->i_mtime_nsec));
+
+		inode->i_size = le64_to_cpu(die->i_size);
+
+		/* total blocks for compressed files */
+		if (erofs_inode_is_data_compressed(vi->datalayout))
+			nblks = le32_to_cpu(die->i_u.compressed_blocks);
+		else if (vi->datalayout == EROFS_INODE_CHUNK_BASED)
+			/* fill chunked inode summary info */
+			vi->chunkformat = le16_to_cpu(die->i_u.c.format);
+		kfree(copied);
+		copied = NULL;
+		break;
+	case EROFS_INODE_LAYOUT_COMPACT:
+		vi->inode_isize = sizeof(struct erofs_inode_compact);
+		*ofs += vi->inode_isize;
+		vi->xattr_isize = erofs_xattr_ibody_size(dic->i_xattr_icount);
+
+		inode->i_mode = le16_to_cpu(dic->i_mode);
+		switch (inode->i_mode & S_IFMT) {
+		case S_IFREG:
+		case S_IFDIR:
+		case S_IFLNK:
+			vi->raw_blkaddr = le32_to_cpu(dic->i_u.raw_blkaddr);
+			break;
+		case S_IFCHR:
+		case S_IFBLK:
+			inode->i_rdev =
+				new_decode_dev(le32_to_cpu(dic->i_u.rdev));
+			break;
+		case S_IFIFO:
+		case S_IFSOCK:
+			inode->i_rdev = 0;
+			break;
+		default:
+			goto bogusimode;
+		}
+		i_uid_write(inode, le16_to_cpu(dic->i_uid));
+		i_gid_write(inode, le16_to_cpu(dic->i_gid));
+		set_nlink(inode, le16_to_cpu(dic->i_nlink));
+
+		/* use build time for compact inodes */
+		inode_set_ctime(inode, sbi->build_time, sbi->build_time_nsec);
+
+		inode->i_size = le32_to_cpu(dic->i_size);
+		if (erofs_inode_is_data_compressed(vi->datalayout))
+			nblks = le32_to_cpu(dic->i_u.compressed_blocks);
+		else if (vi->datalayout == EROFS_INODE_CHUNK_BASED)
+			vi->chunkformat = le16_to_cpu(dic->i_u.c.format);
+		break;
+	default:
+		erofs_err(inode->i_sb,
+			  "unsupported on-disk inode version %u of nid %llu",
+			  erofs_inode_version(ifmt), vi->nid);
+		err = -EOPNOTSUPP;
+		goto err_out;
+	}
+
+	if (vi->datalayout == EROFS_INODE_CHUNK_BASED) {
+		if (vi->chunkformat & ~EROFS_CHUNK_FORMAT_ALL) {
+			erofs_err(inode->i_sb,
+				  "unsupported chunk format %x of nid %llu",
+				  vi->chunkformat, vi->nid);
+			err = -EOPNOTSUPP;
+			goto err_out;
+		}
+		vi->chunkbits = sb->s_blocksize_bits +
+			(vi->chunkformat & EROFS_CHUNK_FORMAT_BLKBITS_MASK);
+	}
+	inode->i_mtime = inode->i_atime = inode_get_ctime(inode);
+
+	inode->i_flags &= ~S_DAX;
+	if (test_opt(&sbi->opt, DAX_ALWAYS) && S_ISREG(inode->i_mode) &&
+	    (vi->datalayout == EROFS_INODE_FLAT_PLAIN ||
+	     vi->datalayout == EROFS_INODE_CHUNK_BASED))
+		inode->i_flags |= S_DAX;
+
+	if (!nblks)
+		/* measure inode.i_blocks as generic filesystems */
+		inode->i_blocks = round_up(inode->i_size, sb->s_blocksize) >> 9;
+	else
+		inode->i_blocks = nblks << (sb->s_blocksize_bits - 9);
+	return kaddr;
+
+bogusimode:
+	erofs_err(inode->i_sb, "bogus i_mode (%o) @ nid %llu",
+		  inode->i_mode, vi->nid);
+	err = -EFSCORRUPTED;
+err_out:
+	DBG_BUGON(1);
+	kfree(copied);
+	erofs_put_metabuf(buf);
+	return ERR_PTR(err);
+}
+
+static int erofs_fill_symlink(struct inode *inode, void *kaddr,
+			      unsigned int m_pofs)
+{
+	struct erofs_inode *vi = EROFS_I(inode);
+	unsigned int bsz = i_blocksize(inode);
+	char *lnk;
+
+	/* if it cannot be handled with fast symlink scheme */
+	if (vi->datalayout != EROFS_INODE_FLAT_INLINE ||
+	    inode->i_size >= bsz || inode->i_size < 0) {
+		inode->i_op = &erofs_symlink_iops;
+		return 0;
+	}
+
+	lnk = kmalloc(inode->i_size + 1, GFP_KERNEL);
+	if (!lnk)
+		return -ENOMEM;
+
+	m_pofs += vi->xattr_isize;
+	/* inline symlink data shouldn't cross block boundary */
+	if (m_pofs + inode->i_size > bsz) {
+		kfree(lnk);
+		erofs_err(inode->i_sb,
+			  "inline data cross block boundary @ nid %llu",
+			  vi->nid);
+		DBG_BUGON(1);
+		return -EFSCORRUPTED;
+	}
+	memcpy(lnk, kaddr + m_pofs, inode->i_size);
+	lnk[inode->i_size] = '\0';
+
+	inode->i_link = lnk;
+	inode->i_op = &erofs_fast_symlink_iops;
+	return 0;
+}
+
+static int erofs_fill_inode(struct inode *inode)
+{
+	struct erofs_inode *vi = EROFS_I(inode);
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	void *kaddr;
+	unsigned int ofs;
+	int err = 0;
+
+	trace_erofs_fill_inode(inode);
+
+	/* read inode base data from disk */
+	kaddr = erofs_read_inode(&buf, inode, &ofs);
+	if (IS_ERR(kaddr))
+		return PTR_ERR(kaddr);
+
+	/* setup the new inode */
+	switch (inode->i_mode & S_IFMT) {
+	case S_IFREG:
+		inode->i_op = &erofs_generic_iops;
+		if (erofs_inode_is_data_compressed(vi->datalayout))
+			inode->i_fop = &generic_ro_fops;
+		else
+			inode->i_fop = &erofs_file_fops;
+		break;
+	case S_IFDIR:
+		inode->i_op = &erofs_dir_iops;
+		inode->i_fop = &erofs_dir_fops;
+		inode_nohighmem(inode);
+		break;
+	case S_IFLNK:
+		err = erofs_fill_symlink(inode, kaddr, ofs);
+		if (err)
+			goto out_unlock;
+		inode_nohighmem(inode);
+		break;
+	case S_IFCHR:
+	case S_IFBLK:
+	case S_IFIFO:
+	case S_IFSOCK:
+		inode->i_op = &erofs_generic_iops;
+		init_special_inode(inode, inode->i_mode, inode->i_rdev);
+		goto out_unlock;
+	default:
+		err = -EFSCORRUPTED;
+		goto out_unlock;
+	}
+
+	if (erofs_inode_is_data_compressed(vi->datalayout)) {
+#ifdef CONFIG_EROFS_FS_ZIP
+		if (!erofs_is_fscache_mode(inode->i_sb) &&
+		    inode->i_sb->s_blocksize_bits == PAGE_SHIFT) {
+			inode->i_mapping->a_ops = &z_erofs_aops;
+			err = 0;
+			goto out_unlock;
+		}
+#endif
+		err = -EOPNOTSUPP;
+		goto out_unlock;
+	}
+	inode->i_mapping->a_ops = &erofs_raw_access_aops;
+	mapping_set_large_folios(inode->i_mapping);
+#ifdef CONFIG_EROFS_FS_ONDEMAND
+	if (erofs_is_fscache_mode(inode->i_sb))
+		inode->i_mapping->a_ops = &erofs_fscache_access_aops;
+#endif
+
+out_unlock:
+	erofs_put_metabuf(&buf);
+	return err;
+}
+
+/*
+ * ino_t is 32-bits on 32-bit arch. We have to squash the 64-bit value down
+ * so that it will fit.
+ */
+static ino_t erofs_squash_ino(erofs_nid_t nid)
+{
+	ino_t ino = (ino_t)nid;
+
+	if (sizeof(ino_t) < sizeof(erofs_nid_t))
+		ino ^= nid >> (sizeof(erofs_nid_t) - sizeof(ino_t)) * 8;
+	return ino;
+}
+
+static int erofs_iget5_eq(struct inode *inode, void *opaque)
+{
+	return EROFS_I(inode)->nid == *(erofs_nid_t *)opaque;
+}
+
+static int erofs_iget5_set(struct inode *inode, void *opaque)
+{
+	const erofs_nid_t nid = *(erofs_nid_t *)opaque;
+
+	inode->i_ino = erofs_squash_ino(nid);
+	EROFS_I(inode)->nid = nid;
+	return 0;
+}
+
+struct inode *erofs_iget(struct super_block *sb, erofs_nid_t nid)
+{
+	struct inode *inode;
+
+	inode = iget5_locked(sb, erofs_squash_ino(nid), erofs_iget5_eq,
+			     erofs_iget5_set, &nid);
+	if (!inode)
+		return ERR_PTR(-ENOMEM);
+
+	if (inode->i_state & I_NEW) {
+		int err = erofs_fill_inode(inode);
+
+		if (err) {
+			iget_failed(inode);
+			return ERR_PTR(err);
+		}
+		unlock_new_inode(inode);
+	}
+	return inode;
+}
+
+int erofs_getattr(struct mnt_idmap *idmap, const struct path *path,
+		  struct kstat *stat, u32 request_mask,
+		  unsigned int query_flags)
+{
+	struct inode *const inode = d_inode(path->dentry);
+
+	if (erofs_inode_is_data_compressed(EROFS_I(inode)->datalayout))
+		stat->attributes |= STATX_ATTR_COMPRESSED;
+
+	stat->attributes |= STATX_ATTR_IMMUTABLE;
+	stat->attributes_mask |= (STATX_ATTR_COMPRESSED |
+				  STATX_ATTR_IMMUTABLE);
+
+	generic_fillattr(idmap, request_mask, inode, stat);
+	return 0;
+}
+
+const struct inode_operations erofs_generic_iops = {
+	.getattr = erofs_getattr,
+	.listxattr = erofs_listxattr,
+	.get_inode_acl = erofs_get_acl,
+	.fiemap = erofs_fiemap,
+};
+
+const struct inode_operations erofs_symlink_iops = {
+	.get_link = page_get_link,
+	.getattr = erofs_getattr,
+	.listxattr = erofs_listxattr,
+	.get_inode_acl = erofs_get_acl,
+};
+
+const struct inode_operations erofs_fast_symlink_iops = {
+	.get_link = simple_get_link,
+	.getattr = erofs_getattr,
+	.listxattr = erofs_listxattr,
+	.get_inode_acl = erofs_get_acl,
+};
diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h
new file mode 100644
index 0000000000..d8de61350d
--- /dev/null
+++ b/fs/erofs/internal.h
@@ -0,0 +1,537 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2021, Alibaba Cloud
+ */
+#ifndef __EROFS_INTERNAL_H
+#define __EROFS_INTERNAL_H
+
+#include <linux/fs.h>
+#include <linux/dcache.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/bio.h>
+#include <linux/magic.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/iomap.h>
+#include "erofs_fs.h"
+
+/* redefine pr_fmt "erofs: " */
+#undef pr_fmt
+#define pr_fmt(fmt) "erofs: " fmt
+
+__printf(3, 4) void _erofs_err(struct super_block *sb,
+			       const char *function, const char *fmt, ...);
+#define erofs_err(sb, fmt, ...)	\
+	_erofs_err(sb, __func__, fmt "\n", ##__VA_ARGS__)
+__printf(3, 4) void _erofs_info(struct super_block *sb,
+			       const char *function, const char *fmt, ...);
+#define erofs_info(sb, fmt, ...) \
+	_erofs_info(sb, __func__, fmt "\n", ##__VA_ARGS__)
+#ifdef CONFIG_EROFS_FS_DEBUG
+#define DBG_BUGON               BUG_ON
+#else
+#define DBG_BUGON(x)            ((void)(x))
+#endif	/* !CONFIG_EROFS_FS_DEBUG */
+
+/* EROFS_SUPER_MAGIC_V1 to represent the whole file system */
+#define EROFS_SUPER_MAGIC   EROFS_SUPER_MAGIC_V1
+
+typedef u64 erofs_nid_t;
+typedef u64 erofs_off_t;
+/* data type for filesystem-wide blocks number */
+typedef u32 erofs_blk_t;
+
+struct erofs_device_info {
+	char *path;
+	struct erofs_fscache *fscache;
+	struct block_device *bdev;
+	struct dax_device *dax_dev;
+	u64 dax_part_off;
+
+	u32 blocks;
+	u32 mapped_blkaddr;
+};
+
+enum {
+	EROFS_SYNC_DECOMPRESS_AUTO,
+	EROFS_SYNC_DECOMPRESS_FORCE_ON,
+	EROFS_SYNC_DECOMPRESS_FORCE_OFF
+};
+
+struct erofs_mount_opts {
+#ifdef CONFIG_EROFS_FS_ZIP
+	/* current strategy of how to use managed cache */
+	unsigned char cache_strategy;
+	/* strategy of sync decompression (0 - auto, 1 - force on, 2 - force off) */
+	unsigned int sync_decompress;
+
+	/* threshold for decompression synchronously */
+	unsigned int max_sync_decompress_pages;
+#endif
+	unsigned int mount_opt;
+};
+
+struct erofs_dev_context {
+	struct idr tree;
+	struct rw_semaphore rwsem;
+
+	unsigned int extra_devices;
+	bool flatdev;
+};
+
+struct erofs_fs_context {
+	struct erofs_mount_opts opt;
+	struct erofs_dev_context *devs;
+	char *fsid;
+	char *domain_id;
+};
+
+/* all filesystem-wide lz4 configurations */
+struct erofs_sb_lz4_info {
+	/* # of pages needed for EROFS lz4 rolling decompression */
+	u16 max_distance_pages;
+	/* maximum possible blocks for pclusters in the filesystem */
+	u16 max_pclusterblks;
+};
+
+struct erofs_domain {
+	refcount_t ref;
+	struct list_head list;
+	struct fscache_volume *volume;
+	char *domain_id;
+};
+
+struct erofs_fscache {
+	struct fscache_cookie *cookie;
+	struct inode *inode;	/* anonymous inode for the blob */
+
+	/* used for share domain mode */
+	struct erofs_domain *domain;
+	struct list_head node;
+	refcount_t ref;
+	char *name;
+};
+
+struct erofs_xattr_prefix_item {
+	struct erofs_xattr_long_prefix *prefix;
+	u8 infix_len;
+};
+
+struct erofs_sb_info {
+	struct erofs_mount_opts opt;	/* options */
+#ifdef CONFIG_EROFS_FS_ZIP
+	/* list for all registered superblocks, mainly for shrinker */
+	struct list_head list;
+	struct mutex umount_mutex;
+
+	/* managed XArray arranged in physical block number */
+	struct xarray managed_pslots;
+
+	unsigned int shrinker_run_no;
+	u16 available_compr_algs;
+
+	/* pseudo inode to manage cached pages */
+	struct inode *managed_cache;
+
+	struct erofs_sb_lz4_info lz4;
+#endif	/* CONFIG_EROFS_FS_ZIP */
+	struct inode *packed_inode;
+	struct erofs_dev_context *devs;
+	struct dax_device *dax_dev;
+	u64 dax_part_off;
+	u64 total_blocks;
+	u32 primarydevice_blocks;
+
+	u32 meta_blkaddr;
+#ifdef CONFIG_EROFS_FS_XATTR
+	u32 xattr_blkaddr;
+	u32 xattr_prefix_start;
+	u8 xattr_prefix_count;
+	struct erofs_xattr_prefix_item *xattr_prefixes;
+	unsigned int xattr_filter_reserved;
+#endif
+	u16 device_id_mask;	/* valid bits of device id to be used */
+
+	unsigned char islotbits;	/* inode slot unit size in bit shift */
+	unsigned char blkszbits;	/* filesystem block size in bit shift */
+
+	u32 sb_size;			/* total superblock size */
+	u32 build_time_nsec;
+	u64 build_time;
+
+	/* what we really care is nid, rather than ino.. */
+	erofs_nid_t root_nid;
+	erofs_nid_t packed_nid;
+	/* used for statfs, f_files - f_favail */
+	u64 inos;
+
+	u8 uuid[16];                    /* 128-bit uuid for volume */
+	u8 volume_name[16];             /* volume name */
+	u32 feature_compat;
+	u32 feature_incompat;
+
+	/* sysfs support */
+	struct kobject s_kobj;		/* /sys/fs/erofs/<devname> */
+	struct completion s_kobj_unregister;
+
+	/* fscache support */
+	struct fscache_volume *volume;
+	struct erofs_fscache *s_fscache;
+	struct erofs_domain *domain;
+	char *fsid;
+	char *domain_id;
+};
+
+#define EROFS_SB(sb) ((struct erofs_sb_info *)(sb)->s_fs_info)
+#define EROFS_I_SB(inode) ((struct erofs_sb_info *)(inode)->i_sb->s_fs_info)
+
+/* Mount flags set via mount options or defaults */
+#define EROFS_MOUNT_XATTR_USER		0x00000010
+#define EROFS_MOUNT_POSIX_ACL		0x00000020
+#define EROFS_MOUNT_DAX_ALWAYS		0x00000040
+#define EROFS_MOUNT_DAX_NEVER		0x00000080
+
+#define clear_opt(opt, option)	((opt)->mount_opt &= ~EROFS_MOUNT_##option)
+#define set_opt(opt, option)	((opt)->mount_opt |= EROFS_MOUNT_##option)
+#define test_opt(opt, option)	((opt)->mount_opt & EROFS_MOUNT_##option)
+
+static inline bool erofs_is_fscache_mode(struct super_block *sb)
+{
+	return IS_ENABLED(CONFIG_EROFS_FS_ONDEMAND) && !sb->s_bdev;
+}
+
+enum {
+	EROFS_ZIP_CACHE_DISABLED,
+	EROFS_ZIP_CACHE_READAHEAD,
+	EROFS_ZIP_CACHE_READAROUND
+};
+
+/* basic unit of the workstation of a super_block */
+struct erofs_workgroup {
+	pgoff_t index;
+	struct lockref lockref;
+};
+
+enum erofs_kmap_type {
+	EROFS_NO_KMAP,		/* don't map the buffer */
+	EROFS_KMAP,		/* use kmap_local_page() to map the buffer */
+};
+
+struct erofs_buf {
+	struct inode *inode;
+	struct page *page;
+	void *base;
+	enum erofs_kmap_type kmap_type;
+};
+#define __EROFS_BUF_INITIALIZER	((struct erofs_buf){ .page = NULL })
+
+#define ROOT_NID(sb)		((sb)->root_nid)
+
+#define erofs_blknr(sb, addr)	((addr) >> (sb)->s_blocksize_bits)
+#define erofs_blkoff(sb, addr)	((addr) & ((sb)->s_blocksize - 1))
+#define erofs_pos(sb, blk)	((erofs_off_t)(blk) << (sb)->s_blocksize_bits)
+#define erofs_iblks(i)	(round_up((i)->i_size, i_blocksize(i)) >> (i)->i_blkbits)
+
+#define EROFS_FEATURE_FUNCS(name, compat, feature) \
+static inline bool erofs_sb_has_##name(struct erofs_sb_info *sbi) \
+{ \
+	return sbi->feature_##compat & EROFS_FEATURE_##feature; \
+}
+
+EROFS_FEATURE_FUNCS(zero_padding, incompat, INCOMPAT_ZERO_PADDING)
+EROFS_FEATURE_FUNCS(compr_cfgs, incompat, INCOMPAT_COMPR_CFGS)
+EROFS_FEATURE_FUNCS(big_pcluster, incompat, INCOMPAT_BIG_PCLUSTER)
+EROFS_FEATURE_FUNCS(chunked_file, incompat, INCOMPAT_CHUNKED_FILE)
+EROFS_FEATURE_FUNCS(device_table, incompat, INCOMPAT_DEVICE_TABLE)
+EROFS_FEATURE_FUNCS(compr_head2, incompat, INCOMPAT_COMPR_HEAD2)
+EROFS_FEATURE_FUNCS(ztailpacking, incompat, INCOMPAT_ZTAILPACKING)
+EROFS_FEATURE_FUNCS(fragments, incompat, INCOMPAT_FRAGMENTS)
+EROFS_FEATURE_FUNCS(dedupe, incompat, INCOMPAT_DEDUPE)
+EROFS_FEATURE_FUNCS(xattr_prefixes, incompat, INCOMPAT_XATTR_PREFIXES)
+EROFS_FEATURE_FUNCS(sb_chksum, compat, COMPAT_SB_CHKSUM)
+EROFS_FEATURE_FUNCS(xattr_filter, compat, COMPAT_XATTR_FILTER)
+
+/* atomic flag definitions */
+#define EROFS_I_EA_INITED_BIT	0
+#define EROFS_I_Z_INITED_BIT	1
+
+/* bitlock definitions (arranged in reverse order) */
+#define EROFS_I_BL_XATTR_BIT	(BITS_PER_LONG - 1)
+#define EROFS_I_BL_Z_BIT	(BITS_PER_LONG - 2)
+
+struct erofs_inode {
+	erofs_nid_t nid;
+
+	/* atomic flags (including bitlocks) */
+	unsigned long flags;
+
+	unsigned char datalayout;
+	unsigned char inode_isize;
+	unsigned int xattr_isize;
+
+	unsigned int xattr_name_filter;
+	unsigned int xattr_shared_count;
+	unsigned int *xattr_shared_xattrs;
+
+	union {
+		erofs_blk_t raw_blkaddr;
+		struct {
+			unsigned short	chunkformat;
+			unsigned char	chunkbits;
+		};
+#ifdef CONFIG_EROFS_FS_ZIP
+		struct {
+			unsigned short z_advise;
+			unsigned char  z_algorithmtype[2];
+			unsigned char  z_logical_clusterbits;
+			unsigned long  z_tailextent_headlcn;
+			union {
+				struct {
+					erofs_off_t    z_idataoff;
+					unsigned short z_idata_size;
+				};
+				erofs_off_t z_fragmentoff;
+			};
+		};
+#endif	/* CONFIG_EROFS_FS_ZIP */
+	};
+	/* the corresponding vfs inode */
+	struct inode vfs_inode;
+};
+
+#define EROFS_I(ptr)	container_of(ptr, struct erofs_inode, vfs_inode)
+
+static inline erofs_off_t erofs_iloc(struct inode *inode)
+{
+	struct erofs_sb_info *sbi = EROFS_I_SB(inode);
+
+	return erofs_pos(inode->i_sb, sbi->meta_blkaddr) +
+		(EROFS_I(inode)->nid << sbi->islotbits);
+}
+
+static inline unsigned int erofs_inode_version(unsigned int ifmt)
+{
+	return (ifmt >> EROFS_I_VERSION_BIT) & EROFS_I_VERSION_MASK;
+}
+
+static inline unsigned int erofs_inode_datalayout(unsigned int ifmt)
+{
+	return (ifmt >> EROFS_I_DATALAYOUT_BIT) & EROFS_I_DATALAYOUT_MASK;
+}
+
+/*
+ * Different from grab_cache_page_nowait(), reclaiming is never triggered
+ * when allocating new pages.
+ */
+static inline
+struct page *erofs_grab_cache_page_nowait(struct address_space *mapping,
+					  pgoff_t index)
+{
+	return pagecache_get_page(mapping, index,
+			FGP_LOCK|FGP_CREAT|FGP_NOFS|FGP_NOWAIT,
+			readahead_gfp_mask(mapping) & ~__GFP_RECLAIM);
+}
+
+/* Has a disk mapping */
+#define EROFS_MAP_MAPPED	0x0001
+/* Located in metadata (could be copied from bd_inode) */
+#define EROFS_MAP_META		0x0002
+/* The extent is encoded */
+#define EROFS_MAP_ENCODED	0x0004
+/* The length of extent is full */
+#define EROFS_MAP_FULL_MAPPED	0x0008
+/* Located in the special packed inode */
+#define EROFS_MAP_FRAGMENT	0x0010
+/* The extent refers to partial decompressed data */
+#define EROFS_MAP_PARTIAL_REF	0x0020
+
+struct erofs_map_blocks {
+	struct erofs_buf buf;
+
+	erofs_off_t m_pa, m_la;
+	u64 m_plen, m_llen;
+
+	unsigned short m_deviceid;
+	char m_algorithmformat;
+	unsigned int m_flags;
+};
+
+/*
+ * Used to get the exact decompressed length, e.g. fiemap (consider lookback
+ * approach instead if possible since it's more metadata lightweight.)
+ */
+#define EROFS_GET_BLOCKS_FIEMAP		0x0001
+/* Used to map the whole extent if non-negligible data is requested for LZMA */
+#define EROFS_GET_BLOCKS_READMORE	0x0002
+/* Used to map tail extent for tailpacking inline or fragment pcluster */
+#define EROFS_GET_BLOCKS_FINDTAIL	0x0004
+
+enum {
+	Z_EROFS_COMPRESSION_SHIFTED = Z_EROFS_COMPRESSION_MAX,
+	Z_EROFS_COMPRESSION_INTERLACED,
+	Z_EROFS_COMPRESSION_RUNTIME_MAX
+};
+
+struct erofs_map_dev {
+	struct erofs_fscache *m_fscache;
+	struct block_device *m_bdev;
+	struct dax_device *m_daxdev;
+	u64 m_dax_part_off;
+
+	erofs_off_t m_pa;
+	unsigned int m_deviceid;
+};
+
+extern struct file_system_type erofs_fs_type;
+extern const struct super_operations erofs_sops;
+
+extern const struct address_space_operations erofs_raw_access_aops;
+extern const struct address_space_operations z_erofs_aops;
+extern const struct address_space_operations erofs_fscache_access_aops;
+
+extern const struct inode_operations erofs_generic_iops;
+extern const struct inode_operations erofs_symlink_iops;
+extern const struct inode_operations erofs_fast_symlink_iops;
+extern const struct inode_operations erofs_dir_iops;
+
+extern const struct file_operations erofs_file_fops;
+extern const struct file_operations erofs_dir_fops;
+
+extern const struct iomap_ops z_erofs_iomap_report_ops;
+
+/* flags for erofs_fscache_register_cookie() */
+#define EROFS_REG_COOKIE_SHARE		0x0001
+#define EROFS_REG_COOKIE_NEED_NOEXIST	0x0002
+
+void *erofs_read_metadata(struct super_block *sb, struct erofs_buf *buf,
+			  erofs_off_t *offset, int *lengthp);
+void erofs_unmap_metabuf(struct erofs_buf *buf);
+void erofs_put_metabuf(struct erofs_buf *buf);
+void *erofs_bread(struct erofs_buf *buf, erofs_blk_t blkaddr,
+		  enum erofs_kmap_type type);
+void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb);
+void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb,
+			 erofs_blk_t blkaddr, enum erofs_kmap_type type);
+int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *dev);
+int erofs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		 u64 start, u64 len);
+int erofs_map_blocks(struct inode *inode, struct erofs_map_blocks *map);
+struct inode *erofs_iget(struct super_block *sb, erofs_nid_t nid);
+int erofs_getattr(struct mnt_idmap *idmap, const struct path *path,
+		  struct kstat *stat, u32 request_mask,
+		  unsigned int query_flags);
+int erofs_namei(struct inode *dir, const struct qstr *name,
+		erofs_nid_t *nid, unsigned int *d_type);
+
+static inline void *erofs_vm_map_ram(struct page **pages, unsigned int count)
+{
+	int retried = 0;
+
+	while (1) {
+		void *p = vm_map_ram(pages, count, -1);
+
+		/* retry two more times (totally 3 times) */
+		if (p || ++retried >= 3)
+			return p;
+		vm_unmap_aliases();
+	}
+	return NULL;
+}
+
+int erofs_register_sysfs(struct super_block *sb);
+void erofs_unregister_sysfs(struct super_block *sb);
+int __init erofs_init_sysfs(void);
+void erofs_exit_sysfs(void);
+
+struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp);
+static inline void erofs_pagepool_add(struct page **pagepool, struct page *page)
+{
+	set_page_private(page, (unsigned long)*pagepool);
+	*pagepool = page;
+}
+void erofs_release_pages(struct page **pagepool);
+
+#ifdef CONFIG_EROFS_FS_ZIP
+void erofs_workgroup_put(struct erofs_workgroup *grp);
+struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb,
+					     pgoff_t index);
+struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb,
+					       struct erofs_workgroup *grp);
+void erofs_workgroup_free_rcu(struct erofs_workgroup *grp);
+void erofs_shrinker_register(struct super_block *sb);
+void erofs_shrinker_unregister(struct super_block *sb);
+int __init erofs_init_shrinker(void);
+void erofs_exit_shrinker(void);
+int __init z_erofs_init_zip_subsystem(void);
+void z_erofs_exit_zip_subsystem(void);
+int erofs_try_to_free_all_cached_pages(struct erofs_sb_info *sbi,
+				       struct erofs_workgroup *egrp);
+int z_erofs_map_blocks_iter(struct inode *inode, struct erofs_map_blocks *map,
+			    int flags);
+void *erofs_get_pcpubuf(unsigned int requiredpages);
+void erofs_put_pcpubuf(void *ptr);
+int erofs_pcpubuf_growsize(unsigned int nrpages);
+void __init erofs_pcpubuf_init(void);
+void erofs_pcpubuf_exit(void);
+int erofs_init_managed_cache(struct super_block *sb);
+int z_erofs_parse_cfgs(struct super_block *sb, struct erofs_super_block *dsb);
+#else
+static inline void erofs_shrinker_register(struct super_block *sb) {}
+static inline void erofs_shrinker_unregister(struct super_block *sb) {}
+static inline int erofs_init_shrinker(void) { return 0; }
+static inline void erofs_exit_shrinker(void) {}
+static inline int z_erofs_init_zip_subsystem(void) { return 0; }
+static inline void z_erofs_exit_zip_subsystem(void) {}
+static inline void erofs_pcpubuf_init(void) {}
+static inline void erofs_pcpubuf_exit(void) {}
+static inline int erofs_init_managed_cache(struct super_block *sb) { return 0; }
+#endif	/* !CONFIG_EROFS_FS_ZIP */
+
+#ifdef CONFIG_EROFS_FS_ZIP_LZMA
+int __init z_erofs_lzma_init(void);
+void z_erofs_lzma_exit(void);
+#else
+static inline int z_erofs_lzma_init(void) { return 0; }
+static inline int z_erofs_lzma_exit(void) { return 0; }
+#endif	/* !CONFIG_EROFS_FS_ZIP_LZMA */
+
+#ifdef CONFIG_EROFS_FS_ZIP_DEFLATE
+int __init z_erofs_deflate_init(void);
+void z_erofs_deflate_exit(void);
+#else
+static inline int z_erofs_deflate_init(void) { return 0; }
+static inline int z_erofs_deflate_exit(void) { return 0; }
+#endif	/* !CONFIG_EROFS_FS_ZIP_DEFLATE */
+
+#ifdef CONFIG_EROFS_FS_ONDEMAND
+int erofs_fscache_register_fs(struct super_block *sb);
+void erofs_fscache_unregister_fs(struct super_block *sb);
+
+struct erofs_fscache *erofs_fscache_register_cookie(struct super_block *sb,
+					char *name, unsigned int flags);
+void erofs_fscache_unregister_cookie(struct erofs_fscache *fscache);
+#else
+static inline int erofs_fscache_register_fs(struct super_block *sb)
+{
+	return -EOPNOTSUPP;
+}
+static inline void erofs_fscache_unregister_fs(struct super_block *sb) {}
+
+static inline
+struct erofs_fscache *erofs_fscache_register_cookie(struct super_block *sb,
+					char *name, unsigned int flags)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
+static inline void erofs_fscache_unregister_cookie(struct erofs_fscache *fscache)
+{
+}
+#endif
+
+#define EFSCORRUPTED    EUCLEAN         /* Filesystem is corrupted */
+
+#endif	/* __EROFS_INTERNAL_H */
diff --git a/fs/erofs/namei.c b/fs/erofs/namei.c
new file mode 100644
index 0000000000..d4f631d39f
--- /dev/null
+++ b/fs/erofs/namei.c
@@ -0,0 +1,224 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2022, Alibaba Cloud
+ */
+#include "xattr.h"
+#include <trace/events/erofs.h>
+
+struct erofs_qstr {
+	const unsigned char *name;
+	const unsigned char *end;
+};
+
+/* based on the end of qn is accurate and it must have the trailing '\0' */
+static inline int erofs_dirnamecmp(const struct erofs_qstr *qn,
+				   const struct erofs_qstr *qd,
+				   unsigned int *matched)
+{
+	unsigned int i = *matched;
+
+	/*
+	 * on-disk error, let's only BUG_ON in the debugging mode.
+	 * otherwise, it will return 1 to just skip the invalid name
+	 * and go on (in consideration of the lookup performance).
+	 */
+	DBG_BUGON(qd->name > qd->end);
+
+	/* qd could not have trailing '\0' */
+	/* However it is absolutely safe if < qd->end */
+	while (qd->name + i < qd->end && qd->name[i] != '\0') {
+		if (qn->name[i] != qd->name[i]) {
+			*matched = i;
+			return qn->name[i] > qd->name[i] ? 1 : -1;
+		}
+		++i;
+	}
+	*matched = i;
+	/* See comments in __d_alloc on the terminating NUL character */
+	return qn->name[i] == '\0' ? 0 : 1;
+}
+
+#define nameoff_from_disk(off, sz)	(le16_to_cpu(off) & ((sz) - 1))
+
+static struct erofs_dirent *find_target_dirent(struct erofs_qstr *name,
+					       u8 *data,
+					       unsigned int dirblksize,
+					       const int ndirents)
+{
+	int head, back;
+	unsigned int startprfx, endprfx;
+	struct erofs_dirent *const de = (struct erofs_dirent *)data;
+
+	/* since the 1st dirent has been evaluated previously */
+	head = 1;
+	back = ndirents - 1;
+	startprfx = endprfx = 0;
+
+	while (head <= back) {
+		const int mid = head + (back - head) / 2;
+		const int nameoff = nameoff_from_disk(de[mid].nameoff,
+						      dirblksize);
+		unsigned int matched = min(startprfx, endprfx);
+		struct erofs_qstr dname = {
+			.name = data + nameoff,
+			.end = mid >= ndirents - 1 ?
+				data + dirblksize :
+				data + nameoff_from_disk(de[mid + 1].nameoff,
+							 dirblksize)
+		};
+
+		/* string comparison without already matched prefix */
+		int ret = erofs_dirnamecmp(name, &dname, &matched);
+
+		if (!ret) {
+			return de + mid;
+		} else if (ret > 0) {
+			head = mid + 1;
+			startprfx = matched;
+		} else {
+			back = mid - 1;
+			endprfx = matched;
+		}
+	}
+
+	return ERR_PTR(-ENOENT);
+}
+
+static void *erofs_find_target_block(struct erofs_buf *target,
+		struct inode *dir, struct erofs_qstr *name, int *_ndirents)
+{
+	unsigned int bsz = i_blocksize(dir);
+	int head = 0, back = erofs_iblks(dir) - 1;
+	unsigned int startprfx = 0, endprfx = 0;
+	void *candidate = ERR_PTR(-ENOENT);
+
+	while (head <= back) {
+		const int mid = head + (back - head) / 2;
+		struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+		struct erofs_dirent *de;
+
+		buf.inode = dir;
+		de = erofs_bread(&buf, mid, EROFS_KMAP);
+		if (!IS_ERR(de)) {
+			const int nameoff = nameoff_from_disk(de->nameoff, bsz);
+			const int ndirents = nameoff / sizeof(*de);
+			int diff;
+			unsigned int matched;
+			struct erofs_qstr dname;
+
+			if (!ndirents) {
+				erofs_put_metabuf(&buf);
+				erofs_err(dir->i_sb,
+					  "corrupted dir block %d @ nid %llu",
+					  mid, EROFS_I(dir)->nid);
+				DBG_BUGON(1);
+				de = ERR_PTR(-EFSCORRUPTED);
+				goto out;
+			}
+
+			matched = min(startprfx, endprfx);
+
+			dname.name = (u8 *)de + nameoff;
+			if (ndirents == 1)
+				dname.end = (u8 *)de + bsz;
+			else
+				dname.end = (u8 *)de +
+					nameoff_from_disk(de[1].nameoff, bsz);
+
+			/* string comparison without already matched prefix */
+			diff = erofs_dirnamecmp(name, &dname, &matched);
+
+			if (!diff) {
+				*_ndirents = 0;
+				goto out;
+			} else if (diff > 0) {
+				head = mid + 1;
+				startprfx = matched;
+
+				if (!IS_ERR(candidate))
+					erofs_put_metabuf(target);
+				*target = buf;
+				candidate = de;
+				*_ndirents = ndirents;
+			} else {
+				erofs_put_metabuf(&buf);
+
+				back = mid - 1;
+				endprfx = matched;
+			}
+			continue;
+		}
+out:		/* free if the candidate is valid */
+		if (!IS_ERR(candidate))
+			erofs_put_metabuf(target);
+		return de;
+	}
+	return candidate;
+}
+
+int erofs_namei(struct inode *dir, const struct qstr *name, erofs_nid_t *nid,
+		unsigned int *d_type)
+{
+	int ndirents;
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	struct erofs_dirent *de;
+	struct erofs_qstr qn;
+
+	if (!dir->i_size)
+		return -ENOENT;
+
+	qn.name = name->name;
+	qn.end = name->name + name->len;
+	buf.inode = dir;
+
+	ndirents = 0;
+	de = erofs_find_target_block(&buf, dir, &qn, &ndirents);
+	if (IS_ERR(de))
+		return PTR_ERR(de);
+
+	if (ndirents)
+		de = find_target_dirent(&qn, (u8 *)de, i_blocksize(dir),
+					ndirents);
+
+	if (!IS_ERR(de)) {
+		*nid = le64_to_cpu(de->nid);
+		*d_type = de->file_type;
+	}
+	erofs_put_metabuf(&buf);
+	return PTR_ERR_OR_ZERO(de);
+}
+
+static struct dentry *erofs_lookup(struct inode *dir, struct dentry *dentry,
+				   unsigned int flags)
+{
+	int err;
+	erofs_nid_t nid;
+	unsigned int d_type;
+	struct inode *inode;
+
+	trace_erofs_lookup(dir, dentry, flags);
+
+	if (dentry->d_name.len > EROFS_NAME_LEN)
+		return ERR_PTR(-ENAMETOOLONG);
+
+	err = erofs_namei(dir, &dentry->d_name, &nid, &d_type);
+
+	if (err == -ENOENT)
+		/* negative dentry */
+		inode = NULL;
+	else if (err)
+		inode = ERR_PTR(err);
+	else
+		inode = erofs_iget(dir->i_sb, nid);
+	return d_splice_alias(inode, dentry);
+}
+
+const struct inode_operations erofs_dir_iops = {
+	.lookup = erofs_lookup,
+	.getattr = erofs_getattr,
+	.listxattr = erofs_listxattr,
+	.get_inode_acl = erofs_get_acl,
+	.fiemap = erofs_fiemap,
+};
diff --git a/fs/erofs/pcpubuf.c b/fs/erofs/pcpubuf.c
new file mode 100644
index 0000000000..c7a4b1d770
--- /dev/null
+++ b/fs/erofs/pcpubuf.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) Gao Xiang <xiang@kernel.org>
+ *
+ * For low-latency decompression algorithms (e.g. lz4), reserve consecutive
+ * per-CPU virtual memory (in pages) in advance to store such inplace I/O
+ * data if inplace decompression is failed (due to unmet inplace margin for
+ * example).
+ */
+#include "internal.h"
+
+struct erofs_pcpubuf {
+	raw_spinlock_t lock;
+	void *ptr;
+	struct page **pages;
+	unsigned int nrpages;
+};
+
+static DEFINE_PER_CPU(struct erofs_pcpubuf, erofs_pcb);
+
+void *erofs_get_pcpubuf(unsigned int requiredpages)
+	__acquires(pcb->lock)
+{
+	struct erofs_pcpubuf *pcb = &get_cpu_var(erofs_pcb);
+
+	raw_spin_lock(&pcb->lock);
+	/* check if the per-CPU buffer is too small */
+	if (requiredpages > pcb->nrpages) {
+		raw_spin_unlock(&pcb->lock);
+		put_cpu_var(erofs_pcb);
+		/* (for sparse checker) pretend pcb->lock is still taken */
+		__acquire(pcb->lock);
+		return NULL;
+	}
+	return pcb->ptr;
+}
+
+void erofs_put_pcpubuf(void *ptr) __releases(pcb->lock)
+{
+	struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, smp_processor_id());
+
+	DBG_BUGON(pcb->ptr != ptr);
+	raw_spin_unlock(&pcb->lock);
+	put_cpu_var(erofs_pcb);
+}
+
+/* the next step: support per-CPU page buffers hotplug */
+int erofs_pcpubuf_growsize(unsigned int nrpages)
+{
+	static DEFINE_MUTEX(pcb_resize_mutex);
+	static unsigned int pcb_nrpages;
+	struct page *pagepool = NULL;
+	int delta, cpu, ret, i;
+
+	mutex_lock(&pcb_resize_mutex);
+	delta = nrpages - pcb_nrpages;
+	ret = 0;
+	/* avoid shrinking pcpubuf, since no idea how many fses rely on */
+	if (delta <= 0)
+		goto out;
+
+	for_each_possible_cpu(cpu) {
+		struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, cpu);
+		struct page **pages, **oldpages;
+		void *ptr, *old_ptr;
+
+		pages = kmalloc_array(nrpages, sizeof(*pages), GFP_KERNEL);
+		if (!pages) {
+			ret = -ENOMEM;
+			break;
+		}
+
+		for (i = 0; i < nrpages; ++i) {
+			pages[i] = erofs_allocpage(&pagepool, GFP_KERNEL);
+			if (!pages[i]) {
+				ret = -ENOMEM;
+				oldpages = pages;
+				goto free_pagearray;
+			}
+		}
+		ptr = vmap(pages, nrpages, VM_MAP, PAGE_KERNEL);
+		if (!ptr) {
+			ret = -ENOMEM;
+			oldpages = pages;
+			goto free_pagearray;
+		}
+		raw_spin_lock(&pcb->lock);
+		old_ptr = pcb->ptr;
+		pcb->ptr = ptr;
+		oldpages = pcb->pages;
+		pcb->pages = pages;
+		i = pcb->nrpages;
+		pcb->nrpages = nrpages;
+		raw_spin_unlock(&pcb->lock);
+
+		if (!oldpages) {
+			DBG_BUGON(old_ptr);
+			continue;
+		}
+
+		if (old_ptr)
+			vunmap(old_ptr);
+free_pagearray:
+		while (i)
+			erofs_pagepool_add(&pagepool, oldpages[--i]);
+		kfree(oldpages);
+		if (ret)
+			break;
+	}
+	pcb_nrpages = nrpages;
+	erofs_release_pages(&pagepool);
+out:
+	mutex_unlock(&pcb_resize_mutex);
+	return ret;
+}
+
+void __init erofs_pcpubuf_init(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, cpu);
+
+		raw_spin_lock_init(&pcb->lock);
+	}
+}
+
+void erofs_pcpubuf_exit(void)
+{
+	int cpu, i;
+
+	for_each_possible_cpu(cpu) {
+		struct erofs_pcpubuf *pcb = &per_cpu(erofs_pcb, cpu);
+
+		if (pcb->ptr) {
+			vunmap(pcb->ptr);
+			pcb->ptr = NULL;
+		}
+		if (!pcb->pages)
+			continue;
+
+		for (i = 0; i < pcb->nrpages; ++i)
+			if (pcb->pages[i])
+				put_page(pcb->pages[i]);
+		kfree(pcb->pages);
+		pcb->pages = NULL;
+	}
+}
diff --git a/fs/erofs/super.c b/fs/erofs/super.c
new file mode 100644
index 0000000000..cc44fb2e00
--- /dev/null
+++ b/fs/erofs/super.c
@@ -0,0 +1,1021 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2021, Alibaba Cloud
+ */
+#include <linux/module.h>
+#include <linux/statfs.h>
+#include <linux/parser.h>
+#include <linux/seq_file.h>
+#include <linux/crc32c.h>
+#include <linux/fs_context.h>
+#include <linux/fs_parser.h>
+#include <linux/dax.h>
+#include <linux/exportfs.h>
+#include "xattr.h"
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/erofs.h>
+
+static struct kmem_cache *erofs_inode_cachep __read_mostly;
+
+void _erofs_err(struct super_block *sb, const char *func, const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	pr_err("(device %s): %s: %pV", sb->s_id, func, &vaf);
+	va_end(args);
+}
+
+void _erofs_info(struct super_block *sb, const char *func, const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	pr_info("(device %s): %pV", sb->s_id, &vaf);
+	va_end(args);
+}
+
+static int erofs_superblock_csum_verify(struct super_block *sb, void *sbdata)
+{
+	size_t len = 1 << EROFS_SB(sb)->blkszbits;
+	struct erofs_super_block *dsb;
+	u32 expected_crc, crc;
+
+	if (len > EROFS_SUPER_OFFSET)
+		len -= EROFS_SUPER_OFFSET;
+
+	dsb = kmemdup(sbdata + EROFS_SUPER_OFFSET, len, GFP_KERNEL);
+	if (!dsb)
+		return -ENOMEM;
+
+	expected_crc = le32_to_cpu(dsb->checksum);
+	dsb->checksum = 0;
+	/* to allow for x86 boot sectors and other oddities. */
+	crc = crc32c(~0, dsb, len);
+	kfree(dsb);
+
+	if (crc != expected_crc) {
+		erofs_err(sb, "invalid checksum 0x%08x, 0x%08x expected",
+			  crc, expected_crc);
+		return -EBADMSG;
+	}
+	return 0;
+}
+
+static void erofs_inode_init_once(void *ptr)
+{
+	struct erofs_inode *vi = ptr;
+
+	inode_init_once(&vi->vfs_inode);
+}
+
+static struct inode *erofs_alloc_inode(struct super_block *sb)
+{
+	struct erofs_inode *vi =
+		alloc_inode_sb(sb, erofs_inode_cachep, GFP_KERNEL);
+
+	if (!vi)
+		return NULL;
+
+	/* zero out everything except vfs_inode */
+	memset(vi, 0, offsetof(struct erofs_inode, vfs_inode));
+	return &vi->vfs_inode;
+}
+
+static void erofs_free_inode(struct inode *inode)
+{
+	struct erofs_inode *vi = EROFS_I(inode);
+
+	if (inode->i_op == &erofs_fast_symlink_iops)
+		kfree(inode->i_link);
+	kfree(vi->xattr_shared_xattrs);
+	kmem_cache_free(erofs_inode_cachep, vi);
+}
+
+static bool check_layout_compatibility(struct super_block *sb,
+				       struct erofs_super_block *dsb)
+{
+	const unsigned int feature = le32_to_cpu(dsb->feature_incompat);
+
+	EROFS_SB(sb)->feature_incompat = feature;
+
+	/* check if current kernel meets all mandatory requirements */
+	if (feature & (~EROFS_ALL_FEATURE_INCOMPAT)) {
+		erofs_err(sb, "unidentified incompatible feature %x, please upgrade kernel",
+			   feature & ~EROFS_ALL_FEATURE_INCOMPAT);
+		return false;
+	}
+	return true;
+}
+
+/* read variable-sized metadata, offset will be aligned by 4-byte */
+void *erofs_read_metadata(struct super_block *sb, struct erofs_buf *buf,
+			  erofs_off_t *offset, int *lengthp)
+{
+	u8 *buffer, *ptr;
+	int len, i, cnt;
+
+	*offset = round_up(*offset, 4);
+	ptr = erofs_bread(buf, erofs_blknr(sb, *offset), EROFS_KMAP);
+	if (IS_ERR(ptr))
+		return ptr;
+
+	len = le16_to_cpu(*(__le16 *)&ptr[erofs_blkoff(sb, *offset)]);
+	if (!len)
+		len = U16_MAX + 1;
+	buffer = kmalloc(len, GFP_KERNEL);
+	if (!buffer)
+		return ERR_PTR(-ENOMEM);
+	*offset += sizeof(__le16);
+	*lengthp = len;
+
+	for (i = 0; i < len; i += cnt) {
+		cnt = min_t(int, sb->s_blocksize - erofs_blkoff(sb, *offset),
+			    len - i);
+		ptr = erofs_bread(buf, erofs_blknr(sb, *offset), EROFS_KMAP);
+		if (IS_ERR(ptr)) {
+			kfree(buffer);
+			return ptr;
+		}
+		memcpy(buffer + i, ptr + erofs_blkoff(sb, *offset), cnt);
+		*offset += cnt;
+	}
+	return buffer;
+}
+
+#ifndef CONFIG_EROFS_FS_ZIP
+static int z_erofs_parse_cfgs(struct super_block *sb,
+			      struct erofs_super_block *dsb)
+{
+	if (!dsb->u1.available_compr_algs)
+		return 0;
+
+	erofs_err(sb, "compression disabled, unable to mount compressed EROFS");
+	return -EOPNOTSUPP;
+}
+#endif
+
+static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb,
+			     struct erofs_device_info *dif, erofs_off_t *pos)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct erofs_fscache *fscache;
+	struct erofs_deviceslot *dis;
+	struct block_device *bdev;
+	void *ptr;
+
+	ptr = erofs_read_metabuf(buf, sb, erofs_blknr(sb, *pos), EROFS_KMAP);
+	if (IS_ERR(ptr))
+		return PTR_ERR(ptr);
+	dis = ptr + erofs_blkoff(sb, *pos);
+
+	if (!sbi->devs->flatdev && !dif->path) {
+		if (!dis->tag[0]) {
+			erofs_err(sb, "empty device tag @ pos %llu", *pos);
+			return -EINVAL;
+		}
+		dif->path = kmemdup_nul(dis->tag, sizeof(dis->tag), GFP_KERNEL);
+		if (!dif->path)
+			return -ENOMEM;
+	}
+
+	if (erofs_is_fscache_mode(sb)) {
+		fscache = erofs_fscache_register_cookie(sb, dif->path, 0);
+		if (IS_ERR(fscache))
+			return PTR_ERR(fscache);
+		dif->fscache = fscache;
+	} else if (!sbi->devs->flatdev) {
+		bdev = blkdev_get_by_path(dif->path, BLK_OPEN_READ, sb->s_type,
+					  NULL);
+		if (IS_ERR(bdev))
+			return PTR_ERR(bdev);
+		dif->bdev = bdev;
+		dif->dax_dev = fs_dax_get_by_bdev(bdev, &dif->dax_part_off,
+						  NULL, NULL);
+	}
+
+	dif->blocks = le32_to_cpu(dis->blocks);
+	dif->mapped_blkaddr = le32_to_cpu(dis->mapped_blkaddr);
+	sbi->total_blocks += dif->blocks;
+	*pos += EROFS_DEVT_SLOT_SIZE;
+	return 0;
+}
+
+static int erofs_scan_devices(struct super_block *sb,
+			      struct erofs_super_block *dsb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	unsigned int ondisk_extradevs;
+	erofs_off_t pos;
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	struct erofs_device_info *dif;
+	int id, err = 0;
+
+	sbi->total_blocks = sbi->primarydevice_blocks;
+	if (!erofs_sb_has_device_table(sbi))
+		ondisk_extradevs = 0;
+	else
+		ondisk_extradevs = le16_to_cpu(dsb->extra_devices);
+
+	if (sbi->devs->extra_devices &&
+	    ondisk_extradevs != sbi->devs->extra_devices) {
+		erofs_err(sb, "extra devices don't match (ondisk %u, given %u)",
+			  ondisk_extradevs, sbi->devs->extra_devices);
+		return -EINVAL;
+	}
+	if (!ondisk_extradevs)
+		return 0;
+
+	if (!sbi->devs->extra_devices && !erofs_is_fscache_mode(sb))
+		sbi->devs->flatdev = true;
+
+	sbi->device_id_mask = roundup_pow_of_two(ondisk_extradevs + 1) - 1;
+	pos = le16_to_cpu(dsb->devt_slotoff) * EROFS_DEVT_SLOT_SIZE;
+	down_read(&sbi->devs->rwsem);
+	if (sbi->devs->extra_devices) {
+		idr_for_each_entry(&sbi->devs->tree, dif, id) {
+			err = erofs_init_device(&buf, sb, dif, &pos);
+			if (err)
+				break;
+		}
+	} else {
+		for (id = 0; id < ondisk_extradevs; id++) {
+			dif = kzalloc(sizeof(*dif), GFP_KERNEL);
+			if (!dif) {
+				err = -ENOMEM;
+				break;
+			}
+
+			err = idr_alloc(&sbi->devs->tree, dif, 0, 0, GFP_KERNEL);
+			if (err < 0) {
+				kfree(dif);
+				break;
+			}
+			++sbi->devs->extra_devices;
+
+			err = erofs_init_device(&buf, sb, dif, &pos);
+			if (err)
+				break;
+		}
+	}
+	up_read(&sbi->devs->rwsem);
+	erofs_put_metabuf(&buf);
+	return err;
+}
+
+static int erofs_read_superblock(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi;
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	struct erofs_super_block *dsb;
+	void *data;
+	int ret;
+
+	data = erofs_read_metabuf(&buf, sb, 0, EROFS_KMAP);
+	if (IS_ERR(data)) {
+		erofs_err(sb, "cannot read erofs superblock");
+		return PTR_ERR(data);
+	}
+
+	sbi = EROFS_SB(sb);
+	dsb = (struct erofs_super_block *)(data + EROFS_SUPER_OFFSET);
+
+	ret = -EINVAL;
+	if (le32_to_cpu(dsb->magic) != EROFS_SUPER_MAGIC_V1) {
+		erofs_err(sb, "cannot find valid erofs superblock");
+		goto out;
+	}
+
+	sbi->blkszbits  = dsb->blkszbits;
+	if (sbi->blkszbits < 9 || sbi->blkszbits > PAGE_SHIFT) {
+		erofs_err(sb, "blkszbits %u isn't supported", sbi->blkszbits);
+		goto out;
+	}
+	if (dsb->dirblkbits) {
+		erofs_err(sb, "dirblkbits %u isn't supported", dsb->dirblkbits);
+		goto out;
+	}
+
+	sbi->feature_compat = le32_to_cpu(dsb->feature_compat);
+	if (erofs_sb_has_sb_chksum(sbi)) {
+		ret = erofs_superblock_csum_verify(sb, data);
+		if (ret)
+			goto out;
+	}
+
+	ret = -EINVAL;
+	if (!check_layout_compatibility(sb, dsb))
+		goto out;
+
+	sbi->sb_size = 128 + dsb->sb_extslots * EROFS_SB_EXTSLOT_SIZE;
+	if (sbi->sb_size > PAGE_SIZE - EROFS_SUPER_OFFSET) {
+		erofs_err(sb, "invalid sb_extslots %u (more than a fs block)",
+			  sbi->sb_size);
+		goto out;
+	}
+	sbi->primarydevice_blocks = le32_to_cpu(dsb->blocks);
+	sbi->meta_blkaddr = le32_to_cpu(dsb->meta_blkaddr);
+#ifdef CONFIG_EROFS_FS_XATTR
+	sbi->xattr_blkaddr = le32_to_cpu(dsb->xattr_blkaddr);
+	sbi->xattr_prefix_start = le32_to_cpu(dsb->xattr_prefix_start);
+	sbi->xattr_prefix_count = dsb->xattr_prefix_count;
+	sbi->xattr_filter_reserved = dsb->xattr_filter_reserved;
+#endif
+	sbi->islotbits = ilog2(sizeof(struct erofs_inode_compact));
+	sbi->root_nid = le16_to_cpu(dsb->root_nid);
+	sbi->packed_nid = le64_to_cpu(dsb->packed_nid);
+	sbi->inos = le64_to_cpu(dsb->inos);
+
+	sbi->build_time = le64_to_cpu(dsb->build_time);
+	sbi->build_time_nsec = le32_to_cpu(dsb->build_time_nsec);
+
+	memcpy(&sb->s_uuid, dsb->uuid, sizeof(dsb->uuid));
+
+	ret = strscpy(sbi->volume_name, dsb->volume_name,
+		      sizeof(dsb->volume_name));
+	if (ret < 0) {	/* -E2BIG */
+		erofs_err(sb, "bad volume name without NIL terminator");
+		ret = -EFSCORRUPTED;
+		goto out;
+	}
+
+	/* parse on-disk compression configurations */
+	ret = z_erofs_parse_cfgs(sb, dsb);
+	if (ret < 0)
+		goto out;
+
+	/* handle multiple devices */
+	ret = erofs_scan_devices(sb, dsb);
+
+	if (erofs_is_fscache_mode(sb))
+		erofs_info(sb, "EXPERIMENTAL fscache-based on-demand read feature in use. Use at your own risk!");
+out:
+	erofs_put_metabuf(&buf);
+	return ret;
+}
+
+static void erofs_default_options(struct erofs_fs_context *ctx)
+{
+#ifdef CONFIG_EROFS_FS_ZIP
+	ctx->opt.cache_strategy = EROFS_ZIP_CACHE_READAROUND;
+	ctx->opt.max_sync_decompress_pages = 3;
+	ctx->opt.sync_decompress = EROFS_SYNC_DECOMPRESS_AUTO;
+#endif
+#ifdef CONFIG_EROFS_FS_XATTR
+	set_opt(&ctx->opt, XATTR_USER);
+#endif
+#ifdef CONFIG_EROFS_FS_POSIX_ACL
+	set_opt(&ctx->opt, POSIX_ACL);
+#endif
+}
+
+enum {
+	Opt_user_xattr,
+	Opt_acl,
+	Opt_cache_strategy,
+	Opt_dax,
+	Opt_dax_enum,
+	Opt_device,
+	Opt_fsid,
+	Opt_domain_id,
+	Opt_err
+};
+
+static const struct constant_table erofs_param_cache_strategy[] = {
+	{"disabled",	EROFS_ZIP_CACHE_DISABLED},
+	{"readahead",	EROFS_ZIP_CACHE_READAHEAD},
+	{"readaround",	EROFS_ZIP_CACHE_READAROUND},
+	{}
+};
+
+static const struct constant_table erofs_dax_param_enums[] = {
+	{"always",	EROFS_MOUNT_DAX_ALWAYS},
+	{"never",	EROFS_MOUNT_DAX_NEVER},
+	{}
+};
+
+static const struct fs_parameter_spec erofs_fs_parameters[] = {
+	fsparam_flag_no("user_xattr",	Opt_user_xattr),
+	fsparam_flag_no("acl",		Opt_acl),
+	fsparam_enum("cache_strategy",	Opt_cache_strategy,
+		     erofs_param_cache_strategy),
+	fsparam_flag("dax",             Opt_dax),
+	fsparam_enum("dax",		Opt_dax_enum, erofs_dax_param_enums),
+	fsparam_string("device",	Opt_device),
+	fsparam_string("fsid",		Opt_fsid),
+	fsparam_string("domain_id",	Opt_domain_id),
+	{}
+};
+
+static bool erofs_fc_set_dax_mode(struct fs_context *fc, unsigned int mode)
+{
+#ifdef CONFIG_FS_DAX
+	struct erofs_fs_context *ctx = fc->fs_private;
+
+	switch (mode) {
+	case EROFS_MOUNT_DAX_ALWAYS:
+		warnfc(fc, "DAX enabled. Warning: EXPERIMENTAL, use at your own risk");
+		set_opt(&ctx->opt, DAX_ALWAYS);
+		clear_opt(&ctx->opt, DAX_NEVER);
+		return true;
+	case EROFS_MOUNT_DAX_NEVER:
+		set_opt(&ctx->opt, DAX_NEVER);
+		clear_opt(&ctx->opt, DAX_ALWAYS);
+		return true;
+	default:
+		DBG_BUGON(1);
+		return false;
+	}
+#else
+	errorfc(fc, "dax options not supported");
+	return false;
+#endif
+}
+
+static int erofs_fc_parse_param(struct fs_context *fc,
+				struct fs_parameter *param)
+{
+	struct erofs_fs_context *ctx = fc->fs_private;
+	struct fs_parse_result result;
+	struct erofs_device_info *dif;
+	int opt, ret;
+
+	opt = fs_parse(fc, erofs_fs_parameters, param, &result);
+	if (opt < 0)
+		return opt;
+
+	switch (opt) {
+	case Opt_user_xattr:
+#ifdef CONFIG_EROFS_FS_XATTR
+		if (result.boolean)
+			set_opt(&ctx->opt, XATTR_USER);
+		else
+			clear_opt(&ctx->opt, XATTR_USER);
+#else
+		errorfc(fc, "{,no}user_xattr options not supported");
+#endif
+		break;
+	case Opt_acl:
+#ifdef CONFIG_EROFS_FS_POSIX_ACL
+		if (result.boolean)
+			set_opt(&ctx->opt, POSIX_ACL);
+		else
+			clear_opt(&ctx->opt, POSIX_ACL);
+#else
+		errorfc(fc, "{,no}acl options not supported");
+#endif
+		break;
+	case Opt_cache_strategy:
+#ifdef CONFIG_EROFS_FS_ZIP
+		ctx->opt.cache_strategy = result.uint_32;
+#else
+		errorfc(fc, "compression not supported, cache_strategy ignored");
+#endif
+		break;
+	case Opt_dax:
+		if (!erofs_fc_set_dax_mode(fc, EROFS_MOUNT_DAX_ALWAYS))
+			return -EINVAL;
+		break;
+	case Opt_dax_enum:
+		if (!erofs_fc_set_dax_mode(fc, result.uint_32))
+			return -EINVAL;
+		break;
+	case Opt_device:
+		dif = kzalloc(sizeof(*dif), GFP_KERNEL);
+		if (!dif)
+			return -ENOMEM;
+		dif->path = kstrdup(param->string, GFP_KERNEL);
+		if (!dif->path) {
+			kfree(dif);
+			return -ENOMEM;
+		}
+		down_write(&ctx->devs->rwsem);
+		ret = idr_alloc(&ctx->devs->tree, dif, 0, 0, GFP_KERNEL);
+		up_write(&ctx->devs->rwsem);
+		if (ret < 0) {
+			kfree(dif->path);
+			kfree(dif);
+			return ret;
+		}
+		++ctx->devs->extra_devices;
+		break;
+#ifdef CONFIG_EROFS_FS_ONDEMAND
+	case Opt_fsid:
+		kfree(ctx->fsid);
+		ctx->fsid = kstrdup(param->string, GFP_KERNEL);
+		if (!ctx->fsid)
+			return -ENOMEM;
+		break;
+	case Opt_domain_id:
+		kfree(ctx->domain_id);
+		ctx->domain_id = kstrdup(param->string, GFP_KERNEL);
+		if (!ctx->domain_id)
+			return -ENOMEM;
+		break;
+#else
+	case Opt_fsid:
+	case Opt_domain_id:
+		errorfc(fc, "%s option not supported", erofs_fs_parameters[opt].name);
+		break;
+#endif
+	default:
+		return -ENOPARAM;
+	}
+	return 0;
+}
+
+static struct inode *erofs_nfs_get_inode(struct super_block *sb,
+					 u64 ino, u32 generation)
+{
+	return erofs_iget(sb, ino);
+}
+
+static struct dentry *erofs_fh_to_dentry(struct super_block *sb,
+		struct fid *fid, int fh_len, int fh_type)
+{
+	return generic_fh_to_dentry(sb, fid, fh_len, fh_type,
+				    erofs_nfs_get_inode);
+}
+
+static struct dentry *erofs_fh_to_parent(struct super_block *sb,
+		struct fid *fid, int fh_len, int fh_type)
+{
+	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
+				    erofs_nfs_get_inode);
+}
+
+static struct dentry *erofs_get_parent(struct dentry *child)
+{
+	erofs_nid_t nid;
+	unsigned int d_type;
+	int err;
+
+	err = erofs_namei(d_inode(child), &dotdot_name, &nid, &d_type);
+	if (err)
+		return ERR_PTR(err);
+	return d_obtain_alias(erofs_iget(child->d_sb, nid));
+}
+
+static const struct export_operations erofs_export_ops = {
+	.fh_to_dentry = erofs_fh_to_dentry,
+	.fh_to_parent = erofs_fh_to_parent,
+	.get_parent = erofs_get_parent,
+};
+
+static int erofs_fc_fill_pseudo_super(struct super_block *sb, struct fs_context *fc)
+{
+	static const struct tree_descr empty_descr = {""};
+
+	return simple_fill_super(sb, EROFS_SUPER_MAGIC, &empty_descr);
+}
+
+static int erofs_fc_fill_super(struct super_block *sb, struct fs_context *fc)
+{
+	struct inode *inode;
+	struct erofs_sb_info *sbi;
+	struct erofs_fs_context *ctx = fc->fs_private;
+	int err;
+
+	sb->s_magic = EROFS_SUPER_MAGIC;
+	sb->s_flags |= SB_RDONLY | SB_NOATIME;
+	sb->s_maxbytes = MAX_LFS_FILESIZE;
+	sb->s_op = &erofs_sops;
+
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
+	if (!sbi)
+		return -ENOMEM;
+
+	sb->s_fs_info = sbi;
+	sbi->opt = ctx->opt;
+	sbi->devs = ctx->devs;
+	ctx->devs = NULL;
+	sbi->fsid = ctx->fsid;
+	ctx->fsid = NULL;
+	sbi->domain_id = ctx->domain_id;
+	ctx->domain_id = NULL;
+
+	sbi->blkszbits = PAGE_SHIFT;
+	if (erofs_is_fscache_mode(sb)) {
+		sb->s_blocksize = PAGE_SIZE;
+		sb->s_blocksize_bits = PAGE_SHIFT;
+
+		err = erofs_fscache_register_fs(sb);
+		if (err)
+			return err;
+
+		err = super_setup_bdi(sb);
+		if (err)
+			return err;
+	} else {
+		if (!sb_set_blocksize(sb, PAGE_SIZE)) {
+			errorfc(fc, "failed to set initial blksize");
+			return -EINVAL;
+		}
+
+		sbi->dax_dev = fs_dax_get_by_bdev(sb->s_bdev,
+						  &sbi->dax_part_off,
+						  NULL, NULL);
+	}
+
+	err = erofs_read_superblock(sb);
+	if (err)
+		return err;
+
+	if (sb->s_blocksize_bits != sbi->blkszbits) {
+		if (erofs_is_fscache_mode(sb)) {
+			errorfc(fc, "unsupported blksize for fscache mode");
+			return -EINVAL;
+		}
+		if (!sb_set_blocksize(sb, 1 << sbi->blkszbits)) {
+			errorfc(fc, "failed to set erofs blksize");
+			return -EINVAL;
+		}
+	}
+
+	if (test_opt(&sbi->opt, DAX_ALWAYS)) {
+		if (!sbi->dax_dev) {
+			errorfc(fc, "DAX unsupported by block device. Turning off DAX.");
+			clear_opt(&sbi->opt, DAX_ALWAYS);
+		} else if (sbi->blkszbits != PAGE_SHIFT) {
+			errorfc(fc, "unsupported blocksize for DAX");
+			clear_opt(&sbi->opt, DAX_ALWAYS);
+		}
+	}
+
+	sb->s_time_gran = 1;
+	sb->s_xattr = erofs_xattr_handlers;
+	sb->s_export_op = &erofs_export_ops;
+
+	if (test_opt(&sbi->opt, POSIX_ACL))
+		sb->s_flags |= SB_POSIXACL;
+	else
+		sb->s_flags &= ~SB_POSIXACL;
+
+#ifdef CONFIG_EROFS_FS_ZIP
+	xa_init(&sbi->managed_pslots);
+#endif
+
+	inode = erofs_iget(sb, ROOT_NID(sbi));
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
+
+	if (!S_ISDIR(inode->i_mode)) {
+		erofs_err(sb, "rootino(nid %llu) is not a directory(i_mode %o)",
+			  ROOT_NID(sbi), inode->i_mode);
+		iput(inode);
+		return -EINVAL;
+	}
+
+	sb->s_root = d_make_root(inode);
+	if (!sb->s_root)
+		return -ENOMEM;
+
+	erofs_shrinker_register(sb);
+	if (erofs_sb_has_fragments(sbi) && sbi->packed_nid) {
+		sbi->packed_inode = erofs_iget(sb, sbi->packed_nid);
+		if (IS_ERR(sbi->packed_inode)) {
+			err = PTR_ERR(sbi->packed_inode);
+			sbi->packed_inode = NULL;
+			return err;
+		}
+	}
+	err = erofs_init_managed_cache(sb);
+	if (err)
+		return err;
+
+	err = erofs_xattr_prefixes_init(sb);
+	if (err)
+		return err;
+
+	err = erofs_register_sysfs(sb);
+	if (err)
+		return err;
+
+	erofs_info(sb, "mounted with root inode @ nid %llu.", ROOT_NID(sbi));
+	return 0;
+}
+
+static int erofs_fc_anon_get_tree(struct fs_context *fc)
+{
+	return get_tree_nodev(fc, erofs_fc_fill_pseudo_super);
+}
+
+static int erofs_fc_get_tree(struct fs_context *fc)
+{
+	struct erofs_fs_context *ctx = fc->fs_private;
+
+	if (IS_ENABLED(CONFIG_EROFS_FS_ONDEMAND) && ctx->fsid)
+		return get_tree_nodev(fc, erofs_fc_fill_super);
+
+	return get_tree_bdev(fc, erofs_fc_fill_super);
+}
+
+static int erofs_fc_reconfigure(struct fs_context *fc)
+{
+	struct super_block *sb = fc->root->d_sb;
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct erofs_fs_context *ctx = fc->fs_private;
+
+	DBG_BUGON(!sb_rdonly(sb));
+
+	if (ctx->fsid || ctx->domain_id)
+		erofs_info(sb, "ignoring reconfiguration for fsid|domain_id.");
+
+	if (test_opt(&ctx->opt, POSIX_ACL))
+		fc->sb_flags |= SB_POSIXACL;
+	else
+		fc->sb_flags &= ~SB_POSIXACL;
+
+	sbi->opt = ctx->opt;
+
+	fc->sb_flags |= SB_RDONLY;
+	return 0;
+}
+
+static int erofs_release_device_info(int id, void *ptr, void *data)
+{
+	struct erofs_device_info *dif = ptr;
+
+	fs_put_dax(dif->dax_dev, NULL);
+	if (dif->bdev)
+		blkdev_put(dif->bdev, &erofs_fs_type);
+	erofs_fscache_unregister_cookie(dif->fscache);
+	dif->fscache = NULL;
+	kfree(dif->path);
+	kfree(dif);
+	return 0;
+}
+
+static void erofs_free_dev_context(struct erofs_dev_context *devs)
+{
+	if (!devs)
+		return;
+	idr_for_each(&devs->tree, &erofs_release_device_info, NULL);
+	idr_destroy(&devs->tree);
+	kfree(devs);
+}
+
+static void erofs_fc_free(struct fs_context *fc)
+{
+	struct erofs_fs_context *ctx = fc->fs_private;
+
+	erofs_free_dev_context(ctx->devs);
+	kfree(ctx->fsid);
+	kfree(ctx->domain_id);
+	kfree(ctx);
+}
+
+static const struct fs_context_operations erofs_context_ops = {
+	.parse_param	= erofs_fc_parse_param,
+	.get_tree       = erofs_fc_get_tree,
+	.reconfigure    = erofs_fc_reconfigure,
+	.free		= erofs_fc_free,
+};
+
+static const struct fs_context_operations erofs_anon_context_ops = {
+	.get_tree       = erofs_fc_anon_get_tree,
+};
+
+static int erofs_init_fs_context(struct fs_context *fc)
+{
+	struct erofs_fs_context *ctx;
+
+	/* pseudo mount for anon inodes */
+	if (fc->sb_flags & SB_KERNMOUNT) {
+		fc->ops = &erofs_anon_context_ops;
+		return 0;
+	}
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+	ctx->devs = kzalloc(sizeof(struct erofs_dev_context), GFP_KERNEL);
+	if (!ctx->devs) {
+		kfree(ctx);
+		return -ENOMEM;
+	}
+	fc->fs_private = ctx;
+
+	idr_init(&ctx->devs->tree);
+	init_rwsem(&ctx->devs->rwsem);
+	erofs_default_options(ctx);
+	fc->ops = &erofs_context_ops;
+	return 0;
+}
+
+static void erofs_kill_sb(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi;
+
+	/* pseudo mount for anon inodes */
+	if (sb->s_flags & SB_KERNMOUNT) {
+		kill_anon_super(sb);
+		return;
+	}
+
+	if (erofs_is_fscache_mode(sb))
+		kill_anon_super(sb);
+	else
+		kill_block_super(sb);
+
+	sbi = EROFS_SB(sb);
+	if (!sbi)
+		return;
+
+	erofs_free_dev_context(sbi->devs);
+	fs_put_dax(sbi->dax_dev, NULL);
+	erofs_fscache_unregister_fs(sb);
+	kfree(sbi->fsid);
+	kfree(sbi->domain_id);
+	kfree(sbi);
+	sb->s_fs_info = NULL;
+}
+
+static void erofs_put_super(struct super_block *sb)
+{
+	struct erofs_sb_info *const sbi = EROFS_SB(sb);
+
+	DBG_BUGON(!sbi);
+
+	erofs_unregister_sysfs(sb);
+	erofs_shrinker_unregister(sb);
+	erofs_xattr_prefixes_cleanup(sb);
+#ifdef CONFIG_EROFS_FS_ZIP
+	iput(sbi->managed_cache);
+	sbi->managed_cache = NULL;
+#endif
+	iput(sbi->packed_inode);
+	sbi->packed_inode = NULL;
+	erofs_free_dev_context(sbi->devs);
+	sbi->devs = NULL;
+	erofs_fscache_unregister_fs(sb);
+}
+
+struct file_system_type erofs_fs_type = {
+	.owner          = THIS_MODULE,
+	.name           = "erofs",
+	.init_fs_context = erofs_init_fs_context,
+	.kill_sb        = erofs_kill_sb,
+	.fs_flags       = FS_REQUIRES_DEV | FS_ALLOW_IDMAP,
+};
+MODULE_ALIAS_FS("erofs");
+
+static int __init erofs_module_init(void)
+{
+	int err;
+
+	erofs_check_ondisk_layout_definitions();
+
+	erofs_inode_cachep = kmem_cache_create("erofs_inode",
+			sizeof(struct erofs_inode), 0,
+			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD | SLAB_ACCOUNT,
+			erofs_inode_init_once);
+	if (!erofs_inode_cachep)
+		return -ENOMEM;
+
+	err = erofs_init_shrinker();
+	if (err)
+		goto shrinker_err;
+
+	err = z_erofs_lzma_init();
+	if (err)
+		goto lzma_err;
+
+	err = z_erofs_deflate_init();
+	if (err)
+		goto deflate_err;
+
+	erofs_pcpubuf_init();
+	err = z_erofs_init_zip_subsystem();
+	if (err)
+		goto zip_err;
+
+	err = erofs_init_sysfs();
+	if (err)
+		goto sysfs_err;
+
+	err = register_filesystem(&erofs_fs_type);
+	if (err)
+		goto fs_err;
+
+	return 0;
+
+fs_err:
+	erofs_exit_sysfs();
+sysfs_err:
+	z_erofs_exit_zip_subsystem();
+zip_err:
+	z_erofs_deflate_exit();
+deflate_err:
+	z_erofs_lzma_exit();
+lzma_err:
+	erofs_exit_shrinker();
+shrinker_err:
+	kmem_cache_destroy(erofs_inode_cachep);
+	return err;
+}
+
+static void __exit erofs_module_exit(void)
+{
+	unregister_filesystem(&erofs_fs_type);
+
+	/* Ensure all RCU free inodes / pclusters are safe to be destroyed. */
+	rcu_barrier();
+
+	erofs_exit_sysfs();
+	z_erofs_exit_zip_subsystem();
+	z_erofs_deflate_exit();
+	z_erofs_lzma_exit();
+	erofs_exit_shrinker();
+	kmem_cache_destroy(erofs_inode_cachep);
+	erofs_pcpubuf_exit();
+}
+
+static int erofs_statfs(struct dentry *dentry, struct kstatfs *buf)
+{
+	struct super_block *sb = dentry->d_sb;
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	u64 id = 0;
+
+	if (!erofs_is_fscache_mode(sb))
+		id = huge_encode_dev(sb->s_bdev->bd_dev);
+
+	buf->f_type = sb->s_magic;
+	buf->f_bsize = sb->s_blocksize;
+	buf->f_blocks = sbi->total_blocks;
+	buf->f_bfree = buf->f_bavail = 0;
+
+	buf->f_files = ULLONG_MAX;
+	buf->f_ffree = ULLONG_MAX - sbi->inos;
+
+	buf->f_namelen = EROFS_NAME_LEN;
+
+	buf->f_fsid    = u64_to_fsid(id);
+	return 0;
+}
+
+static int erofs_show_options(struct seq_file *seq, struct dentry *root)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(root->d_sb);
+	struct erofs_mount_opts *opt = &sbi->opt;
+
+#ifdef CONFIG_EROFS_FS_XATTR
+	if (test_opt(opt, XATTR_USER))
+		seq_puts(seq, ",user_xattr");
+	else
+		seq_puts(seq, ",nouser_xattr");
+#endif
+#ifdef CONFIG_EROFS_FS_POSIX_ACL
+	if (test_opt(opt, POSIX_ACL))
+		seq_puts(seq, ",acl");
+	else
+		seq_puts(seq, ",noacl");
+#endif
+#ifdef CONFIG_EROFS_FS_ZIP
+	if (opt->cache_strategy == EROFS_ZIP_CACHE_DISABLED)
+		seq_puts(seq, ",cache_strategy=disabled");
+	else if (opt->cache_strategy == EROFS_ZIP_CACHE_READAHEAD)
+		seq_puts(seq, ",cache_strategy=readahead");
+	else if (opt->cache_strategy == EROFS_ZIP_CACHE_READAROUND)
+		seq_puts(seq, ",cache_strategy=readaround");
+#endif
+	if (test_opt(opt, DAX_ALWAYS))
+		seq_puts(seq, ",dax=always");
+	if (test_opt(opt, DAX_NEVER))
+		seq_puts(seq, ",dax=never");
+#ifdef CONFIG_EROFS_FS_ONDEMAND
+	if (sbi->fsid)
+		seq_printf(seq, ",fsid=%s", sbi->fsid);
+	if (sbi->domain_id)
+		seq_printf(seq, ",domain_id=%s", sbi->domain_id);
+#endif
+	return 0;
+}
+
+const struct super_operations erofs_sops = {
+	.put_super = erofs_put_super,
+	.alloc_inode = erofs_alloc_inode,
+	.free_inode = erofs_free_inode,
+	.statfs = erofs_statfs,
+	.show_options = erofs_show_options,
+};
+
+module_init(erofs_module_init);
+module_exit(erofs_module_exit);
+
+MODULE_DESCRIPTION("Enhanced ROM File System");
+MODULE_AUTHOR("Gao Xiang, Chao Yu, Miao Xie, CONSUMER BG, HUAWEI Inc.");
+MODULE_LICENSE("GPL");
diff --git a/fs/erofs/sysfs.c b/fs/erofs/sysfs.c
new file mode 100644
index 0000000000..435e515c07
--- /dev/null
+++ b/fs/erofs/sysfs.c
@@ -0,0 +1,277 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C), 2008-2021, OPPO Mobile Comm Corp., Ltd.
+ *             https://www.oppo.com/
+ */
+#include <linux/sysfs.h>
+#include <linux/kobject.h>
+
+#include "internal.h"
+
+enum {
+	attr_feature,
+	attr_pointer_ui,
+	attr_pointer_bool,
+};
+
+enum {
+	struct_erofs_sb_info,
+	struct_erofs_mount_opts,
+};
+
+struct erofs_attr {
+	struct attribute attr;
+	short attr_id;
+	int struct_type, offset;
+};
+
+#define EROFS_ATTR(_name, _mode, _id)					\
+static struct erofs_attr erofs_attr_##_name = {				\
+	.attr = {.name = __stringify(_name), .mode = _mode },		\
+	.attr_id = attr_##_id,						\
+}
+#define EROFS_ATTR_FUNC(_name, _mode)	EROFS_ATTR(_name, _mode, _name)
+#define EROFS_ATTR_FEATURE(_name)	EROFS_ATTR(_name, 0444, feature)
+
+#define EROFS_ATTR_OFFSET(_name, _mode, _id, _struct)	\
+static struct erofs_attr erofs_attr_##_name = {			\
+	.attr = {.name = __stringify(_name), .mode = _mode },	\
+	.attr_id = attr_##_id,					\
+	.struct_type = struct_##_struct,			\
+	.offset = offsetof(struct _struct, _name),\
+}
+
+#define EROFS_ATTR_RW(_name, _id, _struct)	\
+	EROFS_ATTR_OFFSET(_name, 0644, _id, _struct)
+
+#define EROFS_RO_ATTR(_name, _id, _struct)	\
+	EROFS_ATTR_OFFSET(_name, 0444, _id, _struct)
+
+#define EROFS_ATTR_RW_UI(_name, _struct)	\
+	EROFS_ATTR_RW(_name, pointer_ui, _struct)
+
+#define EROFS_ATTR_RW_BOOL(_name, _struct)	\
+	EROFS_ATTR_RW(_name, pointer_bool, _struct)
+
+#define ATTR_LIST(name) (&erofs_attr_##name.attr)
+
+#ifdef CONFIG_EROFS_FS_ZIP
+EROFS_ATTR_RW_UI(sync_decompress, erofs_mount_opts);
+#endif
+
+static struct attribute *erofs_attrs[] = {
+#ifdef CONFIG_EROFS_FS_ZIP
+	ATTR_LIST(sync_decompress),
+#endif
+	NULL,
+};
+ATTRIBUTE_GROUPS(erofs);
+
+/* Features this copy of erofs supports */
+EROFS_ATTR_FEATURE(zero_padding);
+EROFS_ATTR_FEATURE(compr_cfgs);
+EROFS_ATTR_FEATURE(big_pcluster);
+EROFS_ATTR_FEATURE(chunked_file);
+EROFS_ATTR_FEATURE(device_table);
+EROFS_ATTR_FEATURE(compr_head2);
+EROFS_ATTR_FEATURE(sb_chksum);
+EROFS_ATTR_FEATURE(ztailpacking);
+EROFS_ATTR_FEATURE(fragments);
+EROFS_ATTR_FEATURE(dedupe);
+
+static struct attribute *erofs_feat_attrs[] = {
+	ATTR_LIST(zero_padding),
+	ATTR_LIST(compr_cfgs),
+	ATTR_LIST(big_pcluster),
+	ATTR_LIST(chunked_file),
+	ATTR_LIST(device_table),
+	ATTR_LIST(compr_head2),
+	ATTR_LIST(sb_chksum),
+	ATTR_LIST(ztailpacking),
+	ATTR_LIST(fragments),
+	ATTR_LIST(dedupe),
+	NULL,
+};
+ATTRIBUTE_GROUPS(erofs_feat);
+
+static unsigned char *__struct_ptr(struct erofs_sb_info *sbi,
+					  int struct_type, int offset)
+{
+	if (struct_type == struct_erofs_sb_info)
+		return (unsigned char *)sbi + offset;
+	if (struct_type == struct_erofs_mount_opts)
+		return (unsigned char *)&sbi->opt + offset;
+	return NULL;
+}
+
+static ssize_t erofs_attr_show(struct kobject *kobj,
+				struct attribute *attr, char *buf)
+{
+	struct erofs_sb_info *sbi = container_of(kobj, struct erofs_sb_info,
+						s_kobj);
+	struct erofs_attr *a = container_of(attr, struct erofs_attr, attr);
+	unsigned char *ptr = __struct_ptr(sbi, a->struct_type, a->offset);
+
+	switch (a->attr_id) {
+	case attr_feature:
+		return sysfs_emit(buf, "supported\n");
+	case attr_pointer_ui:
+		if (!ptr)
+			return 0;
+		return sysfs_emit(buf, "%u\n", *(unsigned int *)ptr);
+	case attr_pointer_bool:
+		if (!ptr)
+			return 0;
+		return sysfs_emit(buf, "%d\n", *(bool *)ptr);
+	}
+	return 0;
+}
+
+static ssize_t erofs_attr_store(struct kobject *kobj, struct attribute *attr,
+						const char *buf, size_t len)
+{
+	struct erofs_sb_info *sbi = container_of(kobj, struct erofs_sb_info,
+						s_kobj);
+	struct erofs_attr *a = container_of(attr, struct erofs_attr, attr);
+	unsigned char *ptr = __struct_ptr(sbi, a->struct_type, a->offset);
+	unsigned long t;
+	int ret;
+
+	switch (a->attr_id) {
+	case attr_pointer_ui:
+		if (!ptr)
+			return 0;
+		ret = kstrtoul(skip_spaces(buf), 0, &t);
+		if (ret)
+			return ret;
+		if (t != (unsigned int)t)
+			return -ERANGE;
+#ifdef CONFIG_EROFS_FS_ZIP
+		if (!strcmp(a->attr.name, "sync_decompress") &&
+		    (t > EROFS_SYNC_DECOMPRESS_FORCE_OFF))
+			return -EINVAL;
+#endif
+		*(unsigned int *)ptr = t;
+		return len;
+	case attr_pointer_bool:
+		if (!ptr)
+			return 0;
+		ret = kstrtoul(skip_spaces(buf), 0, &t);
+		if (ret)
+			return ret;
+		if (t != 0 && t != 1)
+			return -EINVAL;
+		*(bool *)ptr = !!t;
+		return len;
+	}
+	return 0;
+}
+
+static void erofs_sb_release(struct kobject *kobj)
+{
+	struct erofs_sb_info *sbi = container_of(kobj, struct erofs_sb_info,
+						 s_kobj);
+	complete(&sbi->s_kobj_unregister);
+}
+
+static const struct sysfs_ops erofs_attr_ops = {
+	.show	= erofs_attr_show,
+	.store	= erofs_attr_store,
+};
+
+static const struct kobj_type erofs_sb_ktype = {
+	.default_groups = erofs_groups,
+	.sysfs_ops	= &erofs_attr_ops,
+	.release	= erofs_sb_release,
+};
+
+static const struct kobj_type erofs_ktype = {
+	.sysfs_ops	= &erofs_attr_ops,
+};
+
+static struct kset erofs_root = {
+	.kobj	= {.ktype = &erofs_ktype},
+};
+
+static const struct kobj_type erofs_feat_ktype = {
+	.default_groups = erofs_feat_groups,
+	.sysfs_ops	= &erofs_attr_ops,
+};
+
+static struct kobject erofs_feat = {
+	.kset	= &erofs_root,
+};
+
+int erofs_register_sysfs(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	char *name;
+	char *str = NULL;
+	int err;
+
+	if (erofs_is_fscache_mode(sb)) {
+		if (sbi->domain_id) {
+			str = kasprintf(GFP_KERNEL, "%s,%s", sbi->domain_id,
+					sbi->fsid);
+			if (!str)
+				return -ENOMEM;
+			name = str;
+		} else {
+			name = sbi->fsid;
+		}
+	} else {
+		name = sb->s_id;
+	}
+	sbi->s_kobj.kset = &erofs_root;
+	init_completion(&sbi->s_kobj_unregister);
+	err = kobject_init_and_add(&sbi->s_kobj, &erofs_sb_ktype, NULL, "%s", name);
+	kfree(str);
+	if (err)
+		goto put_sb_kobj;
+	return 0;
+
+put_sb_kobj:
+	kobject_put(&sbi->s_kobj);
+	wait_for_completion(&sbi->s_kobj_unregister);
+	return err;
+}
+
+void erofs_unregister_sysfs(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+
+	if (sbi->s_kobj.state_in_sysfs) {
+		kobject_del(&sbi->s_kobj);
+		kobject_put(&sbi->s_kobj);
+		wait_for_completion(&sbi->s_kobj_unregister);
+	}
+}
+
+int __init erofs_init_sysfs(void)
+{
+	int ret;
+
+	kobject_set_name(&erofs_root.kobj, "erofs");
+	erofs_root.kobj.parent = fs_kobj;
+	ret = kset_register(&erofs_root);
+	if (ret)
+		goto root_err;
+
+	ret = kobject_init_and_add(&erofs_feat, &erofs_feat_ktype,
+				   NULL, "features");
+	if (ret)
+		goto feat_err;
+	return ret;
+
+feat_err:
+	kobject_put(&erofs_feat);
+	kset_unregister(&erofs_root);
+root_err:
+	return ret;
+}
+
+void erofs_exit_sysfs(void)
+{
+	kobject_put(&erofs_feat);
+	kset_unregister(&erofs_root);
+}
diff --git a/fs/erofs/utils.c b/fs/erofs/utils.c
new file mode 100644
index 0000000000..4256a85719
--- /dev/null
+++ b/fs/erofs/utils.c
@@ -0,0 +1,282 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ */
+#include "internal.h"
+
+struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp)
+{
+	struct page *page = *pagepool;
+
+	if (page) {
+		DBG_BUGON(page_ref_count(page) != 1);
+		*pagepool = (struct page *)page_private(page);
+	} else {
+		page = alloc_page(gfp);
+	}
+	return page;
+}
+
+void erofs_release_pages(struct page **pagepool)
+{
+	while (*pagepool) {
+		struct page *page = *pagepool;
+
+		*pagepool = (struct page *)page_private(page);
+		put_page(page);
+	}
+}
+
+#ifdef CONFIG_EROFS_FS_ZIP
+/* global shrink count (for all mounted EROFS instances) */
+static atomic_long_t erofs_global_shrink_cnt;
+
+static bool erofs_workgroup_get(struct erofs_workgroup *grp)
+{
+	if (lockref_get_not_zero(&grp->lockref))
+		return true;
+
+	spin_lock(&grp->lockref.lock);
+	if (__lockref_is_dead(&grp->lockref)) {
+		spin_unlock(&grp->lockref.lock);
+		return false;
+	}
+
+	if (!grp->lockref.count++)
+		atomic_long_dec(&erofs_global_shrink_cnt);
+	spin_unlock(&grp->lockref.lock);
+	return true;
+}
+
+struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb,
+					     pgoff_t index)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct erofs_workgroup *grp;
+
+repeat:
+	rcu_read_lock();
+	grp = xa_load(&sbi->managed_pslots, index);
+	if (grp) {
+		if (!erofs_workgroup_get(grp)) {
+			/* prefer to relax rcu read side */
+			rcu_read_unlock();
+			goto repeat;
+		}
+
+		DBG_BUGON(index != grp->index);
+	}
+	rcu_read_unlock();
+	return grp;
+}
+
+struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb,
+					       struct erofs_workgroup *grp)
+{
+	struct erofs_sb_info *const sbi = EROFS_SB(sb);
+	struct erofs_workgroup *pre;
+
+	DBG_BUGON(grp->lockref.count < 1);
+repeat:
+	xa_lock(&sbi->managed_pslots);
+	pre = __xa_cmpxchg(&sbi->managed_pslots, grp->index,
+			   NULL, grp, GFP_NOFS);
+	if (pre) {
+		if (xa_is_err(pre)) {
+			pre = ERR_PTR(xa_err(pre));
+		} else if (!erofs_workgroup_get(pre)) {
+			/* try to legitimize the current in-tree one */
+			xa_unlock(&sbi->managed_pslots);
+			cond_resched();
+			goto repeat;
+		}
+		grp = pre;
+	}
+	xa_unlock(&sbi->managed_pslots);
+	return grp;
+}
+
+static void  __erofs_workgroup_free(struct erofs_workgroup *grp)
+{
+	atomic_long_dec(&erofs_global_shrink_cnt);
+	erofs_workgroup_free_rcu(grp);
+}
+
+void erofs_workgroup_put(struct erofs_workgroup *grp)
+{
+	if (lockref_put_or_lock(&grp->lockref))
+		return;
+
+	DBG_BUGON(__lockref_is_dead(&grp->lockref));
+	if (grp->lockref.count == 1)
+		atomic_long_inc(&erofs_global_shrink_cnt);
+	--grp->lockref.count;
+	spin_unlock(&grp->lockref.lock);
+}
+
+static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi,
+					   struct erofs_workgroup *grp)
+{
+	int free = false;
+
+	spin_lock(&grp->lockref.lock);
+	if (grp->lockref.count)
+		goto out;
+
+	/*
+	 * Note that all cached pages should be detached before deleted from
+	 * the XArray. Otherwise some cached pages could be still attached to
+	 * the orphan old workgroup when the new one is available in the tree.
+	 */
+	if (erofs_try_to_free_all_cached_pages(sbi, grp))
+		goto out;
+
+	/*
+	 * It's impossible to fail after the workgroup is freezed,
+	 * however in order to avoid some race conditions, add a
+	 * DBG_BUGON to observe this in advance.
+	 */
+	DBG_BUGON(__xa_erase(&sbi->managed_pslots, grp->index) != grp);
+
+	lockref_mark_dead(&grp->lockref);
+	free = true;
+out:
+	spin_unlock(&grp->lockref.lock);
+	if (free)
+		__erofs_workgroup_free(grp);
+	return free;
+}
+
+static unsigned long erofs_shrink_workstation(struct erofs_sb_info *sbi,
+					      unsigned long nr_shrink)
+{
+	struct erofs_workgroup *grp;
+	unsigned int freed = 0;
+	unsigned long index;
+
+	xa_lock(&sbi->managed_pslots);
+	xa_for_each(&sbi->managed_pslots, index, grp) {
+		/* try to shrink each valid workgroup */
+		if (!erofs_try_to_release_workgroup(sbi, grp))
+			continue;
+		xa_unlock(&sbi->managed_pslots);
+
+		++freed;
+		if (!--nr_shrink)
+			return freed;
+		xa_lock(&sbi->managed_pslots);
+	}
+	xa_unlock(&sbi->managed_pslots);
+	return freed;
+}
+
+/* protected by 'erofs_sb_list_lock' */
+static unsigned int shrinker_run_no;
+
+/* protects the mounted 'erofs_sb_list' */
+static DEFINE_SPINLOCK(erofs_sb_list_lock);
+static LIST_HEAD(erofs_sb_list);
+
+void erofs_shrinker_register(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+
+	mutex_init(&sbi->umount_mutex);
+
+	spin_lock(&erofs_sb_list_lock);
+	list_add(&sbi->list, &erofs_sb_list);
+	spin_unlock(&erofs_sb_list_lock);
+}
+
+void erofs_shrinker_unregister(struct super_block *sb)
+{
+	struct erofs_sb_info *const sbi = EROFS_SB(sb);
+
+	mutex_lock(&sbi->umount_mutex);
+	/* clean up all remaining workgroups in memory */
+	erofs_shrink_workstation(sbi, ~0UL);
+
+	spin_lock(&erofs_sb_list_lock);
+	list_del(&sbi->list);
+	spin_unlock(&erofs_sb_list_lock);
+	mutex_unlock(&sbi->umount_mutex);
+}
+
+static unsigned long erofs_shrink_count(struct shrinker *shrink,
+					struct shrink_control *sc)
+{
+	return atomic_long_read(&erofs_global_shrink_cnt);
+}
+
+static unsigned long erofs_shrink_scan(struct shrinker *shrink,
+				       struct shrink_control *sc)
+{
+	struct erofs_sb_info *sbi;
+	struct list_head *p;
+
+	unsigned long nr = sc->nr_to_scan;
+	unsigned int run_no;
+	unsigned long freed = 0;
+
+	spin_lock(&erofs_sb_list_lock);
+	do {
+		run_no = ++shrinker_run_no;
+	} while (run_no == 0);
+
+	/* Iterate over all mounted superblocks and try to shrink them */
+	p = erofs_sb_list.next;
+	while (p != &erofs_sb_list) {
+		sbi = list_entry(p, struct erofs_sb_info, list);
+
+		/*
+		 * We move the ones we do to the end of the list, so we stop
+		 * when we see one we have already done.
+		 */
+		if (sbi->shrinker_run_no == run_no)
+			break;
+
+		if (!mutex_trylock(&sbi->umount_mutex)) {
+			p = p->next;
+			continue;
+		}
+
+		spin_unlock(&erofs_sb_list_lock);
+		sbi->shrinker_run_no = run_no;
+
+		freed += erofs_shrink_workstation(sbi, nr - freed);
+
+		spin_lock(&erofs_sb_list_lock);
+		/* Get the next list element before we move this one */
+		p = p->next;
+
+		/*
+		 * Move this one to the end of the list to provide some
+		 * fairness.
+		 */
+		list_move_tail(&sbi->list, &erofs_sb_list);
+		mutex_unlock(&sbi->umount_mutex);
+
+		if (freed >= nr)
+			break;
+	}
+	spin_unlock(&erofs_sb_list_lock);
+	return freed;
+}
+
+static struct shrinker erofs_shrinker_info = {
+	.scan_objects = erofs_shrink_scan,
+	.count_objects = erofs_shrink_count,
+	.seeks = DEFAULT_SEEKS,
+};
+
+int __init erofs_init_shrinker(void)
+{
+	return register_shrinker(&erofs_shrinker_info, "erofs-shrinker");
+}
+
+void erofs_exit_shrinker(void)
+{
+	unregister_shrinker(&erofs_shrinker_info);
+}
+#endif	/* !CONFIG_EROFS_FS_ZIP */
diff --git a/fs/erofs/xattr.c b/fs/erofs/xattr.c
new file mode 100644
index 0000000000..09d341675e
--- /dev/null
+++ b/fs/erofs/xattr.c
@@ -0,0 +1,560 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2021-2022, Alibaba Cloud
+ */
+#include <linux/security.h>
+#include <linux/xxhash.h>
+#include "xattr.h"
+
+struct erofs_xattr_iter {
+	struct super_block *sb;
+	struct erofs_buf buf;
+	erofs_off_t pos;
+	void *kaddr;
+
+	char *buffer;
+	int buffer_size, buffer_ofs;
+
+	/* getxattr */
+	int index, infix_len;
+	struct qstr name;
+
+	/* listxattr */
+	struct dentry *dentry;
+};
+
+static int erofs_init_inode_xattrs(struct inode *inode)
+{
+	struct erofs_inode *const vi = EROFS_I(inode);
+	struct erofs_xattr_iter it;
+	unsigned int i;
+	struct erofs_xattr_ibody_header *ih;
+	struct super_block *sb = inode->i_sb;
+	int ret = 0;
+
+	/* the most case is that xattrs of this inode are initialized. */
+	if (test_bit(EROFS_I_EA_INITED_BIT, &vi->flags)) {
+		/*
+		 * paired with smp_mb() at the end of the function to ensure
+		 * fields will only be observed after the bit is set.
+		 */
+		smp_mb();
+		return 0;
+	}
+
+	if (wait_on_bit_lock(&vi->flags, EROFS_I_BL_XATTR_BIT, TASK_KILLABLE))
+		return -ERESTARTSYS;
+
+	/* someone has initialized xattrs for us? */
+	if (test_bit(EROFS_I_EA_INITED_BIT, &vi->flags))
+		goto out_unlock;
+
+	/*
+	 * bypass all xattr operations if ->xattr_isize is not greater than
+	 * sizeof(struct erofs_xattr_ibody_header), in detail:
+	 * 1) it is not enough to contain erofs_xattr_ibody_header then
+	 *    ->xattr_isize should be 0 (it means no xattr);
+	 * 2) it is just to contain erofs_xattr_ibody_header, which is on-disk
+	 *    undefined right now (maybe use later with some new sb feature).
+	 */
+	if (vi->xattr_isize == sizeof(struct erofs_xattr_ibody_header)) {
+		erofs_err(sb,
+			  "xattr_isize %d of nid %llu is not supported yet",
+			  vi->xattr_isize, vi->nid);
+		ret = -EOPNOTSUPP;
+		goto out_unlock;
+	} else if (vi->xattr_isize < sizeof(struct erofs_xattr_ibody_header)) {
+		if (vi->xattr_isize) {
+			erofs_err(sb, "bogus xattr ibody @ nid %llu", vi->nid);
+			DBG_BUGON(1);
+			ret = -EFSCORRUPTED;
+			goto out_unlock;	/* xattr ondisk layout error */
+		}
+		ret = -ENOATTR;
+		goto out_unlock;
+	}
+
+	it.buf = __EROFS_BUF_INITIALIZER;
+	erofs_init_metabuf(&it.buf, sb);
+	it.pos = erofs_iloc(inode) + vi->inode_isize;
+
+	/* read in shared xattr array (non-atomic, see kmalloc below) */
+	it.kaddr = erofs_bread(&it.buf, erofs_blknr(sb, it.pos), EROFS_KMAP);
+	if (IS_ERR(it.kaddr)) {
+		ret = PTR_ERR(it.kaddr);
+		goto out_unlock;
+	}
+
+	ih = it.kaddr + erofs_blkoff(sb, it.pos);
+	vi->xattr_name_filter = le32_to_cpu(ih->h_name_filter);
+	vi->xattr_shared_count = ih->h_shared_count;
+	vi->xattr_shared_xattrs = kmalloc_array(vi->xattr_shared_count,
+						sizeof(uint), GFP_KERNEL);
+	if (!vi->xattr_shared_xattrs) {
+		erofs_put_metabuf(&it.buf);
+		ret = -ENOMEM;
+		goto out_unlock;
+	}
+
+	/* let's skip ibody header */
+	it.pos += sizeof(struct erofs_xattr_ibody_header);
+
+	for (i = 0; i < vi->xattr_shared_count; ++i) {
+		it.kaddr = erofs_bread(&it.buf, erofs_blknr(sb, it.pos),
+				       EROFS_KMAP);
+		if (IS_ERR(it.kaddr)) {
+			kfree(vi->xattr_shared_xattrs);
+			vi->xattr_shared_xattrs = NULL;
+			ret = PTR_ERR(it.kaddr);
+			goto out_unlock;
+		}
+		vi->xattr_shared_xattrs[i] = le32_to_cpu(*(__le32 *)
+				(it.kaddr + erofs_blkoff(sb, it.pos)));
+		it.pos += sizeof(__le32);
+	}
+	erofs_put_metabuf(&it.buf);
+
+	/* paired with smp_mb() at the beginning of the function. */
+	smp_mb();
+	set_bit(EROFS_I_EA_INITED_BIT, &vi->flags);
+
+out_unlock:
+	clear_and_wake_up_bit(EROFS_I_BL_XATTR_BIT, &vi->flags);
+	return ret;
+}
+
+static bool erofs_xattr_user_list(struct dentry *dentry)
+{
+	return test_opt(&EROFS_SB(dentry->d_sb)->opt, XATTR_USER);
+}
+
+static bool erofs_xattr_trusted_list(struct dentry *dentry)
+{
+	return capable(CAP_SYS_ADMIN);
+}
+
+static int erofs_xattr_generic_get(const struct xattr_handler *handler,
+				   struct dentry *unused, struct inode *inode,
+				   const char *name, void *buffer, size_t size)
+{
+	if (handler->flags == EROFS_XATTR_INDEX_USER &&
+	    !test_opt(&EROFS_I_SB(inode)->opt, XATTR_USER))
+		return -EOPNOTSUPP;
+
+	return erofs_getxattr(inode, handler->flags, name, buffer, size);
+}
+
+const struct xattr_handler erofs_xattr_user_handler = {
+	.prefix	= XATTR_USER_PREFIX,
+	.flags	= EROFS_XATTR_INDEX_USER,
+	.list	= erofs_xattr_user_list,
+	.get	= erofs_xattr_generic_get,
+};
+
+const struct xattr_handler erofs_xattr_trusted_handler = {
+	.prefix	= XATTR_TRUSTED_PREFIX,
+	.flags	= EROFS_XATTR_INDEX_TRUSTED,
+	.list	= erofs_xattr_trusted_list,
+	.get	= erofs_xattr_generic_get,
+};
+
+#ifdef CONFIG_EROFS_FS_SECURITY
+const struct xattr_handler __maybe_unused erofs_xattr_security_handler = {
+	.prefix	= XATTR_SECURITY_PREFIX,
+	.flags	= EROFS_XATTR_INDEX_SECURITY,
+	.get	= erofs_xattr_generic_get,
+};
+#endif
+
+const struct xattr_handler *erofs_xattr_handlers[] = {
+	&erofs_xattr_user_handler,
+	&erofs_xattr_trusted_handler,
+#ifdef CONFIG_EROFS_FS_SECURITY
+	&erofs_xattr_security_handler,
+#endif
+	NULL,
+};
+
+static int erofs_xattr_copy_to_buffer(struct erofs_xattr_iter *it,
+				      unsigned int len)
+{
+	unsigned int slice, processed;
+	struct super_block *sb = it->sb;
+	void *src;
+
+	for (processed = 0; processed < len; processed += slice) {
+		it->kaddr = erofs_bread(&it->buf, erofs_blknr(sb, it->pos),
+					EROFS_KMAP);
+		if (IS_ERR(it->kaddr))
+			return PTR_ERR(it->kaddr);
+
+		src = it->kaddr + erofs_blkoff(sb, it->pos);
+		slice = min_t(unsigned int, sb->s_blocksize -
+				erofs_blkoff(sb, it->pos), len - processed);
+		memcpy(it->buffer + it->buffer_ofs, src, slice);
+		it->buffer_ofs += slice;
+		it->pos += slice;
+	}
+	return 0;
+}
+
+static int erofs_listxattr_foreach(struct erofs_xattr_iter *it)
+{
+	struct erofs_xattr_entry entry;
+	unsigned int base_index, name_total, prefix_len, infix_len = 0;
+	const char *prefix, *infix = NULL;
+	int err;
+
+	/* 1. handle xattr entry */
+	entry = *(struct erofs_xattr_entry *)
+			(it->kaddr + erofs_blkoff(it->sb, it->pos));
+	it->pos += sizeof(struct erofs_xattr_entry);
+
+	base_index = entry.e_name_index;
+	if (entry.e_name_index & EROFS_XATTR_LONG_PREFIX) {
+		struct erofs_sb_info *sbi = EROFS_SB(it->sb);
+		struct erofs_xattr_prefix_item *pf = sbi->xattr_prefixes +
+			(entry.e_name_index & EROFS_XATTR_LONG_PREFIX_MASK);
+
+		if (pf >= sbi->xattr_prefixes + sbi->xattr_prefix_count)
+			return 0;
+		infix = pf->prefix->infix;
+		infix_len = pf->infix_len;
+		base_index = pf->prefix->base_index;
+	}
+
+	prefix = erofs_xattr_prefix(base_index, it->dentry);
+	if (!prefix)
+		return 0;
+	prefix_len = strlen(prefix);
+	name_total = prefix_len + infix_len + entry.e_name_len + 1;
+
+	if (!it->buffer) {
+		it->buffer_ofs += name_total;
+		return 0;
+	}
+
+	if (it->buffer_ofs + name_total > it->buffer_size)
+		return -ERANGE;
+
+	memcpy(it->buffer + it->buffer_ofs, prefix, prefix_len);
+	memcpy(it->buffer + it->buffer_ofs + prefix_len, infix, infix_len);
+	it->buffer_ofs += prefix_len + infix_len;
+
+	/* 2. handle xattr name */
+	err = erofs_xattr_copy_to_buffer(it, entry.e_name_len);
+	if (err)
+		return err;
+
+	it->buffer[it->buffer_ofs++] = '\0';
+	return 0;
+}
+
+static int erofs_getxattr_foreach(struct erofs_xattr_iter *it)
+{
+	struct super_block *sb = it->sb;
+	struct erofs_xattr_entry entry;
+	unsigned int slice, processed, value_sz;
+
+	/* 1. handle xattr entry */
+	entry = *(struct erofs_xattr_entry *)
+			(it->kaddr + erofs_blkoff(sb, it->pos));
+	it->pos += sizeof(struct erofs_xattr_entry);
+	value_sz = le16_to_cpu(entry.e_value_size);
+
+	/* should also match the infix for long name prefixes */
+	if (entry.e_name_index & EROFS_XATTR_LONG_PREFIX) {
+		struct erofs_sb_info *sbi = EROFS_SB(sb);
+		struct erofs_xattr_prefix_item *pf = sbi->xattr_prefixes +
+			(entry.e_name_index & EROFS_XATTR_LONG_PREFIX_MASK);
+
+		if (pf >= sbi->xattr_prefixes + sbi->xattr_prefix_count)
+			return -ENOATTR;
+
+		if (it->index != pf->prefix->base_index ||
+		    it->name.len != entry.e_name_len + pf->infix_len)
+			return -ENOATTR;
+
+		if (memcmp(it->name.name, pf->prefix->infix, pf->infix_len))
+			return -ENOATTR;
+
+		it->infix_len = pf->infix_len;
+	} else {
+		if (it->index != entry.e_name_index ||
+		    it->name.len != entry.e_name_len)
+			return -ENOATTR;
+
+		it->infix_len = 0;
+	}
+
+	/* 2. handle xattr name */
+	for (processed = 0; processed < entry.e_name_len; processed += slice) {
+		it->kaddr = erofs_bread(&it->buf, erofs_blknr(sb, it->pos),
+					EROFS_KMAP);
+		if (IS_ERR(it->kaddr))
+			return PTR_ERR(it->kaddr);
+
+		slice = min_t(unsigned int,
+				sb->s_blocksize - erofs_blkoff(sb, it->pos),
+				entry.e_name_len - processed);
+		if (memcmp(it->name.name + it->infix_len + processed,
+			   it->kaddr + erofs_blkoff(sb, it->pos), slice))
+			return -ENOATTR;
+		it->pos += slice;
+	}
+
+	/* 3. handle xattr value */
+	if (!it->buffer) {
+		it->buffer_ofs = value_sz;
+		return 0;
+	}
+
+	if (it->buffer_size < value_sz)
+		return -ERANGE;
+
+	return erofs_xattr_copy_to_buffer(it, value_sz);
+}
+
+static int erofs_xattr_iter_inline(struct erofs_xattr_iter *it,
+				   struct inode *inode, bool getxattr)
+{
+	struct erofs_inode *const vi = EROFS_I(inode);
+	unsigned int xattr_header_sz, remaining, entry_sz;
+	erofs_off_t next_pos;
+	int ret;
+
+	xattr_header_sz = sizeof(struct erofs_xattr_ibody_header) +
+			  sizeof(u32) * vi->xattr_shared_count;
+	if (xattr_header_sz >= vi->xattr_isize) {
+		DBG_BUGON(xattr_header_sz > vi->xattr_isize);
+		return -ENOATTR;
+	}
+
+	remaining = vi->xattr_isize - xattr_header_sz;
+	it->pos = erofs_iloc(inode) + vi->inode_isize + xattr_header_sz;
+
+	while (remaining) {
+		it->kaddr = erofs_bread(&it->buf, erofs_blknr(it->sb, it->pos),
+					EROFS_KMAP);
+		if (IS_ERR(it->kaddr))
+			return PTR_ERR(it->kaddr);
+
+		entry_sz = erofs_xattr_entry_size(it->kaddr +
+				erofs_blkoff(it->sb, it->pos));
+		/* xattr on-disk corruption: xattr entry beyond xattr_isize */
+		if (remaining < entry_sz) {
+			DBG_BUGON(1);
+			return -EFSCORRUPTED;
+		}
+		remaining -= entry_sz;
+		next_pos = it->pos + entry_sz;
+
+		if (getxattr)
+			ret = erofs_getxattr_foreach(it);
+		else
+			ret = erofs_listxattr_foreach(it);
+		if ((getxattr && ret != -ENOATTR) || (!getxattr && ret))
+			break;
+
+		it->pos = next_pos;
+	}
+	return ret;
+}
+
+static int erofs_xattr_iter_shared(struct erofs_xattr_iter *it,
+				   struct inode *inode, bool getxattr)
+{
+	struct erofs_inode *const vi = EROFS_I(inode);
+	struct super_block *const sb = it->sb;
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	unsigned int i;
+	int ret = -ENOATTR;
+
+	for (i = 0; i < vi->xattr_shared_count; ++i) {
+		it->pos = erofs_pos(sb, sbi->xattr_blkaddr) +
+				vi->xattr_shared_xattrs[i] * sizeof(__le32);
+		it->kaddr = erofs_bread(&it->buf, erofs_blknr(sb, it->pos),
+					EROFS_KMAP);
+		if (IS_ERR(it->kaddr))
+			return PTR_ERR(it->kaddr);
+
+		if (getxattr)
+			ret = erofs_getxattr_foreach(it);
+		else
+			ret = erofs_listxattr_foreach(it);
+		if ((getxattr && ret != -ENOATTR) || (!getxattr && ret))
+			break;
+	}
+	return ret;
+}
+
+int erofs_getxattr(struct inode *inode, int index, const char *name,
+		   void *buffer, size_t buffer_size)
+{
+	int ret;
+	unsigned int hashbit;
+	struct erofs_xattr_iter it;
+	struct erofs_inode *vi = EROFS_I(inode);
+	struct erofs_sb_info *sbi = EROFS_SB(inode->i_sb);
+
+	if (!name)
+		return -EINVAL;
+
+	ret = erofs_init_inode_xattrs(inode);
+	if (ret)
+		return ret;
+
+	/* reserved flag is non-zero if there's any change of on-disk format */
+	if (erofs_sb_has_xattr_filter(sbi) && !sbi->xattr_filter_reserved) {
+		hashbit = xxh32(name, strlen(name),
+				EROFS_XATTR_FILTER_SEED + index);
+		hashbit &= EROFS_XATTR_FILTER_BITS - 1;
+		if (vi->xattr_name_filter & (1U << hashbit))
+			return -ENOATTR;
+	}
+
+	it.index = index;
+	it.name = (struct qstr)QSTR_INIT(name, strlen(name));
+	if (it.name.len > EROFS_NAME_LEN)
+		return -ERANGE;
+
+	it.sb = inode->i_sb;
+	it.buf = __EROFS_BUF_INITIALIZER;
+	erofs_init_metabuf(&it.buf, it.sb);
+	it.buffer = buffer;
+	it.buffer_size = buffer_size;
+	it.buffer_ofs = 0;
+
+	ret = erofs_xattr_iter_inline(&it, inode, true);
+	if (ret == -ENOATTR)
+		ret = erofs_xattr_iter_shared(&it, inode, true);
+	erofs_put_metabuf(&it.buf);
+	return ret ? ret : it.buffer_ofs;
+}
+
+ssize_t erofs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
+{
+	int ret;
+	struct erofs_xattr_iter it;
+	struct inode *inode = d_inode(dentry);
+
+	ret = erofs_init_inode_xattrs(inode);
+	if (ret == -ENOATTR)
+		return 0;
+	if (ret)
+		return ret;
+
+	it.sb = dentry->d_sb;
+	it.buf = __EROFS_BUF_INITIALIZER;
+	erofs_init_metabuf(&it.buf, it.sb);
+	it.dentry = dentry;
+	it.buffer = buffer;
+	it.buffer_size = buffer_size;
+	it.buffer_ofs = 0;
+
+	ret = erofs_xattr_iter_inline(&it, inode, false);
+	if (!ret || ret == -ENOATTR)
+		ret = erofs_xattr_iter_shared(&it, inode, false);
+	if (ret == -ENOATTR)
+		ret = 0;
+	erofs_put_metabuf(&it.buf);
+	return ret ? ret : it.buffer_ofs;
+}
+
+void erofs_xattr_prefixes_cleanup(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	int i;
+
+	if (sbi->xattr_prefixes) {
+		for (i = 0; i < sbi->xattr_prefix_count; i++)
+			kfree(sbi->xattr_prefixes[i].prefix);
+		kfree(sbi->xattr_prefixes);
+		sbi->xattr_prefixes = NULL;
+	}
+}
+
+int erofs_xattr_prefixes_init(struct super_block *sb)
+{
+	struct erofs_sb_info *sbi = EROFS_SB(sb);
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	erofs_off_t pos = (erofs_off_t)sbi->xattr_prefix_start << 2;
+	struct erofs_xattr_prefix_item *pfs;
+	int ret = 0, i, len;
+
+	if (!sbi->xattr_prefix_count)
+		return 0;
+
+	pfs = kzalloc(sbi->xattr_prefix_count * sizeof(*pfs), GFP_KERNEL);
+	if (!pfs)
+		return -ENOMEM;
+
+	if (sbi->packed_inode)
+		buf.inode = sbi->packed_inode;
+	else
+		erofs_init_metabuf(&buf, sb);
+
+	for (i = 0; i < sbi->xattr_prefix_count; i++) {
+		void *ptr = erofs_read_metadata(sb, &buf, &pos, &len);
+
+		if (IS_ERR(ptr)) {
+			ret = PTR_ERR(ptr);
+			break;
+		} else if (len < sizeof(*pfs->prefix) ||
+			   len > EROFS_NAME_LEN + sizeof(*pfs->prefix)) {
+			kfree(ptr);
+			ret = -EFSCORRUPTED;
+			break;
+		}
+		pfs[i].prefix = ptr;
+		pfs[i].infix_len = len - sizeof(struct erofs_xattr_long_prefix);
+	}
+
+	erofs_put_metabuf(&buf);
+	sbi->xattr_prefixes = pfs;
+	if (ret)
+		erofs_xattr_prefixes_cleanup(sb);
+	return ret;
+}
+
+#ifdef CONFIG_EROFS_FS_POSIX_ACL
+struct posix_acl *erofs_get_acl(struct inode *inode, int type, bool rcu)
+{
+	struct posix_acl *acl;
+	int prefix, rc;
+	char *value = NULL;
+
+	if (rcu)
+		return ERR_PTR(-ECHILD);
+
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		prefix = EROFS_XATTR_INDEX_POSIX_ACL_ACCESS;
+		break;
+	case ACL_TYPE_DEFAULT:
+		prefix = EROFS_XATTR_INDEX_POSIX_ACL_DEFAULT;
+		break;
+	default:
+		return ERR_PTR(-EINVAL);
+	}
+
+	rc = erofs_getxattr(inode, prefix, "", NULL, 0);
+	if (rc > 0) {
+		value = kmalloc(rc, GFP_KERNEL);
+		if (!value)
+			return ERR_PTR(-ENOMEM);
+		rc = erofs_getxattr(inode, prefix, "", value, rc);
+	}
+
+	if (rc == -ENOATTR)
+		acl = NULL;
+	else if (rc < 0)
+		acl = ERR_PTR(rc);
+	else
+		acl = posix_acl_from_xattr(&init_user_ns, value, rc);
+	kfree(value);
+	return acl;
+}
+#endif
diff --git a/fs/erofs/xattr.h b/fs/erofs/xattr.h
new file mode 100644
index 0000000000..f16283cb8c
--- /dev/null
+++ b/fs/erofs/xattr.h
@@ -0,0 +1,73 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2017-2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ */
+#ifndef __EROFS_XATTR_H
+#define __EROFS_XATTR_H
+
+#include "internal.h"
+#include <linux/posix_acl_xattr.h>
+#include <linux/xattr.h>
+
+/* Attribute not found */
+#define ENOATTR         ENODATA
+
+#ifdef CONFIG_EROFS_FS_XATTR
+extern const struct xattr_handler erofs_xattr_user_handler;
+extern const struct xattr_handler erofs_xattr_trusted_handler;
+extern const struct xattr_handler erofs_xattr_security_handler;
+
+static inline const char *erofs_xattr_prefix(unsigned int idx,
+					     struct dentry *dentry)
+{
+	const struct xattr_handler *handler = NULL;
+
+	static const struct xattr_handler *xattr_handler_map[] = {
+		[EROFS_XATTR_INDEX_USER] = &erofs_xattr_user_handler,
+#ifdef CONFIG_EROFS_FS_POSIX_ACL
+		[EROFS_XATTR_INDEX_POSIX_ACL_ACCESS] = &nop_posix_acl_access,
+		[EROFS_XATTR_INDEX_POSIX_ACL_DEFAULT] = &nop_posix_acl_default,
+#endif
+		[EROFS_XATTR_INDEX_TRUSTED] = &erofs_xattr_trusted_handler,
+#ifdef CONFIG_EROFS_FS_SECURITY
+		[EROFS_XATTR_INDEX_SECURITY] = &erofs_xattr_security_handler,
+#endif
+	};
+
+	if (idx && idx < ARRAY_SIZE(xattr_handler_map))
+		handler = xattr_handler_map[idx];
+
+	if (!xattr_handler_can_list(handler, dentry))
+		return NULL;
+
+	return xattr_prefix(handler);
+}
+
+extern const struct xattr_handler *erofs_xattr_handlers[];
+
+int erofs_xattr_prefixes_init(struct super_block *sb);
+void erofs_xattr_prefixes_cleanup(struct super_block *sb);
+int erofs_getxattr(struct inode *, int, const char *, void *, size_t);
+ssize_t erofs_listxattr(struct dentry *, char *, size_t);
+#else
+static inline int erofs_xattr_prefixes_init(struct super_block *sb) { return 0; }
+static inline void erofs_xattr_prefixes_cleanup(struct super_block *sb) {}
+static inline int erofs_getxattr(struct inode *inode, int index,
+				 const char *name, void *buffer,
+				 size_t buffer_size)
+{
+	return -EOPNOTSUPP;
+}
+
+#define erofs_listxattr (NULL)
+#define erofs_xattr_handlers (NULL)
+#endif	/* !CONFIG_EROFS_FS_XATTR */
+
+#ifdef CONFIG_EROFS_FS_POSIX_ACL
+struct posix_acl *erofs_get_acl(struct inode *inode, int type, bool rcu);
+#else
+#define erofs_get_acl	(NULL)
+#endif
+
+#endif
diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
new file mode 100644
index 0000000000..a33cd6757f
--- /dev/null
+++ b/fs/erofs/zdata.c
@@ -0,0 +1,1894 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2018 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ * Copyright (C) 2022 Alibaba Cloud
+ */
+#include "compress.h"
+#include <linux/psi.h>
+#include <linux/cpuhotplug.h>
+#include <trace/events/erofs.h>
+
+#define Z_EROFS_PCLUSTER_MAX_PAGES	(Z_EROFS_PCLUSTER_MAX_SIZE / PAGE_SIZE)
+#define Z_EROFS_INLINE_BVECS		2
+
+/*
+ * let's leave a type here in case of introducing
+ * another tagged pointer later.
+ */
+typedef void *z_erofs_next_pcluster_t;
+
+struct z_erofs_bvec {
+	struct page *page;
+	int offset;
+	unsigned int end;
+};
+
+#define __Z_EROFS_BVSET(name, total) \
+struct name { \
+	/* point to the next page which contains the following bvecs */ \
+	struct page *nextpage; \
+	struct z_erofs_bvec bvec[total]; \
+}
+__Z_EROFS_BVSET(z_erofs_bvset,);
+__Z_EROFS_BVSET(z_erofs_bvset_inline, Z_EROFS_INLINE_BVECS);
+
+/*
+ * Structure fields follow one of the following exclusion rules.
+ *
+ * I: Modifiable by initialization/destruction paths and read-only
+ *    for everyone else;
+ *
+ * L: Field should be protected by the pcluster lock;
+ *
+ * A: Field should be accessed / updated in atomic for parallelized code.
+ */
+struct z_erofs_pcluster {
+	struct erofs_workgroup obj;
+	struct mutex lock;
+
+	/* A: point to next chained pcluster or TAILs */
+	z_erofs_next_pcluster_t next;
+
+	/* L: the maximum decompression size of this round */
+	unsigned int length;
+
+	/* L: total number of bvecs */
+	unsigned int vcnt;
+
+	/* I: page offset of start position of decompression */
+	unsigned short pageofs_out;
+
+	/* I: page offset of inline compressed data */
+	unsigned short pageofs_in;
+
+	union {
+		/* L: inline a certain number of bvec for bootstrap */
+		struct z_erofs_bvset_inline bvset;
+
+		/* I: can be used to free the pcluster by RCU. */
+		struct rcu_head rcu;
+	};
+
+	union {
+		/* I: physical cluster size in pages */
+		unsigned short pclusterpages;
+
+		/* I: tailpacking inline compressed size */
+		unsigned short tailpacking_size;
+	};
+
+	/* I: compression algorithm format */
+	unsigned char algorithmformat;
+
+	/* L: whether partial decompression or not */
+	bool partial;
+
+	/* L: indicate several pageofs_outs or not */
+	bool multibases;
+
+	/* A: compressed bvecs (can be cached or inplaced pages) */
+	struct z_erofs_bvec compressed_bvecs[];
+};
+
+/* the end of a chain of pclusters */
+#define Z_EROFS_PCLUSTER_TAIL           ((void *) 0x700 + POISON_POINTER_DELTA)
+#define Z_EROFS_PCLUSTER_NIL            (NULL)
+
+struct z_erofs_decompressqueue {
+	struct super_block *sb;
+	atomic_t pending_bios;
+	z_erofs_next_pcluster_t head;
+
+	union {
+		struct completion done;
+		struct work_struct work;
+		struct kthread_work kthread_work;
+	} u;
+	bool eio, sync;
+};
+
+static inline bool z_erofs_is_inline_pcluster(struct z_erofs_pcluster *pcl)
+{
+	return !pcl->obj.index;
+}
+
+static inline unsigned int z_erofs_pclusterpages(struct z_erofs_pcluster *pcl)
+{
+	if (z_erofs_is_inline_pcluster(pcl))
+		return 1;
+	return pcl->pclusterpages;
+}
+
+/*
+ * bit 30: I/O error occurred on this page
+ * bit 0 - 29: remaining parts to complete this page
+ */
+#define Z_EROFS_PAGE_EIO			(1 << 30)
+
+static inline void z_erofs_onlinepage_init(struct page *page)
+{
+	union {
+		atomic_t o;
+		unsigned long v;
+	} u = { .o = ATOMIC_INIT(1) };
+
+	set_page_private(page, u.v);
+	smp_wmb();
+	SetPagePrivate(page);
+}
+
+static inline void z_erofs_onlinepage_split(struct page *page)
+{
+	atomic_inc((atomic_t *)&page->private);
+}
+
+static void z_erofs_onlinepage_endio(struct page *page, int err)
+{
+	int orig, v;
+
+	DBG_BUGON(!PagePrivate(page));
+
+	do {
+		orig = atomic_read((atomic_t *)&page->private);
+		v = (orig - 1) | (err ? Z_EROFS_PAGE_EIO : 0);
+	} while (atomic_cmpxchg((atomic_t *)&page->private, orig, v) != orig);
+
+	if (!(v & ~Z_EROFS_PAGE_EIO)) {
+		set_page_private(page, 0);
+		ClearPagePrivate(page);
+		if (!(v & Z_EROFS_PAGE_EIO))
+			SetPageUptodate(page);
+		unlock_page(page);
+	}
+}
+
+#define Z_EROFS_ONSTACK_PAGES		32
+
+/*
+ * since pclustersize is variable for big pcluster feature, introduce slab
+ * pools implementation for different pcluster sizes.
+ */
+struct z_erofs_pcluster_slab {
+	struct kmem_cache *slab;
+	unsigned int maxpages;
+	char name[48];
+};
+
+#define _PCLP(n) { .maxpages = n }
+
+static struct z_erofs_pcluster_slab pcluster_pool[] __read_mostly = {
+	_PCLP(1), _PCLP(4), _PCLP(16), _PCLP(64), _PCLP(128),
+	_PCLP(Z_EROFS_PCLUSTER_MAX_PAGES)
+};
+
+struct z_erofs_bvec_iter {
+	struct page *bvpage;
+	struct z_erofs_bvset *bvset;
+	unsigned int nr, cur;
+};
+
+static struct page *z_erofs_bvec_iter_end(struct z_erofs_bvec_iter *iter)
+{
+	if (iter->bvpage)
+		kunmap_local(iter->bvset);
+	return iter->bvpage;
+}
+
+static struct page *z_erofs_bvset_flip(struct z_erofs_bvec_iter *iter)
+{
+	unsigned long base = (unsigned long)((struct z_erofs_bvset *)0)->bvec;
+	/* have to access nextpage in advance, otherwise it will be unmapped */
+	struct page *nextpage = iter->bvset->nextpage;
+	struct page *oldpage;
+
+	DBG_BUGON(!nextpage);
+	oldpage = z_erofs_bvec_iter_end(iter);
+	iter->bvpage = nextpage;
+	iter->bvset = kmap_local_page(nextpage);
+	iter->nr = (PAGE_SIZE - base) / sizeof(struct z_erofs_bvec);
+	iter->cur = 0;
+	return oldpage;
+}
+
+static void z_erofs_bvec_iter_begin(struct z_erofs_bvec_iter *iter,
+				    struct z_erofs_bvset_inline *bvset,
+				    unsigned int bootstrap_nr,
+				    unsigned int cur)
+{
+	*iter = (struct z_erofs_bvec_iter) {
+		.nr = bootstrap_nr,
+		.bvset = (struct z_erofs_bvset *)bvset,
+	};
+
+	while (cur > iter->nr) {
+		cur -= iter->nr;
+		z_erofs_bvset_flip(iter);
+	}
+	iter->cur = cur;
+}
+
+static int z_erofs_bvec_enqueue(struct z_erofs_bvec_iter *iter,
+				struct z_erofs_bvec *bvec,
+				struct page **candidate_bvpage,
+				struct page **pagepool)
+{
+	if (iter->cur >= iter->nr) {
+		struct page *nextpage = *candidate_bvpage;
+
+		if (!nextpage) {
+			nextpage = erofs_allocpage(pagepool, GFP_NOFS);
+			if (!nextpage)
+				return -ENOMEM;
+			set_page_private(nextpage, Z_EROFS_SHORTLIVED_PAGE);
+		}
+		DBG_BUGON(iter->bvset->nextpage);
+		iter->bvset->nextpage = nextpage;
+		z_erofs_bvset_flip(iter);
+
+		iter->bvset->nextpage = NULL;
+		*candidate_bvpage = NULL;
+	}
+	iter->bvset->bvec[iter->cur++] = *bvec;
+	return 0;
+}
+
+static void z_erofs_bvec_dequeue(struct z_erofs_bvec_iter *iter,
+				 struct z_erofs_bvec *bvec,
+				 struct page **old_bvpage)
+{
+	if (iter->cur == iter->nr)
+		*old_bvpage = z_erofs_bvset_flip(iter);
+	else
+		*old_bvpage = NULL;
+	*bvec = iter->bvset->bvec[iter->cur++];
+}
+
+static void z_erofs_destroy_pcluster_pool(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(pcluster_pool); ++i) {
+		if (!pcluster_pool[i].slab)
+			continue;
+		kmem_cache_destroy(pcluster_pool[i].slab);
+		pcluster_pool[i].slab = NULL;
+	}
+}
+
+static int z_erofs_create_pcluster_pool(void)
+{
+	struct z_erofs_pcluster_slab *pcs;
+	struct z_erofs_pcluster *a;
+	unsigned int size;
+
+	for (pcs = pcluster_pool;
+	     pcs < pcluster_pool + ARRAY_SIZE(pcluster_pool); ++pcs) {
+		size = struct_size(a, compressed_bvecs, pcs->maxpages);
+
+		sprintf(pcs->name, "erofs_pcluster-%u", pcs->maxpages);
+		pcs->slab = kmem_cache_create(pcs->name, size, 0,
+					      SLAB_RECLAIM_ACCOUNT, NULL);
+		if (pcs->slab)
+			continue;
+
+		z_erofs_destroy_pcluster_pool();
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+static struct z_erofs_pcluster *z_erofs_alloc_pcluster(unsigned int nrpages)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(pcluster_pool); ++i) {
+		struct z_erofs_pcluster_slab *pcs = pcluster_pool + i;
+		struct z_erofs_pcluster *pcl;
+
+		if (nrpages > pcs->maxpages)
+			continue;
+
+		pcl = kmem_cache_zalloc(pcs->slab, GFP_NOFS);
+		if (!pcl)
+			return ERR_PTR(-ENOMEM);
+		pcl->pclusterpages = nrpages;
+		return pcl;
+	}
+	return ERR_PTR(-EINVAL);
+}
+
+static void z_erofs_free_pcluster(struct z_erofs_pcluster *pcl)
+{
+	unsigned int pclusterpages = z_erofs_pclusterpages(pcl);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(pcluster_pool); ++i) {
+		struct z_erofs_pcluster_slab *pcs = pcluster_pool + i;
+
+		if (pclusterpages > pcs->maxpages)
+			continue;
+
+		kmem_cache_free(pcs->slab, pcl);
+		return;
+	}
+	DBG_BUGON(1);
+}
+
+static struct workqueue_struct *z_erofs_workqueue __read_mostly;
+
+#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD
+static struct kthread_worker __rcu **z_erofs_pcpu_workers;
+
+static void erofs_destroy_percpu_workers(void)
+{
+	struct kthread_worker *worker;
+	unsigned int cpu;
+
+	for_each_possible_cpu(cpu) {
+		worker = rcu_dereference_protected(
+					z_erofs_pcpu_workers[cpu], 1);
+		rcu_assign_pointer(z_erofs_pcpu_workers[cpu], NULL);
+		if (worker)
+			kthread_destroy_worker(worker);
+	}
+	kfree(z_erofs_pcpu_workers);
+}
+
+static struct kthread_worker *erofs_init_percpu_worker(int cpu)
+{
+	struct kthread_worker *worker =
+		kthread_create_worker_on_cpu(cpu, 0, "erofs_worker/%u", cpu);
+
+	if (IS_ERR(worker))
+		return worker;
+	if (IS_ENABLED(CONFIG_EROFS_FS_PCPU_KTHREAD_HIPRI))
+		sched_set_fifo_low(worker->task);
+	return worker;
+}
+
+static int erofs_init_percpu_workers(void)
+{
+	struct kthread_worker *worker;
+	unsigned int cpu;
+
+	z_erofs_pcpu_workers = kcalloc(num_possible_cpus(),
+			sizeof(struct kthread_worker *), GFP_ATOMIC);
+	if (!z_erofs_pcpu_workers)
+		return -ENOMEM;
+
+	for_each_online_cpu(cpu) {	/* could miss cpu{off,on}line? */
+		worker = erofs_init_percpu_worker(cpu);
+		if (!IS_ERR(worker))
+			rcu_assign_pointer(z_erofs_pcpu_workers[cpu], worker);
+	}
+	return 0;
+}
+#else
+static inline void erofs_destroy_percpu_workers(void) {}
+static inline int erofs_init_percpu_workers(void) { return 0; }
+#endif
+
+#if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_EROFS_FS_PCPU_KTHREAD)
+static DEFINE_SPINLOCK(z_erofs_pcpu_worker_lock);
+static enum cpuhp_state erofs_cpuhp_state;
+
+static int erofs_cpu_online(unsigned int cpu)
+{
+	struct kthread_worker *worker, *old;
+
+	worker = erofs_init_percpu_worker(cpu);
+	if (IS_ERR(worker))
+		return PTR_ERR(worker);
+
+	spin_lock(&z_erofs_pcpu_worker_lock);
+	old = rcu_dereference_protected(z_erofs_pcpu_workers[cpu],
+			lockdep_is_held(&z_erofs_pcpu_worker_lock));
+	if (!old)
+		rcu_assign_pointer(z_erofs_pcpu_workers[cpu], worker);
+	spin_unlock(&z_erofs_pcpu_worker_lock);
+	if (old)
+		kthread_destroy_worker(worker);
+	return 0;
+}
+
+static int erofs_cpu_offline(unsigned int cpu)
+{
+	struct kthread_worker *worker;
+
+	spin_lock(&z_erofs_pcpu_worker_lock);
+	worker = rcu_dereference_protected(z_erofs_pcpu_workers[cpu],
+			lockdep_is_held(&z_erofs_pcpu_worker_lock));
+	rcu_assign_pointer(z_erofs_pcpu_workers[cpu], NULL);
+	spin_unlock(&z_erofs_pcpu_worker_lock);
+
+	synchronize_rcu();
+	if (worker)
+		kthread_destroy_worker(worker);
+	return 0;
+}
+
+static int erofs_cpu_hotplug_init(void)
+{
+	int state;
+
+	state = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
+			"fs/erofs:online", erofs_cpu_online, erofs_cpu_offline);
+	if (state < 0)
+		return state;
+
+	erofs_cpuhp_state = state;
+	return 0;
+}
+
+static void erofs_cpu_hotplug_destroy(void)
+{
+	if (erofs_cpuhp_state)
+		cpuhp_remove_state_nocalls(erofs_cpuhp_state);
+}
+#else /* !CONFIG_HOTPLUG_CPU || !CONFIG_EROFS_FS_PCPU_KTHREAD */
+static inline int erofs_cpu_hotplug_init(void) { return 0; }
+static inline void erofs_cpu_hotplug_destroy(void) {}
+#endif
+
+void z_erofs_exit_zip_subsystem(void)
+{
+	erofs_cpu_hotplug_destroy();
+	erofs_destroy_percpu_workers();
+	destroy_workqueue(z_erofs_workqueue);
+	z_erofs_destroy_pcluster_pool();
+}
+
+int __init z_erofs_init_zip_subsystem(void)
+{
+	int err = z_erofs_create_pcluster_pool();
+
+	if (err)
+		goto out_error_pcluster_pool;
+
+	z_erofs_workqueue = alloc_workqueue("erofs_worker",
+			WQ_UNBOUND | WQ_HIGHPRI, num_possible_cpus());
+	if (!z_erofs_workqueue) {
+		err = -ENOMEM;
+		goto out_error_workqueue_init;
+	}
+
+	err = erofs_init_percpu_workers();
+	if (err)
+		goto out_error_pcpu_worker;
+
+	err = erofs_cpu_hotplug_init();
+	if (err < 0)
+		goto out_error_cpuhp_init;
+	return err;
+
+out_error_cpuhp_init:
+	erofs_destroy_percpu_workers();
+out_error_pcpu_worker:
+	destroy_workqueue(z_erofs_workqueue);
+out_error_workqueue_init:
+	z_erofs_destroy_pcluster_pool();
+out_error_pcluster_pool:
+	return err;
+}
+
+enum z_erofs_pclustermode {
+	Z_EROFS_PCLUSTER_INFLIGHT,
+	/*
+	 * a weak form of Z_EROFS_PCLUSTER_FOLLOWED, the difference is that it
+	 * could be dispatched into bypass queue later due to uptodated managed
+	 * pages. All related online pages cannot be reused for inplace I/O (or
+	 * bvpage) since it can be directly decoded without I/O submission.
+	 */
+	Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE,
+	/*
+	 * The pcluster was just linked to a decompression chain by us.  It can
+	 * also be linked with the remaining pclusters, which means if the
+	 * processing page is the tail page of a pcluster, this pcluster can
+	 * safely use the whole page (since the previous pcluster is within the
+	 * same chain) for in-place I/O, as illustrated below:
+	 *  ___________________________________________________
+	 * |  tail (partial) page  |    head (partial) page    |
+	 * |  (of the current pcl) |   (of the previous pcl)   |
+	 * |___PCLUSTER_FOLLOWED___|_____PCLUSTER_FOLLOWED_____|
+	 *
+	 * [  (*) the page above can be used as inplace I/O.   ]
+	 */
+	Z_EROFS_PCLUSTER_FOLLOWED,
+};
+
+struct z_erofs_decompress_frontend {
+	struct inode *const inode;
+	struct erofs_map_blocks map;
+	struct z_erofs_bvec_iter biter;
+
+	struct page *pagepool;
+	struct page *candidate_bvpage;
+	struct z_erofs_pcluster *pcl;
+	z_erofs_next_pcluster_t owned_head;
+	enum z_erofs_pclustermode mode;
+
+	erofs_off_t headoffset;
+
+	/* a pointer used to pick up inplace I/O pages */
+	unsigned int icur;
+};
+
+#define DECOMPRESS_FRONTEND_INIT(__i) { \
+	.inode = __i, .owned_head = Z_EROFS_PCLUSTER_TAIL, \
+	.mode = Z_EROFS_PCLUSTER_FOLLOWED }
+
+static bool z_erofs_should_alloc_cache(struct z_erofs_decompress_frontend *fe)
+{
+	unsigned int cachestrategy = EROFS_I_SB(fe->inode)->opt.cache_strategy;
+
+	if (cachestrategy <= EROFS_ZIP_CACHE_DISABLED)
+		return false;
+
+	if (!(fe->map.m_flags & EROFS_MAP_FULL_MAPPED))
+		return true;
+
+	if (cachestrategy >= EROFS_ZIP_CACHE_READAROUND &&
+	    fe->map.m_la < fe->headoffset)
+		return true;
+
+	return false;
+}
+
+static void z_erofs_bind_cache(struct z_erofs_decompress_frontend *fe)
+{
+	struct address_space *mc = MNGD_MAPPING(EROFS_I_SB(fe->inode));
+	struct z_erofs_pcluster *pcl = fe->pcl;
+	bool shouldalloc = z_erofs_should_alloc_cache(fe);
+	bool standalone = true;
+	/*
+	 * optimistic allocation without direct reclaim since inplace I/O
+	 * can be used if low memory otherwise.
+	 */
+	gfp_t gfp = (mapping_gfp_mask(mc) & ~__GFP_DIRECT_RECLAIM) |
+			__GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN;
+	unsigned int i;
+
+	if (fe->mode < Z_EROFS_PCLUSTER_FOLLOWED)
+		return;
+
+	for (i = 0; i < pcl->pclusterpages; ++i) {
+		struct page *page;
+		void *t;	/* mark pages just found for debugging */
+		struct page *newpage = NULL;
+
+		/* the compressed page was loaded before */
+		if (READ_ONCE(pcl->compressed_bvecs[i].page))
+			continue;
+
+		page = find_get_page(mc, pcl->obj.index + i);
+
+		if (page) {
+			t = (void *)((unsigned long)page | 1);
+		} else {
+			/* I/O is needed, no possible to decompress directly */
+			standalone = false;
+			if (!shouldalloc)
+				continue;
+
+			/*
+			 * try to use cached I/O if page allocation
+			 * succeeds or fallback to in-place I/O instead
+			 * to avoid any direct reclaim.
+			 */
+			newpage = erofs_allocpage(&fe->pagepool, gfp);
+			if (!newpage)
+				continue;
+			set_page_private(newpage, Z_EROFS_PREALLOCATED_PAGE);
+			t = (void *)((unsigned long)newpage | 1);
+		}
+
+		if (!cmpxchg_relaxed(&pcl->compressed_bvecs[i].page, NULL, t))
+			continue;
+
+		if (page)
+			put_page(page);
+		else if (newpage)
+			erofs_pagepool_add(&fe->pagepool, newpage);
+	}
+
+	/*
+	 * don't do inplace I/O if all compressed pages are available in
+	 * managed cache since it can be moved to the bypass queue instead.
+	 */
+	if (standalone)
+		fe->mode = Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE;
+}
+
+/* called by erofs_shrinker to get rid of all compressed_pages */
+int erofs_try_to_free_all_cached_pages(struct erofs_sb_info *sbi,
+				       struct erofs_workgroup *grp)
+{
+	struct z_erofs_pcluster *const pcl =
+		container_of(grp, struct z_erofs_pcluster, obj);
+	int i;
+
+	DBG_BUGON(z_erofs_is_inline_pcluster(pcl));
+	/*
+	 * refcount of workgroup is now freezed as 0,
+	 * therefore no need to worry about available decompression users.
+	 */
+	for (i = 0; i < pcl->pclusterpages; ++i) {
+		struct page *page = pcl->compressed_bvecs[i].page;
+
+		if (!page)
+			continue;
+
+		/* block other users from reclaiming or migrating the page */
+		if (!trylock_page(page))
+			return -EBUSY;
+
+		if (!erofs_page_is_managed(sbi, page))
+			continue;
+
+		/* barrier is implied in the following 'unlock_page' */
+		WRITE_ONCE(pcl->compressed_bvecs[i].page, NULL);
+		detach_page_private(page);
+		unlock_page(page);
+	}
+	return 0;
+}
+
+static bool z_erofs_cache_release_folio(struct folio *folio, gfp_t gfp)
+{
+	struct z_erofs_pcluster *pcl = folio_get_private(folio);
+	bool ret;
+	int i;
+
+	if (!folio_test_private(folio))
+		return true;
+
+	ret = false;
+	spin_lock(&pcl->obj.lockref.lock);
+	if (pcl->obj.lockref.count > 0)
+		goto out;
+
+	DBG_BUGON(z_erofs_is_inline_pcluster(pcl));
+	for (i = 0; i < pcl->pclusterpages; ++i) {
+		if (pcl->compressed_bvecs[i].page == &folio->page) {
+			WRITE_ONCE(pcl->compressed_bvecs[i].page, NULL);
+			ret = true;
+			break;
+		}
+	}
+	if (ret)
+		folio_detach_private(folio);
+out:
+	spin_unlock(&pcl->obj.lockref.lock);
+	return ret;
+}
+
+/*
+ * It will be called only on inode eviction. In case that there are still some
+ * decompression requests in progress, wait with rescheduling for a bit here.
+ * An extra lock could be introduced instead but it seems unnecessary.
+ */
+static void z_erofs_cache_invalidate_folio(struct folio *folio,
+					   size_t offset, size_t length)
+{
+	const size_t stop = length + offset;
+
+	/* Check for potential overflow in debug mode */
+	DBG_BUGON(stop > folio_size(folio) || stop < length);
+
+	if (offset == 0 && stop == folio_size(folio))
+		while (!z_erofs_cache_release_folio(folio, GFP_NOFS))
+			cond_resched();
+}
+
+static const struct address_space_operations z_erofs_cache_aops = {
+	.release_folio = z_erofs_cache_release_folio,
+	.invalidate_folio = z_erofs_cache_invalidate_folio,
+};
+
+int erofs_init_managed_cache(struct super_block *sb)
+{
+	struct inode *const inode = new_inode(sb);
+
+	if (!inode)
+		return -ENOMEM;
+
+	set_nlink(inode, 1);
+	inode->i_size = OFFSET_MAX;
+	inode->i_mapping->a_ops = &z_erofs_cache_aops;
+	mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);
+	EROFS_SB(sb)->managed_cache = inode;
+	return 0;
+}
+
+static bool z_erofs_try_inplace_io(struct z_erofs_decompress_frontend *fe,
+				   struct z_erofs_bvec *bvec)
+{
+	struct z_erofs_pcluster *const pcl = fe->pcl;
+
+	while (fe->icur > 0) {
+		if (!cmpxchg(&pcl->compressed_bvecs[--fe->icur].page,
+			     NULL, bvec->page)) {
+			pcl->compressed_bvecs[fe->icur] = *bvec;
+			return true;
+		}
+	}
+	return false;
+}
+
+/* callers must be with pcluster lock held */
+static int z_erofs_attach_page(struct z_erofs_decompress_frontend *fe,
+			       struct z_erofs_bvec *bvec, bool exclusive)
+{
+	int ret;
+
+	if (exclusive) {
+		/* give priority for inplaceio to use file pages first */
+		if (z_erofs_try_inplace_io(fe, bvec))
+			return 0;
+		/* otherwise, check if it can be used as a bvpage */
+		if (fe->mode >= Z_EROFS_PCLUSTER_FOLLOWED &&
+		    !fe->candidate_bvpage)
+			fe->candidate_bvpage = bvec->page;
+	}
+	ret = z_erofs_bvec_enqueue(&fe->biter, bvec, &fe->candidate_bvpage,
+				   &fe->pagepool);
+	fe->pcl->vcnt += (ret >= 0);
+	return ret;
+}
+
+static void z_erofs_try_to_claim_pcluster(struct z_erofs_decompress_frontend *f)
+{
+	struct z_erofs_pcluster *pcl = f->pcl;
+	z_erofs_next_pcluster_t *owned_head = &f->owned_head;
+
+	/* type 1, nil pcluster (this pcluster doesn't belong to any chain.) */
+	if (cmpxchg(&pcl->next, Z_EROFS_PCLUSTER_NIL,
+		    *owned_head) == Z_EROFS_PCLUSTER_NIL) {
+		*owned_head = &pcl->next;
+		/* so we can attach this pcluster to our submission chain. */
+		f->mode = Z_EROFS_PCLUSTER_FOLLOWED;
+		return;
+	}
+
+	/* type 2, it belongs to an ongoing chain */
+	f->mode = Z_EROFS_PCLUSTER_INFLIGHT;
+}
+
+static int z_erofs_register_pcluster(struct z_erofs_decompress_frontend *fe)
+{
+	struct erofs_map_blocks *map = &fe->map;
+	bool ztailpacking = map->m_flags & EROFS_MAP_META;
+	struct z_erofs_pcluster *pcl;
+	struct erofs_workgroup *grp;
+	int err;
+
+	if (!(map->m_flags & EROFS_MAP_ENCODED) ||
+	    (!ztailpacking && !(map->m_pa >> PAGE_SHIFT))) {
+		DBG_BUGON(1);
+		return -EFSCORRUPTED;
+	}
+
+	/* no available pcluster, let's allocate one */
+	pcl = z_erofs_alloc_pcluster(ztailpacking ? 1 :
+				     map->m_plen >> PAGE_SHIFT);
+	if (IS_ERR(pcl))
+		return PTR_ERR(pcl);
+
+	spin_lock_init(&pcl->obj.lockref.lock);
+	pcl->obj.lockref.count = 1;	/* one ref for this request */
+	pcl->algorithmformat = map->m_algorithmformat;
+	pcl->length = 0;
+	pcl->partial = true;
+
+	/* new pclusters should be claimed as type 1, primary and followed */
+	pcl->next = fe->owned_head;
+	pcl->pageofs_out = map->m_la & ~PAGE_MASK;
+	fe->mode = Z_EROFS_PCLUSTER_FOLLOWED;
+
+	/*
+	 * lock all primary followed works before visible to others
+	 * and mutex_trylock *never* fails for a new pcluster.
+	 */
+	mutex_init(&pcl->lock);
+	DBG_BUGON(!mutex_trylock(&pcl->lock));
+
+	if (ztailpacking) {
+		pcl->obj.index = 0;	/* which indicates ztailpacking */
+		pcl->pageofs_in = erofs_blkoff(fe->inode->i_sb, map->m_pa);
+		pcl->tailpacking_size = map->m_plen;
+	} else {
+		pcl->obj.index = map->m_pa >> PAGE_SHIFT;
+
+		grp = erofs_insert_workgroup(fe->inode->i_sb, &pcl->obj);
+		if (IS_ERR(grp)) {
+			err = PTR_ERR(grp);
+			goto err_out;
+		}
+
+		if (grp != &pcl->obj) {
+			fe->pcl = container_of(grp,
+					struct z_erofs_pcluster, obj);
+			err = -EEXIST;
+			goto err_out;
+		}
+	}
+	fe->owned_head = &pcl->next;
+	fe->pcl = pcl;
+	return 0;
+
+err_out:
+	mutex_unlock(&pcl->lock);
+	z_erofs_free_pcluster(pcl);
+	return err;
+}
+
+static int z_erofs_pcluster_begin(struct z_erofs_decompress_frontend *fe)
+{
+	struct erofs_map_blocks *map = &fe->map;
+	struct super_block *sb = fe->inode->i_sb;
+	erofs_blk_t blknr = erofs_blknr(sb, map->m_pa);
+	struct erofs_workgroup *grp = NULL;
+	int ret;
+
+	DBG_BUGON(fe->pcl);
+
+	/* must be Z_EROFS_PCLUSTER_TAIL or pointed to previous pcluster */
+	DBG_BUGON(fe->owned_head == Z_EROFS_PCLUSTER_NIL);
+
+	if (!(map->m_flags & EROFS_MAP_META)) {
+		grp = erofs_find_workgroup(sb, blknr);
+	} else if ((map->m_pa & ~PAGE_MASK) + map->m_plen > PAGE_SIZE) {
+		DBG_BUGON(1);
+		return -EFSCORRUPTED;
+	}
+
+	if (grp) {
+		fe->pcl = container_of(grp, struct z_erofs_pcluster, obj);
+		ret = -EEXIST;
+	} else {
+		ret = z_erofs_register_pcluster(fe);
+	}
+
+	if (ret == -EEXIST) {
+		mutex_lock(&fe->pcl->lock);
+		z_erofs_try_to_claim_pcluster(fe);
+	} else if (ret) {
+		return ret;
+	}
+
+	z_erofs_bvec_iter_begin(&fe->biter, &fe->pcl->bvset,
+				Z_EROFS_INLINE_BVECS, fe->pcl->vcnt);
+	if (!z_erofs_is_inline_pcluster(fe->pcl)) {
+		/* bind cache first when cached decompression is preferred */
+		z_erofs_bind_cache(fe);
+	} else {
+		void *mptr;
+
+		mptr = erofs_read_metabuf(&map->buf, sb, blknr, EROFS_NO_KMAP);
+		if (IS_ERR(mptr)) {
+			ret = PTR_ERR(mptr);
+			erofs_err(sb, "failed to get inline data %d", ret);
+			return ret;
+		}
+		get_page(map->buf.page);
+		WRITE_ONCE(fe->pcl->compressed_bvecs[0].page, map->buf.page);
+		fe->mode = Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE;
+	}
+	/* file-backed inplace I/O pages are traversed in reverse order */
+	fe->icur = z_erofs_pclusterpages(fe->pcl);
+	return 0;
+}
+
+/*
+ * keep in mind that no referenced pclusters will be freed
+ * only after a RCU grace period.
+ */
+static void z_erofs_rcu_callback(struct rcu_head *head)
+{
+	z_erofs_free_pcluster(container_of(head,
+			struct z_erofs_pcluster, rcu));
+}
+
+void erofs_workgroup_free_rcu(struct erofs_workgroup *grp)
+{
+	struct z_erofs_pcluster *const pcl =
+		container_of(grp, struct z_erofs_pcluster, obj);
+
+	call_rcu(&pcl->rcu, z_erofs_rcu_callback);
+}
+
+static void z_erofs_pcluster_end(struct z_erofs_decompress_frontend *fe)
+{
+	struct z_erofs_pcluster *pcl = fe->pcl;
+
+	if (!pcl)
+		return;
+
+	z_erofs_bvec_iter_end(&fe->biter);
+	mutex_unlock(&pcl->lock);
+
+	if (fe->candidate_bvpage)
+		fe->candidate_bvpage = NULL;
+
+	/*
+	 * if all pending pages are added, don't hold its reference
+	 * any longer if the pcluster isn't hosted by ourselves.
+	 */
+	if (fe->mode < Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE)
+		erofs_workgroup_put(&pcl->obj);
+
+	fe->pcl = NULL;
+}
+
+static int z_erofs_read_fragment(struct super_block *sb, struct page *page,
+			unsigned int cur, unsigned int end, erofs_off_t pos)
+{
+	struct inode *packed_inode = EROFS_SB(sb)->packed_inode;
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	unsigned int cnt;
+	u8 *src;
+
+	if (!packed_inode)
+		return -EFSCORRUPTED;
+
+	buf.inode = packed_inode;
+	for (; cur < end; cur += cnt, pos += cnt) {
+		cnt = min_t(unsigned int, end - cur,
+			    sb->s_blocksize - erofs_blkoff(sb, pos));
+		src = erofs_bread(&buf, erofs_blknr(sb, pos), EROFS_KMAP);
+		if (IS_ERR(src)) {
+			erofs_put_metabuf(&buf);
+			return PTR_ERR(src);
+		}
+		memcpy_to_page(page, cur, src + erofs_blkoff(sb, pos), cnt);
+	}
+	erofs_put_metabuf(&buf);
+	return 0;
+}
+
+static int z_erofs_do_read_page(struct z_erofs_decompress_frontend *fe,
+				struct page *page)
+{
+	struct inode *const inode = fe->inode;
+	struct erofs_map_blocks *const map = &fe->map;
+	const loff_t offset = page_offset(page);
+	bool tight = true, exclusive;
+	unsigned int cur, end, len, split;
+	int err = 0;
+
+	z_erofs_onlinepage_init(page);
+
+	split = 0;
+	end = PAGE_SIZE;
+repeat:
+	if (offset + end - 1 < map->m_la ||
+	    offset + end - 1 >= map->m_la + map->m_llen) {
+		z_erofs_pcluster_end(fe);
+		map->m_la = offset + end - 1;
+		map->m_llen = 0;
+		err = z_erofs_map_blocks_iter(inode, map, 0);
+		if (err)
+			goto out;
+	}
+
+	cur = offset > map->m_la ? 0 : map->m_la - offset;
+	/* bump split parts first to avoid several separate cases */
+	++split;
+
+	if (!(map->m_flags & EROFS_MAP_MAPPED)) {
+		zero_user_segment(page, cur, end);
+		tight = false;
+		goto next_part;
+	}
+
+	if (map->m_flags & EROFS_MAP_FRAGMENT) {
+		erofs_off_t fpos = offset + cur - map->m_la;
+
+		len = min_t(unsigned int, map->m_llen - fpos, end - cur);
+		err = z_erofs_read_fragment(inode->i_sb, page, cur, cur + len,
+				EROFS_I(inode)->z_fragmentoff + fpos);
+		if (err)
+			goto out;
+		tight = false;
+		goto next_part;
+	}
+
+	if (!fe->pcl) {
+		err = z_erofs_pcluster_begin(fe);
+		if (err)
+			goto out;
+	}
+
+	/*
+	 * Ensure the current partial page belongs to this submit chain rather
+	 * than other concurrent submit chains or the noio(bypass) chain since
+	 * those chains are handled asynchronously thus the page cannot be used
+	 * for inplace I/O or bvpage (should be processed in a strict order.)
+	 */
+	tight &= (fe->mode > Z_EROFS_PCLUSTER_FOLLOWED_NOINPLACE);
+	exclusive = (!cur && ((split <= 1) || tight));
+	if (cur)
+		tight &= (fe->mode >= Z_EROFS_PCLUSTER_FOLLOWED);
+
+	err = z_erofs_attach_page(fe, &((struct z_erofs_bvec) {
+					.page = page,
+					.offset = offset - map->m_la,
+					.end = end,
+				  }), exclusive);
+	if (err)
+		goto out;
+
+	z_erofs_onlinepage_split(page);
+	if (fe->pcl->pageofs_out != (map->m_la & ~PAGE_MASK))
+		fe->pcl->multibases = true;
+	if (fe->pcl->length < offset + end - map->m_la) {
+		fe->pcl->length = offset + end - map->m_la;
+		fe->pcl->pageofs_out = map->m_la & ~PAGE_MASK;
+	}
+	if ((map->m_flags & EROFS_MAP_FULL_MAPPED) &&
+	    !(map->m_flags & EROFS_MAP_PARTIAL_REF) &&
+	    fe->pcl->length == map->m_llen)
+		fe->pcl->partial = false;
+next_part:
+	/* shorten the remaining extent to update progress */
+	map->m_llen = offset + cur - map->m_la;
+	map->m_flags &= ~EROFS_MAP_FULL_MAPPED;
+
+	end = cur;
+	if (end > 0)
+		goto repeat;
+
+out:
+	z_erofs_onlinepage_endio(page, err);
+	return err;
+}
+
+static bool z_erofs_is_sync_decompress(struct erofs_sb_info *sbi,
+				       unsigned int readahead_pages)
+{
+	/* auto: enable for read_folio, disable for readahead */
+	if ((sbi->opt.sync_decompress == EROFS_SYNC_DECOMPRESS_AUTO) &&
+	    !readahead_pages)
+		return true;
+
+	if ((sbi->opt.sync_decompress == EROFS_SYNC_DECOMPRESS_FORCE_ON) &&
+	    (readahead_pages <= sbi->opt.max_sync_decompress_pages))
+		return true;
+
+	return false;
+}
+
+static bool z_erofs_page_is_invalidated(struct page *page)
+{
+	return !page->mapping && !z_erofs_is_shortlived_page(page);
+}
+
+struct z_erofs_decompress_backend {
+	struct page *onstack_pages[Z_EROFS_ONSTACK_PAGES];
+	struct super_block *sb;
+	struct z_erofs_pcluster *pcl;
+
+	/* pages with the longest decompressed length for deduplication */
+	struct page **decompressed_pages;
+	/* pages to keep the compressed data */
+	struct page **compressed_pages;
+
+	struct list_head decompressed_secondary_bvecs;
+	struct page **pagepool;
+	unsigned int onstack_used, nr_pages;
+};
+
+struct z_erofs_bvec_item {
+	struct z_erofs_bvec bvec;
+	struct list_head list;
+};
+
+static void z_erofs_do_decompressed_bvec(struct z_erofs_decompress_backend *be,
+					 struct z_erofs_bvec *bvec)
+{
+	struct z_erofs_bvec_item *item;
+	unsigned int pgnr;
+
+	if (!((bvec->offset + be->pcl->pageofs_out) & ~PAGE_MASK) &&
+	    (bvec->end == PAGE_SIZE ||
+	     bvec->offset + bvec->end == be->pcl->length)) {
+		pgnr = (bvec->offset + be->pcl->pageofs_out) >> PAGE_SHIFT;
+		DBG_BUGON(pgnr >= be->nr_pages);
+		if (!be->decompressed_pages[pgnr]) {
+			be->decompressed_pages[pgnr] = bvec->page;
+			return;
+		}
+	}
+
+	/* (cold path) one pcluster is requested multiple times */
+	item = kmalloc(sizeof(*item), GFP_KERNEL | __GFP_NOFAIL);
+	item->bvec = *bvec;
+	list_add(&item->list, &be->decompressed_secondary_bvecs);
+}
+
+static void z_erofs_fill_other_copies(struct z_erofs_decompress_backend *be,
+				      int err)
+{
+	unsigned int off0 = be->pcl->pageofs_out;
+	struct list_head *p, *n;
+
+	list_for_each_safe(p, n, &be->decompressed_secondary_bvecs) {
+		struct z_erofs_bvec_item *bvi;
+		unsigned int end, cur;
+		void *dst, *src;
+
+		bvi = container_of(p, struct z_erofs_bvec_item, list);
+		cur = bvi->bvec.offset < 0 ? -bvi->bvec.offset : 0;
+		end = min_t(unsigned int, be->pcl->length - bvi->bvec.offset,
+			    bvi->bvec.end);
+		dst = kmap_local_page(bvi->bvec.page);
+		while (cur < end) {
+			unsigned int pgnr, scur, len;
+
+			pgnr = (bvi->bvec.offset + cur + off0) >> PAGE_SHIFT;
+			DBG_BUGON(pgnr >= be->nr_pages);
+
+			scur = bvi->bvec.offset + cur -
+					((pgnr << PAGE_SHIFT) - off0);
+			len = min_t(unsigned int, end - cur, PAGE_SIZE - scur);
+			if (!be->decompressed_pages[pgnr]) {
+				err = -EFSCORRUPTED;
+				cur += len;
+				continue;
+			}
+			src = kmap_local_page(be->decompressed_pages[pgnr]);
+			memcpy(dst + cur, src + scur, len);
+			kunmap_local(src);
+			cur += len;
+		}
+		kunmap_local(dst);
+		z_erofs_onlinepage_endio(bvi->bvec.page, err);
+		list_del(p);
+		kfree(bvi);
+	}
+}
+
+static void z_erofs_parse_out_bvecs(struct z_erofs_decompress_backend *be)
+{
+	struct z_erofs_pcluster *pcl = be->pcl;
+	struct z_erofs_bvec_iter biter;
+	struct page *old_bvpage;
+	int i;
+
+	z_erofs_bvec_iter_begin(&biter, &pcl->bvset, Z_EROFS_INLINE_BVECS, 0);
+	for (i = 0; i < pcl->vcnt; ++i) {
+		struct z_erofs_bvec bvec;
+
+		z_erofs_bvec_dequeue(&biter, &bvec, &old_bvpage);
+
+		if (old_bvpage)
+			z_erofs_put_shortlivedpage(be->pagepool, old_bvpage);
+
+		DBG_BUGON(z_erofs_page_is_invalidated(bvec.page));
+		z_erofs_do_decompressed_bvec(be, &bvec);
+	}
+
+	old_bvpage = z_erofs_bvec_iter_end(&biter);
+	if (old_bvpage)
+		z_erofs_put_shortlivedpage(be->pagepool, old_bvpage);
+}
+
+static int z_erofs_parse_in_bvecs(struct z_erofs_decompress_backend *be,
+				  bool *overlapped)
+{
+	struct z_erofs_pcluster *pcl = be->pcl;
+	unsigned int pclusterpages = z_erofs_pclusterpages(pcl);
+	int i, err = 0;
+
+	*overlapped = false;
+	for (i = 0; i < pclusterpages; ++i) {
+		struct z_erofs_bvec *bvec = &pcl->compressed_bvecs[i];
+		struct page *page = bvec->page;
+
+		/* compressed pages ought to be present before decompressing */
+		if (!page) {
+			DBG_BUGON(1);
+			continue;
+		}
+		be->compressed_pages[i] = page;
+
+		if (z_erofs_is_inline_pcluster(pcl)) {
+			if (!PageUptodate(page))
+				err = -EIO;
+			continue;
+		}
+
+		DBG_BUGON(z_erofs_page_is_invalidated(page));
+		if (!z_erofs_is_shortlived_page(page)) {
+			if (erofs_page_is_managed(EROFS_SB(be->sb), page)) {
+				if (!PageUptodate(page))
+					err = -EIO;
+				continue;
+			}
+			z_erofs_do_decompressed_bvec(be, bvec);
+			*overlapped = true;
+		}
+	}
+
+	if (err)
+		return err;
+	return 0;
+}
+
+static int z_erofs_decompress_pcluster(struct z_erofs_decompress_backend *be,
+				       int err)
+{
+	struct erofs_sb_info *const sbi = EROFS_SB(be->sb);
+	struct z_erofs_pcluster *pcl = be->pcl;
+	unsigned int pclusterpages = z_erofs_pclusterpages(pcl);
+	const struct z_erofs_decompressor *decompressor =
+				&erofs_decompressors[pcl->algorithmformat];
+	unsigned int i, inputsize;
+	int err2;
+	struct page *page;
+	bool overlapped;
+
+	mutex_lock(&pcl->lock);
+	be->nr_pages = PAGE_ALIGN(pcl->length + pcl->pageofs_out) >> PAGE_SHIFT;
+
+	/* allocate (de)compressed page arrays if cannot be kept on stack */
+	be->decompressed_pages = NULL;
+	be->compressed_pages = NULL;
+	be->onstack_used = 0;
+	if (be->nr_pages <= Z_EROFS_ONSTACK_PAGES) {
+		be->decompressed_pages = be->onstack_pages;
+		be->onstack_used = be->nr_pages;
+		memset(be->decompressed_pages, 0,
+		       sizeof(struct page *) * be->nr_pages);
+	}
+
+	if (pclusterpages + be->onstack_used <= Z_EROFS_ONSTACK_PAGES)
+		be->compressed_pages = be->onstack_pages + be->onstack_used;
+
+	if (!be->decompressed_pages)
+		be->decompressed_pages =
+			kvcalloc(be->nr_pages, sizeof(struct page *),
+				 GFP_KERNEL | __GFP_NOFAIL);
+	if (!be->compressed_pages)
+		be->compressed_pages =
+			kvcalloc(pclusterpages, sizeof(struct page *),
+				 GFP_KERNEL | __GFP_NOFAIL);
+
+	z_erofs_parse_out_bvecs(be);
+	err2 = z_erofs_parse_in_bvecs(be, &overlapped);
+	if (err2)
+		err = err2;
+	if (err)
+		goto out;
+
+	if (z_erofs_is_inline_pcluster(pcl))
+		inputsize = pcl->tailpacking_size;
+	else
+		inputsize = pclusterpages * PAGE_SIZE;
+
+	err = decompressor->decompress(&(struct z_erofs_decompress_req) {
+					.sb = be->sb,
+					.in = be->compressed_pages,
+					.out = be->decompressed_pages,
+					.pageofs_in = pcl->pageofs_in,
+					.pageofs_out = pcl->pageofs_out,
+					.inputsize = inputsize,
+					.outputsize = pcl->length,
+					.alg = pcl->algorithmformat,
+					.inplace_io = overlapped,
+					.partial_decoding = pcl->partial,
+					.fillgaps = pcl->multibases,
+				 }, be->pagepool);
+
+out:
+	/* must handle all compressed pages before actual file pages */
+	if (z_erofs_is_inline_pcluster(pcl)) {
+		page = pcl->compressed_bvecs[0].page;
+		WRITE_ONCE(pcl->compressed_bvecs[0].page, NULL);
+		put_page(page);
+	} else {
+		for (i = 0; i < pclusterpages; ++i) {
+			/* consider shortlived pages added when decompressing */
+			page = be->compressed_pages[i];
+
+			if (erofs_page_is_managed(sbi, page))
+				continue;
+			(void)z_erofs_put_shortlivedpage(be->pagepool, page);
+			WRITE_ONCE(pcl->compressed_bvecs[i].page, NULL);
+		}
+	}
+	if (be->compressed_pages < be->onstack_pages ||
+	    be->compressed_pages >= be->onstack_pages + Z_EROFS_ONSTACK_PAGES)
+		kvfree(be->compressed_pages);
+	z_erofs_fill_other_copies(be, err);
+
+	for (i = 0; i < be->nr_pages; ++i) {
+		page = be->decompressed_pages[i];
+		if (!page)
+			continue;
+
+		DBG_BUGON(z_erofs_page_is_invalidated(page));
+
+		/* recycle all individual short-lived pages */
+		if (z_erofs_put_shortlivedpage(be->pagepool, page))
+			continue;
+		z_erofs_onlinepage_endio(page, err);
+	}
+
+	if (be->decompressed_pages != be->onstack_pages)
+		kvfree(be->decompressed_pages);
+
+	pcl->length = 0;
+	pcl->partial = true;
+	pcl->multibases = false;
+	pcl->bvset.nextpage = NULL;
+	pcl->vcnt = 0;
+
+	/* pcluster lock MUST be taken before the following line */
+	WRITE_ONCE(pcl->next, Z_EROFS_PCLUSTER_NIL);
+	mutex_unlock(&pcl->lock);
+	return err;
+}
+
+static void z_erofs_decompress_queue(const struct z_erofs_decompressqueue *io,
+				     struct page **pagepool)
+{
+	struct z_erofs_decompress_backend be = {
+		.sb = io->sb,
+		.pagepool = pagepool,
+		.decompressed_secondary_bvecs =
+			LIST_HEAD_INIT(be.decompressed_secondary_bvecs),
+	};
+	z_erofs_next_pcluster_t owned = io->head;
+
+	while (owned != Z_EROFS_PCLUSTER_TAIL) {
+		DBG_BUGON(owned == Z_EROFS_PCLUSTER_NIL);
+
+		be.pcl = container_of(owned, struct z_erofs_pcluster, next);
+		owned = READ_ONCE(be.pcl->next);
+
+		z_erofs_decompress_pcluster(&be, io->eio ? -EIO : 0);
+		if (z_erofs_is_inline_pcluster(be.pcl))
+			z_erofs_free_pcluster(be.pcl);
+		else
+			erofs_workgroup_put(&be.pcl->obj);
+	}
+}
+
+static void z_erofs_decompressqueue_work(struct work_struct *work)
+{
+	struct z_erofs_decompressqueue *bgq =
+		container_of(work, struct z_erofs_decompressqueue, u.work);
+	struct page *pagepool = NULL;
+
+	DBG_BUGON(bgq->head == Z_EROFS_PCLUSTER_TAIL);
+	z_erofs_decompress_queue(bgq, &pagepool);
+	erofs_release_pages(&pagepool);
+	kvfree(bgq);
+}
+
+#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD
+static void z_erofs_decompressqueue_kthread_work(struct kthread_work *work)
+{
+	z_erofs_decompressqueue_work((struct work_struct *)work);
+}
+#endif
+
+static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io,
+				       int bios)
+{
+	struct erofs_sb_info *const sbi = EROFS_SB(io->sb);
+
+	/* wake up the caller thread for sync decompression */
+	if (io->sync) {
+		if (!atomic_add_return(bios, &io->pending_bios))
+			complete(&io->u.done);
+		return;
+	}
+
+	if (atomic_add_return(bios, &io->pending_bios))
+		return;
+	/* Use (kthread_)work and sync decompression for atomic contexts only */
+	if (!in_task() || irqs_disabled() || rcu_read_lock_any_held()) {
+#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD
+		struct kthread_worker *worker;
+
+		rcu_read_lock();
+		worker = rcu_dereference(
+				z_erofs_pcpu_workers[raw_smp_processor_id()]);
+		if (!worker) {
+			INIT_WORK(&io->u.work, z_erofs_decompressqueue_work);
+			queue_work(z_erofs_workqueue, &io->u.work);
+		} else {
+			kthread_queue_work(worker, &io->u.kthread_work);
+		}
+		rcu_read_unlock();
+#else
+		queue_work(z_erofs_workqueue, &io->u.work);
+#endif
+		/* enable sync decompression for readahead */
+		if (sbi->opt.sync_decompress == EROFS_SYNC_DECOMPRESS_AUTO)
+			sbi->opt.sync_decompress = EROFS_SYNC_DECOMPRESS_FORCE_ON;
+		return;
+	}
+	z_erofs_decompressqueue_work(&io->u.work);
+}
+
+static struct page *pickup_page_for_submission(struct z_erofs_pcluster *pcl,
+					       unsigned int nr,
+					       struct page **pagepool,
+					       struct address_space *mc)
+{
+	const pgoff_t index = pcl->obj.index;
+	gfp_t gfp = mapping_gfp_mask(mc);
+	bool tocache = false;
+
+	struct address_space *mapping;
+	struct page *oldpage, *page;
+	int justfound;
+
+repeat:
+	page = READ_ONCE(pcl->compressed_bvecs[nr].page);
+	oldpage = page;
+
+	if (!page)
+		goto out_allocpage;
+
+	justfound = (unsigned long)page & 1UL;
+	page = (struct page *)((unsigned long)page & ~1UL);
+
+	/*
+	 * preallocated cached pages, which is used to avoid direct reclaim
+	 * otherwise, it will go inplace I/O path instead.
+	 */
+	if (page->private == Z_EROFS_PREALLOCATED_PAGE) {
+		WRITE_ONCE(pcl->compressed_bvecs[nr].page, page);
+		set_page_private(page, 0);
+		tocache = true;
+		goto out_tocache;
+	}
+	mapping = READ_ONCE(page->mapping);
+
+	/*
+	 * file-backed online pages in plcuster are all locked steady,
+	 * therefore it is impossible for `mapping' to be NULL.
+	 */
+	if (mapping && mapping != mc)
+		/* ought to be unmanaged pages */
+		goto out;
+
+	/* directly return for shortlived page as well */
+	if (z_erofs_is_shortlived_page(page))
+		goto out;
+
+	lock_page(page);
+
+	/* only true if page reclaim goes wrong, should never happen */
+	DBG_BUGON(justfound && PagePrivate(page));
+
+	/* the page is still in manage cache */
+	if (page->mapping == mc) {
+		WRITE_ONCE(pcl->compressed_bvecs[nr].page, page);
+
+		if (!PagePrivate(page)) {
+			/*
+			 * impossible to be !PagePrivate(page) for
+			 * the current restriction as well if
+			 * the page is already in compressed_bvecs[].
+			 */
+			DBG_BUGON(!justfound);
+
+			justfound = 0;
+			set_page_private(page, (unsigned long)pcl);
+			SetPagePrivate(page);
+		}
+
+		/* no need to submit io if it is already up-to-date */
+		if (PageUptodate(page)) {
+			unlock_page(page);
+			page = NULL;
+		}
+		goto out;
+	}
+
+	/*
+	 * the managed page has been truncated, it's unsafe to
+	 * reuse this one, let's allocate a new cache-managed page.
+	 */
+	DBG_BUGON(page->mapping);
+	DBG_BUGON(!justfound);
+
+	tocache = true;
+	unlock_page(page);
+	put_page(page);
+out_allocpage:
+	page = erofs_allocpage(pagepool, gfp | __GFP_NOFAIL);
+	if (oldpage != cmpxchg(&pcl->compressed_bvecs[nr].page,
+			       oldpage, page)) {
+		erofs_pagepool_add(pagepool, page);
+		cond_resched();
+		goto repeat;
+	}
+out_tocache:
+	if (!tocache || add_to_page_cache_lru(page, mc, index + nr, gfp)) {
+		/* turn into temporary page if fails (1 ref) */
+		set_page_private(page, Z_EROFS_SHORTLIVED_PAGE);
+		goto out;
+	}
+	attach_page_private(page, pcl);
+	/* drop a refcount added by allocpage (then we have 2 refs here) */
+	put_page(page);
+
+out:	/* the only exit (for tracing and debugging) */
+	return page;
+}
+
+static struct z_erofs_decompressqueue *jobqueue_init(struct super_block *sb,
+			      struct z_erofs_decompressqueue *fgq, bool *fg)
+{
+	struct z_erofs_decompressqueue *q;
+
+	if (fg && !*fg) {
+		q = kvzalloc(sizeof(*q), GFP_KERNEL | __GFP_NOWARN);
+		if (!q) {
+			*fg = true;
+			goto fg_out;
+		}
+#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD
+		kthread_init_work(&q->u.kthread_work,
+				  z_erofs_decompressqueue_kthread_work);
+#else
+		INIT_WORK(&q->u.work, z_erofs_decompressqueue_work);
+#endif
+	} else {
+fg_out:
+		q = fgq;
+		init_completion(&fgq->u.done);
+		atomic_set(&fgq->pending_bios, 0);
+		q->eio = false;
+		q->sync = true;
+	}
+	q->sb = sb;
+	q->head = Z_EROFS_PCLUSTER_TAIL;
+	return q;
+}
+
+/* define decompression jobqueue types */
+enum {
+	JQ_BYPASS,
+	JQ_SUBMIT,
+	NR_JOBQUEUES,
+};
+
+static void move_to_bypass_jobqueue(struct z_erofs_pcluster *pcl,
+				    z_erofs_next_pcluster_t qtail[],
+				    z_erofs_next_pcluster_t owned_head)
+{
+	z_erofs_next_pcluster_t *const submit_qtail = qtail[JQ_SUBMIT];
+	z_erofs_next_pcluster_t *const bypass_qtail = qtail[JQ_BYPASS];
+
+	WRITE_ONCE(pcl->next, Z_EROFS_PCLUSTER_TAIL);
+
+	WRITE_ONCE(*submit_qtail, owned_head);
+	WRITE_ONCE(*bypass_qtail, &pcl->next);
+
+	qtail[JQ_BYPASS] = &pcl->next;
+}
+
+static void z_erofs_decompressqueue_endio(struct bio *bio)
+{
+	struct z_erofs_decompressqueue *q = bio->bi_private;
+	blk_status_t err = bio->bi_status;
+	struct bio_vec *bvec;
+	struct bvec_iter_all iter_all;
+
+	bio_for_each_segment_all(bvec, bio, iter_all) {
+		struct page *page = bvec->bv_page;
+
+		DBG_BUGON(PageUptodate(page));
+		DBG_BUGON(z_erofs_page_is_invalidated(page));
+
+		if (erofs_page_is_managed(EROFS_SB(q->sb), page)) {
+			if (!err)
+				SetPageUptodate(page);
+			unlock_page(page);
+		}
+	}
+	if (err)
+		q->eio = true;
+	z_erofs_decompress_kickoff(q, -1);
+	bio_put(bio);
+}
+
+static void z_erofs_submit_queue(struct z_erofs_decompress_frontend *f,
+				 struct z_erofs_decompressqueue *fgq,
+				 bool *force_fg, bool readahead)
+{
+	struct super_block *sb = f->inode->i_sb;
+	struct address_space *mc = MNGD_MAPPING(EROFS_SB(sb));
+	z_erofs_next_pcluster_t qtail[NR_JOBQUEUES];
+	struct z_erofs_decompressqueue *q[NR_JOBQUEUES];
+	z_erofs_next_pcluster_t owned_head = f->owned_head;
+	/* bio is NULL initially, so no need to initialize last_{index,bdev} */
+	pgoff_t last_index;
+	struct block_device *last_bdev;
+	unsigned int nr_bios = 0;
+	struct bio *bio = NULL;
+	unsigned long pflags;
+	int memstall = 0;
+
+	/*
+	 * if managed cache is enabled, bypass jobqueue is needed,
+	 * no need to read from device for all pclusters in this queue.
+	 */
+	q[JQ_BYPASS] = jobqueue_init(sb, fgq + JQ_BYPASS, NULL);
+	q[JQ_SUBMIT] = jobqueue_init(sb, fgq + JQ_SUBMIT, force_fg);
+
+	qtail[JQ_BYPASS] = &q[JQ_BYPASS]->head;
+	qtail[JQ_SUBMIT] = &q[JQ_SUBMIT]->head;
+
+	/* by default, all need io submission */
+	q[JQ_SUBMIT]->head = owned_head;
+
+	do {
+		struct erofs_map_dev mdev;
+		struct z_erofs_pcluster *pcl;
+		pgoff_t cur, end;
+		unsigned int i = 0;
+		bool bypass = true;
+
+		DBG_BUGON(owned_head == Z_EROFS_PCLUSTER_NIL);
+		pcl = container_of(owned_head, struct z_erofs_pcluster, next);
+		owned_head = READ_ONCE(pcl->next);
+
+		if (z_erofs_is_inline_pcluster(pcl)) {
+			move_to_bypass_jobqueue(pcl, qtail, owned_head);
+			continue;
+		}
+
+		/* no device id here, thus it will always succeed */
+		mdev = (struct erofs_map_dev) {
+			.m_pa = erofs_pos(sb, pcl->obj.index),
+		};
+		(void)erofs_map_dev(sb, &mdev);
+
+		cur = erofs_blknr(sb, mdev.m_pa);
+		end = cur + pcl->pclusterpages;
+
+		do {
+			struct page *page;
+
+			page = pickup_page_for_submission(pcl, i++,
+					&f->pagepool, mc);
+			if (!page)
+				continue;
+
+			if (bio && (cur != last_index + 1 ||
+				    last_bdev != mdev.m_bdev)) {
+submit_bio_retry:
+				submit_bio(bio);
+				if (memstall) {
+					psi_memstall_leave(&pflags);
+					memstall = 0;
+				}
+				bio = NULL;
+			}
+
+			if (unlikely(PageWorkingset(page)) && !memstall) {
+				psi_memstall_enter(&pflags);
+				memstall = 1;
+			}
+
+			if (!bio) {
+				bio = bio_alloc(mdev.m_bdev, BIO_MAX_VECS,
+						REQ_OP_READ, GFP_NOIO);
+				bio->bi_end_io = z_erofs_decompressqueue_endio;
+
+				last_bdev = mdev.m_bdev;
+				bio->bi_iter.bi_sector = (sector_t)cur <<
+					(sb->s_blocksize_bits - 9);
+				bio->bi_private = q[JQ_SUBMIT];
+				if (readahead)
+					bio->bi_opf |= REQ_RAHEAD;
+				++nr_bios;
+			}
+
+			if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE)
+				goto submit_bio_retry;
+
+			last_index = cur;
+			bypass = false;
+		} while (++cur < end);
+
+		if (!bypass)
+			qtail[JQ_SUBMIT] = &pcl->next;
+		else
+			move_to_bypass_jobqueue(pcl, qtail, owned_head);
+	} while (owned_head != Z_EROFS_PCLUSTER_TAIL);
+
+	if (bio) {
+		submit_bio(bio);
+		if (memstall)
+			psi_memstall_leave(&pflags);
+	}
+
+	/*
+	 * although background is preferred, no one is pending for submission.
+	 * don't issue decompression but drop it directly instead.
+	 */
+	if (!*force_fg && !nr_bios) {
+		kvfree(q[JQ_SUBMIT]);
+		return;
+	}
+	z_erofs_decompress_kickoff(q[JQ_SUBMIT], nr_bios);
+}
+
+static void z_erofs_runqueue(struct z_erofs_decompress_frontend *f,
+			     bool force_fg, bool ra)
+{
+	struct z_erofs_decompressqueue io[NR_JOBQUEUES];
+
+	if (f->owned_head == Z_EROFS_PCLUSTER_TAIL)
+		return;
+	z_erofs_submit_queue(f, io, &force_fg, ra);
+
+	/* handle bypass queue (no i/o pclusters) immediately */
+	z_erofs_decompress_queue(&io[JQ_BYPASS], &f->pagepool);
+
+	if (!force_fg)
+		return;
+
+	/* wait until all bios are completed */
+	wait_for_completion_io(&io[JQ_SUBMIT].u.done);
+
+	/* handle synchronous decompress queue in the caller context */
+	z_erofs_decompress_queue(&io[JQ_SUBMIT], &f->pagepool);
+}
+
+/*
+ * Since partial uptodate is still unimplemented for now, we have to use
+ * approximate readmore strategies as a start.
+ */
+static void z_erofs_pcluster_readmore(struct z_erofs_decompress_frontend *f,
+		struct readahead_control *rac, bool backmost)
+{
+	struct inode *inode = f->inode;
+	struct erofs_map_blocks *map = &f->map;
+	erofs_off_t cur, end, headoffset = f->headoffset;
+	int err;
+
+	if (backmost) {
+		if (rac)
+			end = headoffset + readahead_length(rac) - 1;
+		else
+			end = headoffset + PAGE_SIZE - 1;
+		map->m_la = end;
+		err = z_erofs_map_blocks_iter(inode, map,
+					      EROFS_GET_BLOCKS_READMORE);
+		if (err)
+			return;
+
+		/* expand ra for the trailing edge if readahead */
+		if (rac) {
+			cur = round_up(map->m_la + map->m_llen, PAGE_SIZE);
+			readahead_expand(rac, headoffset, cur - headoffset);
+			return;
+		}
+		end = round_up(end, PAGE_SIZE);
+	} else {
+		end = round_up(map->m_la, PAGE_SIZE);
+
+		if (!map->m_llen)
+			return;
+	}
+
+	cur = map->m_la + map->m_llen - 1;
+	while ((cur >= end) && (cur < i_size_read(inode))) {
+		pgoff_t index = cur >> PAGE_SHIFT;
+		struct page *page;
+
+		page = erofs_grab_cache_page_nowait(inode->i_mapping, index);
+		if (page) {
+			if (PageUptodate(page))
+				unlock_page(page);
+			else
+				(void)z_erofs_do_read_page(f, page);
+			put_page(page);
+		}
+
+		if (cur < PAGE_SIZE)
+			break;
+		cur = (index << PAGE_SHIFT) - 1;
+	}
+}
+
+static int z_erofs_read_folio(struct file *file, struct folio *folio)
+{
+	struct inode *const inode = folio->mapping->host;
+	struct erofs_sb_info *const sbi = EROFS_I_SB(inode);
+	struct z_erofs_decompress_frontend f = DECOMPRESS_FRONTEND_INIT(inode);
+	int err;
+
+	trace_erofs_read_folio(folio, false);
+	f.headoffset = (erofs_off_t)folio->index << PAGE_SHIFT;
+
+	z_erofs_pcluster_readmore(&f, NULL, true);
+	err = z_erofs_do_read_page(&f, &folio->page);
+	z_erofs_pcluster_readmore(&f, NULL, false);
+	z_erofs_pcluster_end(&f);
+
+	/* if some compressed cluster ready, need submit them anyway */
+	z_erofs_runqueue(&f, z_erofs_is_sync_decompress(sbi, 0), false);
+
+	if (err && err != -EINTR)
+		erofs_err(inode->i_sb, "read error %d @ %lu of nid %llu",
+			  err, folio->index, EROFS_I(inode)->nid);
+
+	erofs_put_metabuf(&f.map.buf);
+	erofs_release_pages(&f.pagepool);
+	return err;
+}
+
+static void z_erofs_readahead(struct readahead_control *rac)
+{
+	struct inode *const inode = rac->mapping->host;
+	struct erofs_sb_info *const sbi = EROFS_I_SB(inode);
+	struct z_erofs_decompress_frontend f = DECOMPRESS_FRONTEND_INIT(inode);
+	struct folio *head = NULL, *folio;
+	unsigned int nr_folios;
+	int err;
+
+	f.headoffset = readahead_pos(rac);
+
+	z_erofs_pcluster_readmore(&f, rac, true);
+	nr_folios = readahead_count(rac);
+	trace_erofs_readpages(inode, readahead_index(rac), nr_folios, false);
+
+	while ((folio = readahead_folio(rac))) {
+		folio->private = head;
+		head = folio;
+	}
+
+	/* traverse in reverse order for best metadata I/O performance */
+	while (head) {
+		folio = head;
+		head = folio_get_private(folio);
+
+		err = z_erofs_do_read_page(&f, &folio->page);
+		if (err && err != -EINTR)
+			erofs_err(inode->i_sb, "readahead error at folio %lu @ nid %llu",
+				  folio->index, EROFS_I(inode)->nid);
+	}
+	z_erofs_pcluster_readmore(&f, rac, false);
+	z_erofs_pcluster_end(&f);
+
+	z_erofs_runqueue(&f, z_erofs_is_sync_decompress(sbi, nr_folios), true);
+	erofs_put_metabuf(&f.map.buf);
+	erofs_release_pages(&f.pagepool);
+}
+
+const struct address_space_operations z_erofs_aops = {
+	.read_folio = z_erofs_read_folio,
+	.readahead = z_erofs_readahead,
+};
diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
new file mode 100644
index 0000000000..7a1a24ae4a
--- /dev/null
+++ b/fs/erofs/zmap.c
@@ -0,0 +1,774 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2018-2019 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ */
+#include "internal.h"
+#include <asm/unaligned.h>
+#include <trace/events/erofs.h>
+
+struct z_erofs_maprecorder {
+	struct inode *inode;
+	struct erofs_map_blocks *map;
+	void *kaddr;
+
+	unsigned long lcn;
+	/* compression extent information gathered */
+	u8  type, headtype;
+	u16 clusterofs;
+	u16 delta[2];
+	erofs_blk_t pblk, compressedblks;
+	erofs_off_t nextpackoff;
+	bool partialref;
+};
+
+static int z_erofs_load_full_lcluster(struct z_erofs_maprecorder *m,
+				      unsigned long lcn)
+{
+	struct inode *const inode = m->inode;
+	struct erofs_inode *const vi = EROFS_I(inode);
+	const erofs_off_t pos = Z_EROFS_FULL_INDEX_ALIGN(erofs_iloc(inode) +
+			vi->inode_isize + vi->xattr_isize) +
+			lcn * sizeof(struct z_erofs_lcluster_index);
+	struct z_erofs_lcluster_index *di;
+	unsigned int advise, type;
+
+	m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb,
+				      erofs_blknr(inode->i_sb, pos), EROFS_KMAP);
+	if (IS_ERR(m->kaddr))
+		return PTR_ERR(m->kaddr);
+
+	m->nextpackoff = pos + sizeof(struct z_erofs_lcluster_index);
+	m->lcn = lcn;
+	di = m->kaddr + erofs_blkoff(inode->i_sb, pos);
+
+	advise = le16_to_cpu(di->di_advise);
+	type = (advise >> Z_EROFS_LI_LCLUSTER_TYPE_BIT) &
+		((1 << Z_EROFS_LI_LCLUSTER_TYPE_BITS) - 1);
+	switch (type) {
+	case Z_EROFS_LCLUSTER_TYPE_NONHEAD:
+		m->clusterofs = 1 << vi->z_logical_clusterbits;
+		m->delta[0] = le16_to_cpu(di->di_u.delta[0]);
+		if (m->delta[0] & Z_EROFS_LI_D0_CBLKCNT) {
+			if (!(vi->z_advise & (Z_EROFS_ADVISE_BIG_PCLUSTER_1 |
+					Z_EROFS_ADVISE_BIG_PCLUSTER_2))) {
+				DBG_BUGON(1);
+				return -EFSCORRUPTED;
+			}
+			m->compressedblks = m->delta[0] &
+				~Z_EROFS_LI_D0_CBLKCNT;
+			m->delta[0] = 1;
+		}
+		m->delta[1] = le16_to_cpu(di->di_u.delta[1]);
+		break;
+	case Z_EROFS_LCLUSTER_TYPE_PLAIN:
+	case Z_EROFS_LCLUSTER_TYPE_HEAD1:
+	case Z_EROFS_LCLUSTER_TYPE_HEAD2:
+		if (advise & Z_EROFS_LI_PARTIAL_REF)
+			m->partialref = true;
+		m->clusterofs = le16_to_cpu(di->di_clusterofs);
+		if (m->clusterofs >= 1 << vi->z_logical_clusterbits) {
+			DBG_BUGON(1);
+			return -EFSCORRUPTED;
+		}
+		m->pblk = le32_to_cpu(di->di_u.blkaddr);
+		break;
+	default:
+		DBG_BUGON(1);
+		return -EOPNOTSUPP;
+	}
+	m->type = type;
+	return 0;
+}
+
+static unsigned int decode_compactedbits(unsigned int lobits,
+					 unsigned int lomask,
+					 u8 *in, unsigned int pos, u8 *type)
+{
+	const unsigned int v = get_unaligned_le32(in + pos / 8) >> (pos & 7);
+	const unsigned int lo = v & lomask;
+
+	*type = (v >> lobits) & 3;
+	return lo;
+}
+
+static int get_compacted_la_distance(unsigned int lclusterbits,
+				     unsigned int encodebits,
+				     unsigned int vcnt, u8 *in, int i)
+{
+	const unsigned int lomask = (1 << lclusterbits) - 1;
+	unsigned int lo, d1 = 0;
+	u8 type;
+
+	DBG_BUGON(i >= vcnt);
+
+	do {
+		lo = decode_compactedbits(lclusterbits, lomask,
+					  in, encodebits * i, &type);
+
+		if (type != Z_EROFS_LCLUSTER_TYPE_NONHEAD)
+			return d1;
+		++d1;
+	} while (++i < vcnt);
+
+	/* vcnt - 1 (Z_EROFS_LCLUSTER_TYPE_NONHEAD) item */
+	if (!(lo & Z_EROFS_LI_D0_CBLKCNT))
+		d1 += lo - 1;
+	return d1;
+}
+
+static int unpack_compacted_index(struct z_erofs_maprecorder *m,
+				  unsigned int amortizedshift,
+				  erofs_off_t pos, bool lookahead)
+{
+	struct erofs_inode *const vi = EROFS_I(m->inode);
+	const unsigned int lclusterbits = vi->z_logical_clusterbits;
+	const unsigned int lomask = (1 << lclusterbits) - 1;
+	unsigned int vcnt, base, lo, encodebits, nblk, eofs;
+	int i;
+	u8 *in, type;
+	bool big_pcluster;
+
+	if (1 << amortizedshift == 4 && lclusterbits <= 14)
+		vcnt = 2;
+	else if (1 << amortizedshift == 2 && lclusterbits == 12)
+		vcnt = 16;
+	else
+		return -EOPNOTSUPP;
+
+	/* it doesn't equal to round_up(..) */
+	m->nextpackoff = round_down(pos, vcnt << amortizedshift) +
+			 (vcnt << amortizedshift);
+	big_pcluster = vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_1;
+	encodebits = ((vcnt << amortizedshift) - sizeof(__le32)) * 8 / vcnt;
+	eofs = erofs_blkoff(m->inode->i_sb, pos);
+	base = round_down(eofs, vcnt << amortizedshift);
+	in = m->kaddr + base;
+
+	i = (eofs - base) >> amortizedshift;
+
+	lo = decode_compactedbits(lclusterbits, lomask,
+				  in, encodebits * i, &type);
+	m->type = type;
+	if (type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+		m->clusterofs = 1 << lclusterbits;
+
+		/* figure out lookahead_distance: delta[1] if needed */
+		if (lookahead)
+			m->delta[1] = get_compacted_la_distance(lclusterbits,
+						encodebits, vcnt, in, i);
+		if (lo & Z_EROFS_LI_D0_CBLKCNT) {
+			if (!big_pcluster) {
+				DBG_BUGON(1);
+				return -EFSCORRUPTED;
+			}
+			m->compressedblks = lo & ~Z_EROFS_LI_D0_CBLKCNT;
+			m->delta[0] = 1;
+			return 0;
+		} else if (i + 1 != (int)vcnt) {
+			m->delta[0] = lo;
+			return 0;
+		}
+		/*
+		 * since the last lcluster in the pack is special,
+		 * of which lo saves delta[1] rather than delta[0].
+		 * Hence, get delta[0] by the previous lcluster indirectly.
+		 */
+		lo = decode_compactedbits(lclusterbits, lomask,
+					  in, encodebits * (i - 1), &type);
+		if (type != Z_EROFS_LCLUSTER_TYPE_NONHEAD)
+			lo = 0;
+		else if (lo & Z_EROFS_LI_D0_CBLKCNT)
+			lo = 1;
+		m->delta[0] = lo + 1;
+		return 0;
+	}
+	m->clusterofs = lo;
+	m->delta[0] = 0;
+	/* figout out blkaddr (pblk) for HEAD lclusters */
+	if (!big_pcluster) {
+		nblk = 1;
+		while (i > 0) {
+			--i;
+			lo = decode_compactedbits(lclusterbits, lomask,
+						  in, encodebits * i, &type);
+			if (type == Z_EROFS_LCLUSTER_TYPE_NONHEAD)
+				i -= lo;
+
+			if (i >= 0)
+				++nblk;
+		}
+	} else {
+		nblk = 0;
+		while (i > 0) {
+			--i;
+			lo = decode_compactedbits(lclusterbits, lomask,
+						  in, encodebits * i, &type);
+			if (type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+				if (lo & Z_EROFS_LI_D0_CBLKCNT) {
+					--i;
+					nblk += lo & ~Z_EROFS_LI_D0_CBLKCNT;
+					continue;
+				}
+				/* bigpcluster shouldn't have plain d0 == 1 */
+				if (lo <= 1) {
+					DBG_BUGON(1);
+					return -EFSCORRUPTED;
+				}
+				i -= lo - 2;
+				continue;
+			}
+			++nblk;
+		}
+	}
+	in += (vcnt << amortizedshift) - sizeof(__le32);
+	m->pblk = le32_to_cpu(*(__le32 *)in) + nblk;
+	return 0;
+}
+
+static int z_erofs_load_compact_lcluster(struct z_erofs_maprecorder *m,
+					 unsigned long lcn, bool lookahead)
+{
+	struct inode *const inode = m->inode;
+	struct erofs_inode *const vi = EROFS_I(inode);
+	const erofs_off_t ebase = sizeof(struct z_erofs_map_header) +
+		ALIGN(erofs_iloc(inode) + vi->inode_isize + vi->xattr_isize, 8);
+	unsigned int totalidx = erofs_iblks(inode);
+	unsigned int compacted_4b_initial, compacted_2b;
+	unsigned int amortizedshift;
+	erofs_off_t pos;
+
+	if (lcn >= totalidx)
+		return -EINVAL;
+
+	m->lcn = lcn;
+	/* used to align to 32-byte (compacted_2b) alignment */
+	compacted_4b_initial = (32 - ebase % 32) / 4;
+	if (compacted_4b_initial == 32 / 4)
+		compacted_4b_initial = 0;
+
+	if ((vi->z_advise & Z_EROFS_ADVISE_COMPACTED_2B) &&
+	    compacted_4b_initial < totalidx)
+		compacted_2b = rounddown(totalidx - compacted_4b_initial, 16);
+	else
+		compacted_2b = 0;
+
+	pos = ebase;
+	if (lcn < compacted_4b_initial) {
+		amortizedshift = 2;
+		goto out;
+	}
+	pos += compacted_4b_initial * 4;
+	lcn -= compacted_4b_initial;
+
+	if (lcn < compacted_2b) {
+		amortizedshift = 1;
+		goto out;
+	}
+	pos += compacted_2b * 2;
+	lcn -= compacted_2b;
+	amortizedshift = 2;
+out:
+	pos += lcn * (1 << amortizedshift);
+	m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb,
+				      erofs_blknr(inode->i_sb, pos), EROFS_KMAP);
+	if (IS_ERR(m->kaddr))
+		return PTR_ERR(m->kaddr);
+	return unpack_compacted_index(m, amortizedshift, pos, lookahead);
+}
+
+static int z_erofs_load_lcluster_from_disk(struct z_erofs_maprecorder *m,
+					   unsigned int lcn, bool lookahead)
+{
+	switch (EROFS_I(m->inode)->datalayout) {
+	case EROFS_INODE_COMPRESSED_FULL:
+		return z_erofs_load_full_lcluster(m, lcn);
+	case EROFS_INODE_COMPRESSED_COMPACT:
+		return z_erofs_load_compact_lcluster(m, lcn, lookahead);
+	default:
+		return -EINVAL;
+	}
+}
+
+static int z_erofs_extent_lookback(struct z_erofs_maprecorder *m,
+				   unsigned int lookback_distance)
+{
+	struct super_block *sb = m->inode->i_sb;
+	struct erofs_inode *const vi = EROFS_I(m->inode);
+	const unsigned int lclusterbits = vi->z_logical_clusterbits;
+
+	while (m->lcn >= lookback_distance) {
+		unsigned long lcn = m->lcn - lookback_distance;
+		int err;
+
+		err = z_erofs_load_lcluster_from_disk(m, lcn, false);
+		if (err)
+			return err;
+
+		switch (m->type) {
+		case Z_EROFS_LCLUSTER_TYPE_NONHEAD:
+			lookback_distance = m->delta[0];
+			if (!lookback_distance)
+				goto err_bogus;
+			continue;
+		case Z_EROFS_LCLUSTER_TYPE_PLAIN:
+		case Z_EROFS_LCLUSTER_TYPE_HEAD1:
+		case Z_EROFS_LCLUSTER_TYPE_HEAD2:
+			m->headtype = m->type;
+			m->map->m_la = (lcn << lclusterbits) | m->clusterofs;
+			return 0;
+		default:
+			erofs_err(sb, "unknown type %u @ lcn %lu of nid %llu",
+				  m->type, lcn, vi->nid);
+			DBG_BUGON(1);
+			return -EOPNOTSUPP;
+		}
+	}
+err_bogus:
+	erofs_err(sb, "bogus lookback distance %u @ lcn %lu of nid %llu",
+		  lookback_distance, m->lcn, vi->nid);
+	DBG_BUGON(1);
+	return -EFSCORRUPTED;
+}
+
+static int z_erofs_get_extent_compressedlen(struct z_erofs_maprecorder *m,
+					    unsigned int initial_lcn)
+{
+	struct super_block *sb = m->inode->i_sb;
+	struct erofs_inode *const vi = EROFS_I(m->inode);
+	struct erofs_map_blocks *const map = m->map;
+	const unsigned int lclusterbits = vi->z_logical_clusterbits;
+	unsigned long lcn;
+	int err;
+
+	DBG_BUGON(m->type != Z_EROFS_LCLUSTER_TYPE_PLAIN &&
+		  m->type != Z_EROFS_LCLUSTER_TYPE_HEAD1 &&
+		  m->type != Z_EROFS_LCLUSTER_TYPE_HEAD2);
+	DBG_BUGON(m->type != m->headtype);
+
+	if (m->headtype == Z_EROFS_LCLUSTER_TYPE_PLAIN ||
+	    ((m->headtype == Z_EROFS_LCLUSTER_TYPE_HEAD1) &&
+	     !(vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_1)) ||
+	    ((m->headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2) &&
+	     !(vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_2))) {
+		map->m_plen = 1ULL << lclusterbits;
+		return 0;
+	}
+	lcn = m->lcn + 1;
+	if (m->compressedblks)
+		goto out;
+
+	err = z_erofs_load_lcluster_from_disk(m, lcn, false);
+	if (err)
+		return err;
+
+	/*
+	 * If the 1st NONHEAD lcluster has already been handled initially w/o
+	 * valid compressedblks, which means at least it mustn't be CBLKCNT, or
+	 * an internal implemenatation error is detected.
+	 *
+	 * The following code can also handle it properly anyway, but let's
+	 * BUG_ON in the debugging mode only for developers to notice that.
+	 */
+	DBG_BUGON(lcn == initial_lcn &&
+		  m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD);
+
+	switch (m->type) {
+	case Z_EROFS_LCLUSTER_TYPE_PLAIN:
+	case Z_EROFS_LCLUSTER_TYPE_HEAD1:
+	case Z_EROFS_LCLUSTER_TYPE_HEAD2:
+		/*
+		 * if the 1st NONHEAD lcluster is actually PLAIN or HEAD type
+		 * rather than CBLKCNT, it's a 1 lcluster-sized pcluster.
+		 */
+		m->compressedblks = 1 << (lclusterbits - sb->s_blocksize_bits);
+		break;
+	case Z_EROFS_LCLUSTER_TYPE_NONHEAD:
+		if (m->delta[0] != 1)
+			goto err_bonus_cblkcnt;
+		if (m->compressedblks)
+			break;
+		fallthrough;
+	default:
+		erofs_err(sb, "cannot found CBLKCNT @ lcn %lu of nid %llu", lcn,
+			  vi->nid);
+		DBG_BUGON(1);
+		return -EFSCORRUPTED;
+	}
+out:
+	map->m_plen = erofs_pos(sb, m->compressedblks);
+	return 0;
+err_bonus_cblkcnt:
+	erofs_err(sb, "bogus CBLKCNT @ lcn %lu of nid %llu", lcn, vi->nid);
+	DBG_BUGON(1);
+	return -EFSCORRUPTED;
+}
+
+static int z_erofs_get_extent_decompressedlen(struct z_erofs_maprecorder *m)
+{
+	struct inode *inode = m->inode;
+	struct erofs_inode *vi = EROFS_I(inode);
+	struct erofs_map_blocks *map = m->map;
+	unsigned int lclusterbits = vi->z_logical_clusterbits;
+	u64 lcn = m->lcn, headlcn = map->m_la >> lclusterbits;
+	int err;
+
+	do {
+		/* handle the last EOF pcluster (no next HEAD lcluster) */
+		if ((lcn << lclusterbits) >= inode->i_size) {
+			map->m_llen = inode->i_size - map->m_la;
+			return 0;
+		}
+
+		err = z_erofs_load_lcluster_from_disk(m, lcn, true);
+		if (err)
+			return err;
+
+		if (m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+			DBG_BUGON(!m->delta[1] &&
+				  m->clusterofs != 1 << lclusterbits);
+		} else if (m->type == Z_EROFS_LCLUSTER_TYPE_PLAIN ||
+			   m->type == Z_EROFS_LCLUSTER_TYPE_HEAD1 ||
+			   m->type == Z_EROFS_LCLUSTER_TYPE_HEAD2) {
+			/* go on until the next HEAD lcluster */
+			if (lcn != headlcn)
+				break;
+			m->delta[1] = 1;
+		} else {
+			erofs_err(inode->i_sb, "unknown type %u @ lcn %llu of nid %llu",
+				  m->type, lcn, vi->nid);
+			DBG_BUGON(1);
+			return -EOPNOTSUPP;
+		}
+		lcn += m->delta[1];
+	} while (m->delta[1]);
+
+	map->m_llen = (lcn << lclusterbits) + m->clusterofs - map->m_la;
+	return 0;
+}
+
+static int z_erofs_do_map_blocks(struct inode *inode,
+				 struct erofs_map_blocks *map, int flags)
+{
+	struct erofs_inode *const vi = EROFS_I(inode);
+	bool ztailpacking = vi->z_advise & Z_EROFS_ADVISE_INLINE_PCLUSTER;
+	bool fragment = vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER;
+	struct z_erofs_maprecorder m = {
+		.inode = inode,
+		.map = map,
+	};
+	int err = 0;
+	unsigned int lclusterbits, endoff, afmt;
+	unsigned long initial_lcn;
+	unsigned long long ofs, end;
+
+	lclusterbits = vi->z_logical_clusterbits;
+	ofs = flags & EROFS_GET_BLOCKS_FINDTAIL ? inode->i_size - 1 : map->m_la;
+	initial_lcn = ofs >> lclusterbits;
+	endoff = ofs & ((1 << lclusterbits) - 1);
+
+	err = z_erofs_load_lcluster_from_disk(&m, initial_lcn, false);
+	if (err)
+		goto unmap_out;
+
+	if (ztailpacking && (flags & EROFS_GET_BLOCKS_FINDTAIL))
+		vi->z_idataoff = m.nextpackoff;
+
+	map->m_flags = EROFS_MAP_MAPPED | EROFS_MAP_ENCODED;
+	end = (m.lcn + 1ULL) << lclusterbits;
+
+	switch (m.type) {
+	case Z_EROFS_LCLUSTER_TYPE_PLAIN:
+	case Z_EROFS_LCLUSTER_TYPE_HEAD1:
+	case Z_EROFS_LCLUSTER_TYPE_HEAD2:
+		if (endoff >= m.clusterofs) {
+			m.headtype = m.type;
+			map->m_la = (m.lcn << lclusterbits) | m.clusterofs;
+			/*
+			 * For ztailpacking files, in order to inline data more
+			 * effectively, special EOF lclusters are now supported
+			 * which can have three parts at most.
+			 */
+			if (ztailpacking && end > inode->i_size)
+				end = inode->i_size;
+			break;
+		}
+		/* m.lcn should be >= 1 if endoff < m.clusterofs */
+		if (!m.lcn) {
+			erofs_err(inode->i_sb,
+				  "invalid logical cluster 0 at nid %llu",
+				  vi->nid);
+			err = -EFSCORRUPTED;
+			goto unmap_out;
+		}
+		end = (m.lcn << lclusterbits) | m.clusterofs;
+		map->m_flags |= EROFS_MAP_FULL_MAPPED;
+		m.delta[0] = 1;
+		fallthrough;
+	case Z_EROFS_LCLUSTER_TYPE_NONHEAD:
+		/* get the corresponding first chunk */
+		err = z_erofs_extent_lookback(&m, m.delta[0]);
+		if (err)
+			goto unmap_out;
+		break;
+	default:
+		erofs_err(inode->i_sb,
+			  "unknown type %u @ offset %llu of nid %llu",
+			  m.type, ofs, vi->nid);
+		err = -EOPNOTSUPP;
+		goto unmap_out;
+	}
+	if (m.partialref)
+		map->m_flags |= EROFS_MAP_PARTIAL_REF;
+	map->m_llen = end - map->m_la;
+
+	if (flags & EROFS_GET_BLOCKS_FINDTAIL) {
+		vi->z_tailextent_headlcn = m.lcn;
+		/* for non-compact indexes, fragmentoff is 64 bits */
+		if (fragment && vi->datalayout == EROFS_INODE_COMPRESSED_FULL)
+			vi->z_fragmentoff |= (u64)m.pblk << 32;
+	}
+	if (ztailpacking && m.lcn == vi->z_tailextent_headlcn) {
+		map->m_flags |= EROFS_MAP_META;
+		map->m_pa = vi->z_idataoff;
+		map->m_plen = vi->z_idata_size;
+	} else if (fragment && m.lcn == vi->z_tailextent_headlcn) {
+		map->m_flags |= EROFS_MAP_FRAGMENT;
+	} else {
+		map->m_pa = erofs_pos(inode->i_sb, m.pblk);
+		err = z_erofs_get_extent_compressedlen(&m, initial_lcn);
+		if (err)
+			goto unmap_out;
+	}
+
+	if (m.headtype == Z_EROFS_LCLUSTER_TYPE_PLAIN) {
+		if (map->m_llen > map->m_plen) {
+			DBG_BUGON(1);
+			err = -EFSCORRUPTED;
+			goto unmap_out;
+		}
+		afmt = vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER ?
+			Z_EROFS_COMPRESSION_INTERLACED :
+			Z_EROFS_COMPRESSION_SHIFTED;
+	} else {
+		afmt = m.headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2 ?
+			vi->z_algorithmtype[1] : vi->z_algorithmtype[0];
+		if (!(EROFS_I_SB(inode)->available_compr_algs & (1 << afmt))) {
+			erofs_err(inode->i_sb, "inconsistent algorithmtype %u for nid %llu",
+				  afmt, vi->nid);
+			err = -EFSCORRUPTED;
+			goto unmap_out;
+		}
+	}
+	map->m_algorithmformat = afmt;
+
+	if ((flags & EROFS_GET_BLOCKS_FIEMAP) ||
+	    ((flags & EROFS_GET_BLOCKS_READMORE) &&
+	     (map->m_algorithmformat == Z_EROFS_COMPRESSION_LZMA ||
+	      map->m_algorithmformat == Z_EROFS_COMPRESSION_DEFLATE) &&
+	      map->m_llen >= i_blocksize(inode))) {
+		err = z_erofs_get_extent_decompressedlen(&m);
+		if (!err)
+			map->m_flags |= EROFS_MAP_FULL_MAPPED;
+	}
+
+unmap_out:
+	erofs_unmap_metabuf(&m.map->buf);
+	return err;
+}
+
+static int z_erofs_fill_inode_lazy(struct inode *inode)
+{
+	struct erofs_inode *const vi = EROFS_I(inode);
+	struct super_block *const sb = inode->i_sb;
+	int err, headnr;
+	erofs_off_t pos;
+	struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+	void *kaddr;
+	struct z_erofs_map_header *h;
+
+	if (test_bit(EROFS_I_Z_INITED_BIT, &vi->flags)) {
+		/*
+		 * paired with smp_mb() at the end of the function to ensure
+		 * fields will only be observed after the bit is set.
+		 */
+		smp_mb();
+		return 0;
+	}
+
+	if (wait_on_bit_lock(&vi->flags, EROFS_I_BL_Z_BIT, TASK_KILLABLE))
+		return -ERESTARTSYS;
+
+	err = 0;
+	if (test_bit(EROFS_I_Z_INITED_BIT, &vi->flags))
+		goto out_unlock;
+
+	pos = ALIGN(erofs_iloc(inode) + vi->inode_isize + vi->xattr_isize, 8);
+	kaddr = erofs_read_metabuf(&buf, sb, erofs_blknr(sb, pos), EROFS_KMAP);
+	if (IS_ERR(kaddr)) {
+		err = PTR_ERR(kaddr);
+		goto out_unlock;
+	}
+
+	h = kaddr + erofs_blkoff(sb, pos);
+	/*
+	 * if the highest bit of the 8-byte map header is set, the whole file
+	 * is stored in the packed inode. The rest bits keeps z_fragmentoff.
+	 */
+	if (h->h_clusterbits >> Z_EROFS_FRAGMENT_INODE_BIT) {
+		vi->z_advise = Z_EROFS_ADVISE_FRAGMENT_PCLUSTER;
+		vi->z_fragmentoff = le64_to_cpu(*(__le64 *)h) ^ (1ULL << 63);
+		vi->z_tailextent_headlcn = 0;
+		goto done;
+	}
+	vi->z_advise = le16_to_cpu(h->h_advise);
+	vi->z_algorithmtype[0] = h->h_algorithmtype & 15;
+	vi->z_algorithmtype[1] = h->h_algorithmtype >> 4;
+
+	headnr = 0;
+	if (vi->z_algorithmtype[0] >= Z_EROFS_COMPRESSION_MAX ||
+	    vi->z_algorithmtype[++headnr] >= Z_EROFS_COMPRESSION_MAX) {
+		erofs_err(sb, "unknown HEAD%u format %u for nid %llu, please upgrade kernel",
+			  headnr + 1, vi->z_algorithmtype[headnr], vi->nid);
+		err = -EOPNOTSUPP;
+		goto out_put_metabuf;
+	}
+
+	vi->z_logical_clusterbits = sb->s_blocksize_bits + (h->h_clusterbits & 7);
+	if (!erofs_sb_has_big_pcluster(EROFS_SB(sb)) &&
+	    vi->z_advise & (Z_EROFS_ADVISE_BIG_PCLUSTER_1 |
+			    Z_EROFS_ADVISE_BIG_PCLUSTER_2)) {
+		erofs_err(sb, "per-inode big pcluster without sb feature for nid %llu",
+			  vi->nid);
+		err = -EFSCORRUPTED;
+		goto out_put_metabuf;
+	}
+	if (vi->datalayout == EROFS_INODE_COMPRESSED_COMPACT &&
+	    !(vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_1) ^
+	    !(vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_2)) {
+		erofs_err(sb, "big pcluster head1/2 of compact indexes should be consistent for nid %llu",
+			  vi->nid);
+		err = -EFSCORRUPTED;
+		goto out_put_metabuf;
+	}
+
+	if (vi->z_advise & Z_EROFS_ADVISE_INLINE_PCLUSTER) {
+		struct erofs_map_blocks map = {
+			.buf = __EROFS_BUF_INITIALIZER
+		};
+
+		vi->z_idata_size = le16_to_cpu(h->h_idata_size);
+		err = z_erofs_do_map_blocks(inode, &map,
+					    EROFS_GET_BLOCKS_FINDTAIL);
+		erofs_put_metabuf(&map.buf);
+
+		if (!map.m_plen ||
+		    erofs_blkoff(sb, map.m_pa) + map.m_plen > sb->s_blocksize) {
+			erofs_err(sb, "invalid tail-packing pclustersize %llu",
+				  map.m_plen);
+			err = -EFSCORRUPTED;
+		}
+		if (err < 0)
+			goto out_put_metabuf;
+	}
+
+	if (vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER &&
+	    !(h->h_clusterbits >> Z_EROFS_FRAGMENT_INODE_BIT)) {
+		struct erofs_map_blocks map = {
+			.buf = __EROFS_BUF_INITIALIZER
+		};
+
+		vi->z_fragmentoff = le32_to_cpu(h->h_fragmentoff);
+		err = z_erofs_do_map_blocks(inode, &map,
+					    EROFS_GET_BLOCKS_FINDTAIL);
+		erofs_put_metabuf(&map.buf);
+		if (err < 0)
+			goto out_put_metabuf;
+	}
+done:
+	/* paired with smp_mb() at the beginning of the function */
+	smp_mb();
+	set_bit(EROFS_I_Z_INITED_BIT, &vi->flags);
+out_put_metabuf:
+	erofs_put_metabuf(&buf);
+out_unlock:
+	clear_and_wake_up_bit(EROFS_I_BL_Z_BIT, &vi->flags);
+	return err;
+}
+
+int z_erofs_map_blocks_iter(struct inode *inode, struct erofs_map_blocks *map,
+			    int flags)
+{
+	struct erofs_inode *const vi = EROFS_I(inode);
+	int err = 0;
+
+	trace_z_erofs_map_blocks_iter_enter(inode, map, flags);
+
+	/* when trying to read beyond EOF, leave it unmapped */
+	if (map->m_la >= inode->i_size) {
+		map->m_llen = map->m_la + 1 - inode->i_size;
+		map->m_la = inode->i_size;
+		map->m_flags = 0;
+		goto out;
+	}
+
+	err = z_erofs_fill_inode_lazy(inode);
+	if (err)
+		goto out;
+
+	if ((vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER) &&
+	    !vi->z_tailextent_headlcn) {
+		map->m_la = 0;
+		map->m_llen = inode->i_size;
+		map->m_flags = EROFS_MAP_MAPPED | EROFS_MAP_FULL_MAPPED |
+				EROFS_MAP_FRAGMENT;
+		goto out;
+	}
+
+	err = z_erofs_do_map_blocks(inode, map, flags);
+out:
+	trace_z_erofs_map_blocks_iter_exit(inode, map, flags, err);
+	return err;
+}
+
+static int z_erofs_iomap_begin_report(struct inode *inode, loff_t offset,
+				loff_t length, unsigned int flags,
+				struct iomap *iomap, struct iomap *srcmap)
+{
+	int ret;
+	struct erofs_map_blocks map = { .m_la = offset };
+
+	ret = z_erofs_map_blocks_iter(inode, &map, EROFS_GET_BLOCKS_FIEMAP);
+	erofs_put_metabuf(&map.buf);
+	if (ret < 0)
+		return ret;
+
+	iomap->bdev = inode->i_sb->s_bdev;
+	iomap->offset = map.m_la;
+	iomap->length = map.m_llen;
+	if (map.m_flags & EROFS_MAP_MAPPED) {
+		iomap->type = IOMAP_MAPPED;
+		iomap->addr = map.m_flags & EROFS_MAP_FRAGMENT ?
+			      IOMAP_NULL_ADDR : map.m_pa;
+	} else {
+		iomap->type = IOMAP_HOLE;
+		iomap->addr = IOMAP_NULL_ADDR;
+		/*
+		 * No strict rule on how to describe extents for post EOF, yet
+		 * we need to do like below. Otherwise, iomap itself will get
+		 * into an endless loop on post EOF.
+		 *
+		 * Calculate the effective offset by subtracting extent start
+		 * (map.m_la) from the requested offset, and add it to length.
+		 * (NB: offset >= map.m_la always)
+		 */
+		if (iomap->offset >= inode->i_size)
+			iomap->length = length + offset - map.m_la;
+	}
+	iomap->flags = 0;
+	return 0;
+}
+
+const struct iomap_ops z_erofs_iomap_report_ops = {
+	.iomap_begin = z_erofs_iomap_begin_report,
+};