1 files changed, 499 insertions, 457 deletions

diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
index 18d7fd1a5064..c8d8e129eb4b 100644
--- a/fs/erofs/zmap.c
+++ b/fs/erofs/zmap.c
@@ -4,296 +4,172 @@
  *             https://www.huawei.com/
  */
 #include "internal.h"
-#include <asm/unaligned.h>
+#include <linux/unaligned.h>
 #include <trace/events/erofs.h>
 
-static int z_erofs_do_map_blocks(struct inode *inode,
-				 struct erofs_map_blocks *map,
-				 int flags);
-
-int z_erofs_fill_inode(struct inode *inode)
-{
-	struct erofs_inode *const vi = EROFS_I(inode);
-	struct erofs_sb_info *sbi = EROFS_SB(inode->i_sb);
-
-	if (!erofs_sb_has_big_pcluster(sbi) &&
-	    !erofs_sb_has_ztailpacking(sbi) &&
-	    vi->datalayout == EROFS_INODE_FLAT_COMPRESSION_LEGACY) {
-		vi->z_advise = 0;
-		vi->z_algorithmtype[0] = 0;
-		vi->z_algorithmtype[1] = 0;
-		vi->z_logical_clusterbits = LOG_BLOCK_SIZE;
-		set_bit(EROFS_I_Z_INITED_BIT, &vi->flags);
-	}
-	inode->i_mapping->a_ops = &z_erofs_aops;
-	return 0;
-}
-
-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;
-
-	DBG_BUGON(!erofs_sb_has_big_pcluster(EROFS_SB(sb)) &&
-		  !erofs_sb_has_ztailpacking(EROFS_SB(sb)) &&
-		  vi->datalayout == EROFS_INODE_FLAT_COMPRESSION_LEGACY);
-
-	pos = ALIGN(iloc(EROFS_SB(sb), vi->nid) + vi->inode_isize +
-		    vi->xattr_isize, 8);
-	kaddr = erofs_read_metabuf(&buf, sb, erofs_blknr(pos),
-				   EROFS_KMAP_ATOMIC);
-	if (IS_ERR(kaddr)) {
-		err = PTR_ERR(kaddr);
-		goto out_unlock;
-	}
-
-	h = kaddr + erofs_blkoff(pos);
-	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 unmap_done;
-	}
-
-	vi->z_logical_clusterbits = LOG_BLOCK_SIZE + (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 unmap_done;
-	}
-	if (vi->datalayout == EROFS_INODE_FLAT_COMPRESSION &&
-	    !(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 unmap_done;
-	}
-unmap_done:
-	erofs_put_metabuf(&buf);
-	if (err)
-		goto out_unlock;
-
-	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(map.m_pa) + map.m_plen > EROFS_BLKSIZ) {
-			erofs_err(sb, "invalid tail-packing pclustersize %llu",
-				  map.m_plen);
-			err = -EFSCORRUPTED;
-		}
-		if (err < 0)
-			goto out_unlock;
-	}
-	/* paired with smp_mb() at the beginning of the function */
-	smp_mb();
-	set_bit(EROFS_I_Z_INITED_BIT, &vi->flags);
-out_unlock:
-	clear_and_wake_up_bit(EROFS_I_BL_Z_BIT, &vi->flags);
-	return err;
-}
-
 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, compressedlcs;
+	erofs_blk_t pblk, compressedblks;
 	erofs_off_t nextpackoff;
+	bool partialref, in_mbox;
 };
 
-static int z_erofs_reload_indexes(struct z_erofs_maprecorder *m,
-				  erofs_blk_t eblk)
-{
-	struct super_block *const sb = m->inode->i_sb;
-
-	m->kaddr = erofs_read_metabuf(&m->map->buf, sb, eblk,
-				      EROFS_KMAP_ATOMIC);
-	if (IS_ERR(m->kaddr))
-		return PTR_ERR(m->kaddr);
-	return 0;
-}
-
-static int legacy_load_cluster_from_disk(struct z_erofs_maprecorder *m,
-					 unsigned long lcn)
+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 ibase = iloc(EROFS_I_SB(inode), vi->nid);
-	const erofs_off_t pos =
-		Z_EROFS_VLE_LEGACY_INDEX_ALIGN(ibase + vi->inode_isize +
-					       vi->xattr_isize) +
-		lcn * sizeof(struct z_erofs_vle_decompressed_index);
-	struct z_erofs_vle_decompressed_index *di;
-	unsigned int advise, type;
-	int err;
-
-	err = z_erofs_reload_indexes(m, erofs_blknr(pos));
-	if (err)
-		return err;
-
-	m->nextpackoff = pos + sizeof(struct z_erofs_vle_decompressed_index);
+	const erofs_off_t pos = Z_EROFS_FULL_INDEX_START(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;
+
+	di = erofs_read_metabuf(&m->map->buf, inode->i_sb, pos, m->in_mbox);
+	if (IS_ERR(di))
+		return PTR_ERR(di);
 	m->lcn = lcn;
-	di = m->kaddr + erofs_blkoff(pos);
+	m->nextpackoff = pos + sizeof(struct z_erofs_lcluster_index);
 
 	advise = le16_to_cpu(di->di_advise);
-	type = (advise >> Z_EROFS_VLE_DI_CLUSTER_TYPE_BIT) &
-		((1 << Z_EROFS_VLE_DI_CLUSTER_TYPE_BITS) - 1);
-	switch (type) {
-	case Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD:
-		m->clusterofs = 1 << vi->z_logical_clusterbits;
+	m->type = advise & Z_EROFS_LI_LCLUSTER_TYPE_MASK;
+	if (m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+		m->clusterofs = 1 << vi->z_lclusterbits;
 		m->delta[0] = le16_to_cpu(di->di_u.delta[0]);
-		if (m->delta[0] & Z_EROFS_VLE_DI_D0_CBLKCNT) {
+		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->compressedlcs = m->delta[0] &
-				~Z_EROFS_VLE_DI_D0_CBLKCNT;
+			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_VLE_CLUSTER_TYPE_PLAIN:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD1:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD2:
+	} else {
+		m->partialref = !!(advise & Z_EROFS_LI_PARTIAL_REF);
 		m->clusterofs = le16_to_cpu(di->di_clusterofs);
 		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;
+	const unsigned int lo = v & ((1 << lobits) - 1);
 
 	*type = (v >> lobits) & 3;
 	return lo;
 }
 
-static int get_compacted_la_distance(unsigned int lclusterbits,
+static int get_compacted_la_distance(unsigned int lobits,
 				     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);
+		lo = decode_compactedbits(lobits, in, encodebits * i, &type);
 
-		if (type != Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD)
+		if (type != Z_EROFS_LCLUSTER_TYPE_NONHEAD)
 			return d1;
 		++d1;
 	} while (++i < vcnt);
 
-	/* vcnt - 1 (Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD) item */
-	if (!(lo & Z_EROFS_VLE_DI_D0_CBLKCNT))
+	/* 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)
+static int z_erofs_load_compact_lcluster(struct z_erofs_maprecorder *m,
+					 unsigned long lcn, 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;
+	struct inode *const inode = m->inode;
+	struct erofs_inode *const vi = EROFS_I(inode);
+	const erofs_off_t ebase = Z_EROFS_MAP_HEADER_END(erofs_iloc(inode) +
+			vi->inode_isize + vi->xattr_isize);
+	const unsigned int lclusterbits = vi->z_lclusterbits;
+	const unsigned int totalidx = erofs_iblks(inode);
+	unsigned int compacted_4b_initial, compacted_2b, amortizedshift;
+	unsigned int vcnt, lo, lobits, encodebits, nblk, bytes;
+	bool big_pcluster = vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_1;
+	erofs_off_t pos;
 	u8 *in, type;
-	bool big_pcluster;
+	int i;
+
+	if (lcn >= totalidx || lclusterbits > 14)
+		return -EINVAL;
+
+	m->lcn = lcn;
+	/* used to align to 32-byte (compacted_2b) alignment */
+	compacted_4b_initial = ((32 - ebase % 32) / 4) & 7;
+	compacted_2b = 0;
+	if ((vi->z_advise & Z_EROFS_ADVISE_COMPACTED_2B) &&
+	    compacted_4b_initial < totalidx)
+		compacted_2b = rounddown(totalidx - compacted_4b_initial, 16);
 
-	if (1 << amortizedshift == 4)
+	pos = ebase;
+	amortizedshift = 2;	/* compact_4b */
+	if (lcn >= compacted_4b_initial) {
+		pos += compacted_4b_initial * 4;
+		lcn -= compacted_4b_initial;
+		if (lcn < compacted_2b) {
+			amortizedshift = 1;
+		} else {
+			pos += compacted_2b * 2;
+			lcn -= compacted_2b;
+		}
+	}
+	pos += lcn * (1 << amortizedshift);
+
+	/* figure out the lcluster count in this pack */
+	if (1 << amortizedshift == 4 && lclusterbits <= 14)
 		vcnt = 2;
-	else if (1 << amortizedshift == 2 && lclusterbits == 12)
+	else if (1 << amortizedshift == 2 && lclusterbits <= 12)
 		vcnt = 16;
 	else
 		return -EOPNOTSUPP;
 
+	in = erofs_read_metabuf(&m->map->buf, inode->i_sb, pos, m->in_mbox);
+	if (IS_ERR(in))
+		return PTR_ERR(in);
+
 	/* 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;
+	lobits = max(lclusterbits, ilog2(Z_EROFS_LI_D0_CBLKCNT) + 1U);
 	encodebits = ((vcnt << amortizedshift) - sizeof(__le32)) * 8 / vcnt;
-	eofs = erofs_blkoff(pos);
-	base = round_down(eofs, vcnt << amortizedshift);
-	in = m->kaddr + base;
-
-	i = (eofs - base) >> amortizedshift;
+	bytes = pos & ((vcnt << amortizedshift) - 1);
+	in -= bytes;
+	i = bytes >> amortizedshift;
 
-	lo = decode_compactedbits(lclusterbits, lomask,
-				  in, encodebits * i, &type);
+	lo = decode_compactedbits(lobits, in, encodebits * i, &type);
 	m->type = type;
-	if (type == Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD) {
+	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,
+			m->delta[1] = get_compacted_la_distance(lobits,
 						encodebits, vcnt, in, i);
-		if (lo & Z_EROFS_VLE_DI_D0_CBLKCNT) {
+		if (lo & Z_EROFS_LI_D0_CBLKCNT) {
 			if (!big_pcluster) {
 				DBG_BUGON(1);
 				return -EFSCORRUPTED;
 			}
-			m->compressedlcs = lo & ~Z_EROFS_VLE_DI_D0_CBLKCNT;
+			m->compressedblks = lo & ~Z_EROFS_LI_D0_CBLKCNT;
 			m->delta[0] = 1;
 			return 0;
 		} else if (i + 1 != (int)vcnt) {
@@ -305,11 +181,11 @@ static int unpack_compacted_index(struct z_erofs_maprecorder *m,
 		 * 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_VLE_CLUSTER_TYPE_NONHEAD)
+		lo = decode_compactedbits(lobits, in,
+					  encodebits * (i - 1), &type);
+		if (type != Z_EROFS_LCLUSTER_TYPE_NONHEAD)
 			lo = 0;
-		else if (lo & Z_EROFS_VLE_DI_D0_CBLKCNT)
+		else if (lo & Z_EROFS_LI_D0_CBLKCNT)
 			lo = 1;
 		m->delta[0] = lo + 1;
 		return 0;
@@ -321,9 +197,9 @@ static int unpack_compacted_index(struct z_erofs_maprecorder *m,
 		nblk = 1;
 		while (i > 0) {
 			--i;
-			lo = decode_compactedbits(lclusterbits, lomask,
-						  in, encodebits * i, &type);
-			if (type == Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD)
+			lo = decode_compactedbits(lobits, in,
+						  encodebits * i, &type);
+			if (type == Z_EROFS_LCLUSTER_TYPE_NONHEAD)
 				i -= lo;
 
 			if (i >= 0)
@@ -333,12 +209,12 @@ static int unpack_compacted_index(struct z_erofs_maprecorder *m,
 		nblk = 0;
 		while (i > 0) {
 			--i;
-			lo = decode_compactedbits(lclusterbits, lomask,
-						  in, encodebits * i, &type);
-			if (type == Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD) {
-				if (lo & Z_EROFS_VLE_DI_D0_CBLKCNT) {
+			lo = decode_compactedbits(lobits, in,
+						  encodebits * i, &type);
+			if (type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+				if (lo & Z_EROFS_LI_D0_CBLKCNT) {
 					--i;
-					nblk += lo & ~Z_EROFS_VLE_DI_D0_CBLKCNT;
+					nblk += lo & ~Z_EROFS_LI_D0_CBLKCNT;
 					continue;
 				}
 				/* bigpcluster shouldn't have plain d0 == 1 */
@@ -357,197 +233,118 @@ static int unpack_compacted_index(struct z_erofs_maprecorder *m,
 	return 0;
 }
 
-static int compacted_load_cluster_from_disk(struct z_erofs_maprecorder *m,
-					    unsigned long lcn, bool lookahead)
+static int z_erofs_load_lcluster_from_disk(struct z_erofs_maprecorder *m,
+					   unsigned int lcn, bool lookahead)
 {
-	struct inode *const inode = m->inode;
-	struct erofs_inode *const vi = EROFS_I(inode);
-	const unsigned int lclusterbits = vi->z_logical_clusterbits;
-	const erofs_off_t ebase = ALIGN(iloc(EROFS_I_SB(inode), vi->nid) +
-					vi->inode_isize + vi->xattr_isize, 8) +
-		sizeof(struct z_erofs_map_header);
-	const unsigned int totalidx = DIV_ROUND_UP(inode->i_size, EROFS_BLKSIZ);
-	unsigned int compacted_4b_initial, compacted_2b;
-	unsigned int amortizedshift;
-	erofs_off_t pos;
+	struct erofs_inode *vi = EROFS_I(m->inode);
 	int err;
 
-	if (lclusterbits != 12)
-		return -EOPNOTSUPP;
-
-	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;
+	if (vi->datalayout == EROFS_INODE_COMPRESSED_COMPACT) {
+		err = z_erofs_load_compact_lcluster(m, lcn, lookahead);
+	} else {
+		DBG_BUGON(vi->datalayout != EROFS_INODE_COMPRESSED_FULL);
+		err = z_erofs_load_full_lcluster(m, lcn);
 	}
-	pos += compacted_2b * 2;
-	lcn -= compacted_2b;
-	amortizedshift = 2;
-out:
-	pos += lcn * (1 << amortizedshift);
-	err = z_erofs_reload_indexes(m, erofs_blknr(pos));
 	if (err)
 		return err;
-	return unpack_compacted_index(m, amortizedshift, pos, lookahead);
-}
-
-static int z_erofs_load_cluster_from_disk(struct z_erofs_maprecorder *m,
-					  unsigned int lcn, bool lookahead)
-{
-	const unsigned int datamode = EROFS_I(m->inode)->datalayout;
-
-	if (datamode == EROFS_INODE_FLAT_COMPRESSION_LEGACY)
-		return legacy_load_cluster_from_disk(m, lcn);
-
-	if (datamode == EROFS_INODE_FLAT_COMPRESSION)
-		return compacted_load_cluster_from_disk(m, lcn, lookahead);
 
-	return -EINVAL;
+	if (m->type >= Z_EROFS_LCLUSTER_TYPE_MAX) {
+		erofs_err(m->inode->i_sb, "unknown type %u @ lcn %u of nid %llu",
+			  m->type, lcn, EROFS_I(m->inode)->nid);
+		DBG_BUGON(1);
+		return -EOPNOTSUPP;
+	} else if (m->type != Z_EROFS_LCLUSTER_TYPE_NONHEAD &&
+		   m->clusterofs >= (1 << vi->z_lclusterbits)) {
+		DBG_BUGON(1);
+		return -EFSCORRUPTED;
+	}
+	return 0;
 }
 
 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);
-	struct erofs_map_blocks *const map = m->map;
-	const unsigned int lclusterbits = vi->z_logical_clusterbits;
-	unsigned long lcn = m->lcn;
-	int err;
+	const unsigned int lclusterbits = vi->z_lclusterbits;
 
-	if (lcn < lookback_distance) {
-		erofs_err(m->inode->i_sb,
-			  "bogus lookback distance @ nid %llu", vi->nid);
-		DBG_BUGON(1);
-		return -EFSCORRUPTED;
-	}
+	while (m->lcn >= lookback_distance) {
+		unsigned long lcn = m->lcn - lookback_distance;
+		int err;
 
-	/* load extent head logical cluster if needed */
-	lcn -= lookback_distance;
-	err = z_erofs_load_cluster_from_disk(m, lcn, false);
-	if (err)
-		return err;
+		if (!lookback_distance)
+			break;
 
-	switch (m->type) {
-	case Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD:
-		if (!m->delta[0]) {
-			erofs_err(m->inode->i_sb,
-				  "invalid lookback distance 0 @ nid %llu",
-				  vi->nid);
-			DBG_BUGON(1);
-			return -EFSCORRUPTED;
+		err = z_erofs_load_lcluster_from_disk(m, lcn, false);
+		if (err)
+			return err;
+		if (m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+			lookback_distance = m->delta[0];
+			continue;
 		}
-		return z_erofs_extent_lookback(m, m->delta[0]);
-	case Z_EROFS_VLE_CLUSTER_TYPE_PLAIN:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD1:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD2:
 		m->headtype = m->type;
-		map->m_la = (lcn << lclusterbits) | m->clusterofs;
-		break;
-	default:
-		erofs_err(m->inode->i_sb,
-			  "unknown type %u @ lcn %lu of nid %llu",
-			  m->type, lcn, vi->nid);
-		DBG_BUGON(1);
-		return -EOPNOTSUPP;
+		m->map->m_la = (lcn << lclusterbits) | m->clusterofs;
+		return 0;
 	}
-	return 0;
+	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 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;
+	struct inode *inode = m->inode;
+	struct super_block *sb = inode->i_sb;
+	struct erofs_inode *vi = EROFS_I(inode);
+	bool bigpcl1 = vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_1;
+	bool bigpcl2 = vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_2;
+	unsigned long lcn = m->lcn + 1;
 	int err;
 
-	DBG_BUGON(m->type != Z_EROFS_VLE_CLUSTER_TYPE_PLAIN &&
-		  m->type != Z_EROFS_VLE_CLUSTER_TYPE_HEAD1 &&
-		  m->type != Z_EROFS_VLE_CLUSTER_TYPE_HEAD2);
+	DBG_BUGON(m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD);
 	DBG_BUGON(m->type != m->headtype);
 
-	if (m->headtype == Z_EROFS_VLE_CLUSTER_TYPE_PLAIN ||
-	    ((m->headtype == Z_EROFS_VLE_CLUSTER_TYPE_HEAD1) &&
-	     !(vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_1)) ||
-	    ((m->headtype == Z_EROFS_VLE_CLUSTER_TYPE_HEAD2) &&
-	     !(vi->z_advise & Z_EROFS_ADVISE_BIG_PCLUSTER_2))) {
-		map->m_plen = 1 << lclusterbits;
-		return 0;
-	}
-	lcn = m->lcn + 1;
-	if (m->compressedlcs)
+	if ((m->headtype == Z_EROFS_LCLUSTER_TYPE_HEAD1 && !bigpcl1) ||
+	    ((m->headtype == Z_EROFS_LCLUSTER_TYPE_PLAIN ||
+	      m->headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2) && !bigpcl2) ||
+	    (lcn << vi->z_lclusterbits) >= inode->i_size)
+		m->compressedblks = 1;
+
+	if (m->compressedblks)
 		goto out;
 
-	err = z_erofs_load_cluster_from_disk(m, lcn, false);
+	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 compressedlcs, which means at least it mustn't be CBLKCNT, or
+	 * 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_VLE_CLUSTER_TYPE_NONHEAD);
+		  m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD);
 
-	switch (m->type) {
-	case Z_EROFS_VLE_CLUSTER_TYPE_PLAIN:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD1:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD2:
-		/*
-		 * if the 1st NONHEAD lcluster is actually PLAIN or HEAD type
-		 * rather than CBLKCNT, it's a 1 lcluster-sized pcluster.
-		 */
-		m->compressedlcs = 1;
-		break;
-	case Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD:
-		if (m->delta[0] != 1)
-			goto err_bonus_cblkcnt;
-		if (m->compressedlcs)
-			break;
-		fallthrough;
-	default:
-		erofs_err(m->inode->i_sb,
-			  "cannot found CBLKCNT @ lcn %lu of nid %llu",
-			  lcn, vi->nid);
+	if (m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD && m->delta[0] != 1) {
+		erofs_err(sb, "bogus CBLKCNT @ lcn %lu of nid %llu", lcn, vi->nid);
 		DBG_BUGON(1);
 		return -EFSCORRUPTED;
 	}
+
+	/*
+	 * if the 1st NONHEAD lcluster is actually PLAIN or HEAD type rather
+	 * than CBLKCNT, it's a 1 block-sized pcluster.
+	 */
+	if (m->type != Z_EROFS_LCLUSTER_TYPE_NONHEAD || !m->compressedblks)
+		m->compressedblks = 1;
 out:
-	map->m_plen = m->compressedlcs << lclusterbits;
+	m->map->m_plen = erofs_pos(sb, m->compressedblks);
 	return 0;
-err_bonus_cblkcnt:
-	erofs_err(m->inode->i_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)
@@ -555,176 +352,416 @@ 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;
+	unsigned int lclusterbits = vi->z_lclusterbits;
 	u64 lcn = m->lcn, headlcn = map->m_la >> lclusterbits;
 	int err;
 
-	do {
+	while (1) {
 		/* 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_cluster_from_disk(m, lcn, true);
+		err = z_erofs_load_lcluster_from_disk(m, lcn, true);
 		if (err)
 			return err;
 
-		if (m->type == Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD) {
-			DBG_BUGON(!m->delta[1] &&
-				  m->clusterofs != 1 << lclusterbits);
-		} else if (m->type == Z_EROFS_VLE_CLUSTER_TYPE_PLAIN ||
-			   m->type == Z_EROFS_VLE_CLUSTER_TYPE_HEAD1 ||
-			   m->type == Z_EROFS_VLE_CLUSTER_TYPE_HEAD2) {
-			/* go on until the next HEAD lcluster */
+		if (m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+			/* work around invalid d1 generated by pre-1.0 mkfs */
+			if (unlikely(!m->delta[1])) {
+				m->delta[1] = 1;
+				DBG_BUGON(1);
+			}
+		} else if (m->type < Z_EROFS_LCLUSTER_TYPE_MAX) {
 			if (lcn != headlcn)
-				break;
+				break;	/* ends at the next HEAD lcluster */
 			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)
+static int z_erofs_map_blocks_fo(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;
+	struct erofs_inode *vi = EROFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	bool fragment = vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER;
+	bool ztailpacking = vi->z_idata_size;
+	unsigned int lclusterbits = vi->z_lclusterbits;
 	struct z_erofs_maprecorder m = {
 		.inode = inode,
 		.map = map,
+		.in_mbox = erofs_inode_in_metabox(inode),
 	};
-	int err = 0;
-	unsigned int lclusterbits, endoff;
+	unsigned int endoff;
 	unsigned long initial_lcn;
 	unsigned long long ofs, end;
+	int err;
 
-	lclusterbits = vi->z_logical_clusterbits;
 	ofs = flags & EROFS_GET_BLOCKS_FINDTAIL ? inode->i_size - 1 : map->m_la;
+	if (fragment && !(flags & EROFS_GET_BLOCKS_FINDTAIL) &&
+	    !vi->z_tailextent_headlcn) {
+		map->m_la = 0;
+		map->m_llen = inode->i_size;
+		map->m_flags = EROFS_MAP_FRAGMENT;
+		return 0;
+	}
 	initial_lcn = ofs >> lclusterbits;
 	endoff = ofs & ((1 << lclusterbits) - 1);
 
-	err = z_erofs_load_cluster_from_disk(&m, initial_lcn, false);
+	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;
-
+	if ((flags & EROFS_GET_BLOCKS_FINDTAIL) && ztailpacking)
+		vi->z_fragmentoff = m.nextpackoff;
 	map->m_flags = EROFS_MAP_MAPPED | EROFS_MAP_ENCODED;
 	end = (m.lcn + 1ULL) << lclusterbits;
 
-	switch (m.type) {
-	case Z_EROFS_VLE_CLUSTER_TYPE_PLAIN:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD1:
-	case Z_EROFS_VLE_CLUSTER_TYPE_HEAD2:
-		if (endoff >= m.clusterofs) {
-			m.headtype = m.type;
-			map->m_la = (m.lcn << lclusterbits) | m.clusterofs;
-			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;
+	if (m.type != Z_EROFS_LCLUSTER_TYPE_NONHEAD && 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;
+	} else {
+		if (m.type != Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+			end = (m.lcn << lclusterbits) | m.clusterofs;
+			map->m_flags |= EROFS_MAP_FULL_MAPPED;
+			m.delta[0] = 1;
 		}
-		end = (m.lcn << lclusterbits) | m.clusterofs;
-		map->m_flags |= EROFS_MAP_FULL_MAPPED;
-		m.delta[0] = 1;
-		fallthrough;
-	case Z_EROFS_VLE_CLUSTER_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)
+	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_pa = vi->z_fragmentoff;
 		map->m_plen = vi->z_idata_size;
+		if (erofs_blkoff(sb, map->m_pa) + map->m_plen > sb->s_blocksize) {
+			erofs_err(sb, "ztailpacking inline data across blocks @ nid %llu",
+				  vi->nid);
+			err = -EFSCORRUPTED;
+			goto unmap_out;
+		}
+	} else if (fragment && m.lcn == vi->z_tailextent_headlcn) {
+		map->m_flags = EROFS_MAP_FRAGMENT;
 	} else {
-		map->m_pa = blknr_to_addr(m.pblk);
+		map->m_pa = erofs_pos(sb, m.pblk);
 		err = z_erofs_get_extent_compressedlen(&m, initial_lcn);
 		if (err)
-			goto out;
+			goto unmap_out;
 	}
 
-	if (m.headtype == Z_EROFS_VLE_CLUSTER_TYPE_PLAIN)
-		map->m_algorithmformat = Z_EROFS_COMPRESSION_SHIFTED;
-	else if (m.headtype == Z_EROFS_VLE_CLUSTER_TYPE_HEAD2)
+	if (m.headtype == Z_EROFS_LCLUSTER_TYPE_PLAIN) {
+		if (map->m_llen > map->m_plen) {
+			DBG_BUGON(1);
+			err = -EFSCORRUPTED;
+			goto unmap_out;
+		}
+		if (vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER)
+			map->m_algorithmformat = Z_EROFS_COMPRESSION_INTERLACED;
+		else
+			map->m_algorithmformat = Z_EROFS_COMPRESSION_SHIFTED;
+	} else if (m.headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2) {
 		map->m_algorithmformat = vi->z_algorithmtype[1];
-	else
+	} else {
 		map->m_algorithmformat = vi->z_algorithmtype[0];
+	}
 
 	if ((flags & EROFS_GET_BLOCKS_FIEMAP) ||
 	    ((flags & EROFS_GET_BLOCKS_READMORE) &&
-	     map->m_algorithmformat == Z_EROFS_COMPRESSION_LZMA &&
-	     map->m_llen >= EROFS_BLKSIZ)) {
+	     (map->m_algorithmformat == Z_EROFS_COMPRESSION_LZMA ||
+	      map->m_algorithmformat == Z_EROFS_COMPRESSION_DEFLATE ||
+	      map->m_algorithmformat == Z_EROFS_COMPRESSION_ZSTD) &&
+	      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;
+}
 
-out:
-	erofs_dbg("%s, m_la %llu m_pa %llu m_llen %llu m_plen %llu m_flags 0%o",
-		  __func__, map->m_la, map->m_pa,
-		  map->m_llen, map->m_plen, map->m_flags);
+static int z_erofs_map_blocks_ext(struct inode *inode,
+				  struct erofs_map_blocks *map, int flags)
+{
+	struct erofs_inode *vi = EROFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	bool interlaced = vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER;
+	unsigned int recsz = z_erofs_extent_recsize(vi->z_advise);
+	erofs_off_t pos = round_up(Z_EROFS_MAP_HEADER_END(erofs_iloc(inode) +
+				   vi->inode_isize + vi->xattr_isize), recsz);
+	bool in_mbox = erofs_inode_in_metabox(inode);
+	erofs_off_t lend = inode->i_size;
+	erofs_off_t l, r, mid, pa, la, lstart;
+	struct z_erofs_extent *ext;
+	unsigned int fmt;
+	bool last;
+
+	map->m_flags = 0;
+	if (recsz <= offsetof(struct z_erofs_extent, pstart_hi)) {
+		if (recsz <= offsetof(struct z_erofs_extent, pstart_lo)) {
+			ext = erofs_read_metabuf(&map->buf, sb, pos, in_mbox);
+			if (IS_ERR(ext))
+				return PTR_ERR(ext);
+			pa = le64_to_cpu(*(__le64 *)ext);
+			pos += sizeof(__le64);
+			lstart = 0;
+		} else {
+			lstart = round_down(map->m_la, 1 << vi->z_lclusterbits);
+			pos += (lstart >> vi->z_lclusterbits) * recsz;
+			pa = EROFS_NULL_ADDR;
+		}
+
+		for (; lstart <= map->m_la; lstart += 1 << vi->z_lclusterbits) {
+			ext = erofs_read_metabuf(&map->buf, sb, pos, in_mbox);
+			if (IS_ERR(ext))
+				return PTR_ERR(ext);
+			map->m_plen = le32_to_cpu(ext->plen);
+			if (pa != EROFS_NULL_ADDR) {
+				map->m_pa = pa;
+				pa += map->m_plen & Z_EROFS_EXTENT_PLEN_MASK;
+			} else {
+				map->m_pa = le32_to_cpu(ext->pstart_lo);
+			}
+			pos += recsz;
+		}
+		last = (lstart >= round_up(lend, 1 << vi->z_lclusterbits));
+		lend = min(lstart, lend);
+		lstart -= 1 << vi->z_lclusterbits;
+	} else {
+		lstart = lend;
+		for (l = 0, r = vi->z_extents; l < r; ) {
+			mid = l + (r - l) / 2;
+			ext = erofs_read_metabuf(&map->buf, sb,
+						 pos + mid * recsz, in_mbox);
+			if (IS_ERR(ext))
+				return PTR_ERR(ext);
+
+			la = le32_to_cpu(ext->lstart_lo);
+			pa = le32_to_cpu(ext->pstart_lo) |
+				(u64)le32_to_cpu(ext->pstart_hi) << 32;
+			if (recsz > offsetof(struct z_erofs_extent, lstart_hi))
+				la |= (u64)le32_to_cpu(ext->lstart_hi) << 32;
+
+			if (la > map->m_la) {
+				r = mid;
+				if (la > lend) {
+					DBG_BUGON(1);
+					return -EFSCORRUPTED;
+				}
+				lend = la;
+			} else {
+				l = mid + 1;
+				if (map->m_la == la)
+					r = min(l + 1, r);
+				lstart = la;
+				map->m_plen = le32_to_cpu(ext->plen);
+				map->m_pa = pa;
+			}
+		}
+		last = (l >= vi->z_extents);
+	}
+
+	if (lstart < lend) {
+		map->m_la = lstart;
+		if (last && (vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER)) {
+			map->m_flags = EROFS_MAP_FRAGMENT;
+			vi->z_fragmentoff = map->m_plen;
+			if (recsz > offsetof(struct z_erofs_extent, pstart_lo))
+				vi->z_fragmentoff |= map->m_pa << 32;
+		} else if (map->m_plen & Z_EROFS_EXTENT_PLEN_MASK) {
+			map->m_flags |= EROFS_MAP_MAPPED |
+				EROFS_MAP_FULL_MAPPED | EROFS_MAP_ENCODED;
+			fmt = map->m_plen >> Z_EROFS_EXTENT_PLEN_FMT_BIT;
+			if (fmt)
+				map->m_algorithmformat = fmt - 1;
+			else if (interlaced && !erofs_blkoff(sb, map->m_pa))
+				map->m_algorithmformat =
+					Z_EROFS_COMPRESSION_INTERLACED;
+			else
+				map->m_algorithmformat =
+					Z_EROFS_COMPRESSION_SHIFTED;
+			if (map->m_plen & Z_EROFS_EXTENT_PLEN_PARTIAL)
+				map->m_flags |= EROFS_MAP_PARTIAL_REF;
+			map->m_plen &= Z_EROFS_EXTENT_PLEN_MASK;
+		}
+	}
+	map->m_llen = lend - map->m_la;
+	return 0;
+}
+
+static int z_erofs_fill_inode(struct inode *inode, struct erofs_map_blocks *map)
+{
+	struct erofs_inode *const vi = EROFS_I(inode);
+	struct super_block *const sb = inode->i_sb;
+	struct z_erofs_map_header *h;
+	erofs_off_t pos;
+	int err = 0;
 
+	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;
+
+	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);
+	h = erofs_read_metabuf(&map->buf, sb, pos, erofs_inode_in_metabox(inode));
+	if (IS_ERR(h)) {
+		err = PTR_ERR(h);
+		goto out_unlock;
+	}
+
+	/*
+	 * 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_lclusterbits = sb->s_blocksize_bits + (h->h_clusterbits & 15);
+	if (vi->datalayout == EROFS_INODE_COMPRESSED_FULL &&
+	    (vi->z_advise & Z_EROFS_ADVISE_EXTENTS)) {
+		vi->z_extents = le32_to_cpu(h->h_extents_lo) |
+			((u64)le16_to_cpu(h->h_extents_hi) << 32);
+		goto done;
+	}
+
+	vi->z_algorithmtype[0] = h->h_algorithmtype & 15;
+	vi->z_algorithmtype[1] = h->h_algorithmtype >> 4;
+	if (vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER)
+		vi->z_fragmentoff = le32_to_cpu(h->h_fragmentoff);
+	else if (vi->z_advise & Z_EROFS_ADVISE_INLINE_PCLUSTER)
+		vi->z_idata_size = le16_to_cpu(h->h_idata_size);
+
+	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_unlock;
+	}
+	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_unlock;
+	}
+
+	if (vi->z_idata_size ||
+	    (vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER)) {
+		struct erofs_map_blocks tm = {
+			.buf = __EROFS_BUF_INITIALIZER
+		};
+
+		err = z_erofs_map_blocks_fo(inode, &tm,
+					    EROFS_GET_BLOCKS_FINDTAIL);
+		erofs_put_metabuf(&tm.buf);
+		if (err < 0)
+			goto out_unlock;
+	}
+done:
+	/* paired with smp_mb() at the beginning of the function */
+	smp_mb();
+	set_bit(EROFS_I_Z_INITED_BIT, &vi->flags);
+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,
+static int z_erofs_map_sanity_check(struct inode *inode,
+				    struct erofs_map_blocks *map)
+{
+	struct erofs_sb_info *sbi = EROFS_I_SB(inode);
+	u64 pend;
+
+	if (!(map->m_flags & EROFS_MAP_ENCODED))
+		return 0;
+	if (unlikely(map->m_algorithmformat >= Z_EROFS_COMPRESSION_RUNTIME_MAX)) {
+		erofs_err(inode->i_sb, "unknown algorithm %d @ pos %llu for nid %llu, please upgrade kernel",
+			  map->m_algorithmformat, map->m_la, EROFS_I(inode)->nid);
+		return -EOPNOTSUPP;
+	}
+	if (unlikely(map->m_algorithmformat < Z_EROFS_COMPRESSION_MAX &&
+		     !(sbi->available_compr_algs & (1 << map->m_algorithmformat)))) {
+		erofs_err(inode->i_sb, "inconsistent algorithmtype %u for nid %llu",
+			  map->m_algorithmformat, EROFS_I(inode)->nid);
+		return -EFSCORRUPTED;
+	}
+	if (unlikely(map->m_plen > Z_EROFS_PCLUSTER_MAX_SIZE ||
+		     map->m_llen > Z_EROFS_PCLUSTER_MAX_DSIZE))
+		return -EOPNOTSUPP;
+	/* Filesystems beyond 48-bit physical block addresses are invalid */
+	if (unlikely(check_add_overflow(map->m_pa, map->m_plen, &pend) ||
+		     (pend >> sbi->blkszbits) >= BIT_ULL(48)))
+		return -EFSCORRUPTED;
+	return 0;
+}
+
+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) {
+	trace_erofs_map_blocks_enter(inode, map, flags);
+	if (map->m_la >= inode->i_size) {	/* post-EOF unmapped extent */
 		map->m_llen = map->m_la + 1 - inode->i_size;
 		map->m_la = inode->i_size;
 		map->m_flags = 0;
-		goto out;
+	} else {
+		err = z_erofs_fill_inode(inode, map);
+		if (!err) {
+			if (vi->datalayout == EROFS_INODE_COMPRESSED_FULL &&
+			    (vi->z_advise & Z_EROFS_ADVISE_EXTENTS))
+				err = z_erofs_map_blocks_ext(inode, map, flags);
+			else
+				err = z_erofs_map_blocks_fo(inode, map, flags);
+		}
+		if (!err)
+			err = z_erofs_map_sanity_check(inode, map);
+		if (err)
+			map->m_llen = 0;
 	}
-
-	err = z_erofs_fill_inode_lazy(inode);
-	if (err)
-		goto out;
-
-	err = z_erofs_do_map_blocks(inode, map, flags);
-out:
-	trace_z_erofs_map_blocks_iter_exit(inode, map, flags, err);
-
-	/* aggressively BUG_ON iff CONFIG_EROFS_FS_DEBUG is on */
-	DBG_BUGON(err < 0 && err != -ENOMEM);
+	trace_erofs_map_blocks_exit(inode, map, flags, err);
 	return err;
 }
 
@@ -745,17 +782,22 @@ static int z_erofs_iomap_begin_report(struct inode *inode, loff_t offset,
 	iomap->length = map.m_llen;
 	if (map.m_flags & EROFS_MAP_MAPPED) {
 		iomap->type = IOMAP_MAPPED;
-		iomap->addr = map.m_pa;
+		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 how to describe extents for post EOF, yet
-		 * we need do like below. Otherwise, iomap itself will get
+		 * 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 + map.m_la - offset;
+			iomap->length = length + offset - map.m_la;
 	}
 	iomap->flags = 0;
 	return 0;