Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Fix error output of boot instance creation failure

   Currently if a boot instance creation fails, instead of printing out
   the name of the instance that failed, it prints "(null)". That is
   because it prints "cur_str" that had already been processed by
   strsep(). Print the saved name instead.

   While at it, print the error code of the failure.

 - Fix use-after-free for same named historgrams

   Histograms can be named so that they can be used in multiple events.
   But if the named histogram has a variable attached, the second event
   that uses the named histogram which duplicates it and needs to free
   the original after duplication leaves the old variable in place and
   still visible. If another histogram uses than variable, it will use
   the stale one which will try to reference the freed duplicate
   histogram and crash the kernel.

   Free the duplicate variables along with the duplicated histogram
   data.

 - Check return value of kthread_run() in event self test

   The events self tests uses a kthread for testing but does not check
   if it succeeded in creating a kthread. If the kthread creation were
   to fail, the code will still try to call kthread_stop() on the error
   returned.

 - Fix race between reading trace_pipe and updating subbuffer size

   If a user is reading the trace_pipe file at the same time they update
   the ring buffer sub-buffer size, can cause the trace_pipe read to
   read stale data. Add trace_access_lock() around updating the ring
   buffer sub-buffer size.

 - Fix eventfs_inode on failure path in creation of the events directory

   In the creation of the "events" directory, if after allocating the
   eventfs_inode a failure is detected, it calls cleanup_ei() which
   calls free_ei(). The free_ei() will test if eventfs_inode being freed
   has no children. It is a bug if it does. But on the failure case of
   the creation of the "events" directory, the children lists have not
   yet been initialized and the free will trigger a warning because
   list_empty() on an uninitialized list returns false.

   Move the initialization into init_ei() where it makes more sense and
   makes sure that a created eventfs_inode has its lists initialized
   upon creation.

 - Check return value of kthread_run() in ftrace direct sample code

   The sample code that shows how to use the ftrace direct calls does
   not test the return of kthread_run() to see if it succeeds. Return a
   failure if the kthread_run() doesn't succeed.

 - Clear user events state on fork in case of alloc failure

   On fork, the child gets a pointer to the parent's user events state.
   It makes a copy of it then updates the child's pointer to it. But if
   the allocation fails, the duplication function leaves the child with
   a pointer to its parent's descriptor. When the child cleans up its
   data, it will free the parent's descriptor while the parent is still
   using it.

   In the duplication function, set the child's user_event_mm to NULL
   before testing if the allocation succeeded, and when it exits it will
   not free the parent's descriptor.

 - Fix retry exhaustion in simple ring buffer reader swap

   simple_ring_buffer_swap_reader_page() starts with retry set to 8 and
   post-decrements it only after a failed link replacement. On the final
   attempt, a successful replacement leaves retry at zero, while a
   failed replacement leaves it at -1.

   But the check for success expects the retry value to be non-zero and
   exits with an error on zero. This is the opposite result. Fix it.

 - Fail nicely when the remote swap_reader_page() returns an error

   Currently, if the swap_reader_page() of a remote buffer fails, it
   triggers a WARN_ON_ONCE() and continues normally. Instead, have it
   exit with an error and a pr_warn() print instead of a full WARNING.

* tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Stop remote reader update when page swap fails
  tracing: Fix retry exhaustion in simple ring buffer reader swap
  tracing/user_events: Clear copied tracing state before fork duplication
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify
  eventfs: Initialize ei->children and ei->list in init_ei()
  tracing: Fix use-after-free in trace_pipe read on sub-buffer order change
  tracing: Fix crash passing ERR_PTR to kthread_stop()
  tracing: Fix use-after-free with same-name named triggers
  tracing: Fix logged instance name on creation failure

1 files changed, 838 insertions, 0 deletions

diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
new file mode 100644
index 000000000..f8981fc74
--- /dev/null
+++ b/fs/erofs/zmap.c
@@ -0,0 +1,838 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2018-2019 HUAWEI, Inc.
+ *             https://www.huawei.com/
+ */
+#include "internal.h"
+#include <linux/unaligned.h>
+#include <trace/events/erofs.h>
+
+struct z_erofs_maprecorder {
+	struct inode *inode;
+	struct erofs_map_blocks *map;
+	u64 lcn;
+	/* compression extent information gathered */
+	u8  type, headtype;
+	u16 clusterofs;
+	u16 delta[2];
+	erofs_blk_t pblk;
+	erofs_off_t nextpackoff;
+	int compressedblks;
+	bool partialref, in_mbox;
+};
+
+static int z_erofs_load_full_lcluster(struct z_erofs_maprecorder *m, u64 lcn)
+{
+	struct inode *const inode = m->inode;
+	struct erofs_inode *const vi = EROFS_I(inode);
+	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;
+	m->nextpackoff = pos + sizeof(struct z_erofs_lcluster_index);
+
+	advise = le16_to_cpu(di->di_advise);
+	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_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]);
+	} else {
+		m->partialref = !!(advise & Z_EROFS_LI_PARTIAL_REF);
+		m->clusterofs = le16_to_cpu(di->di_clusterofs);
+		if (advise & Z_EROFS_LI_HOLE) {
+			m->compressedblks = 0;
+			m->pblk = EROFS_NULL_ADDR;
+		} else {
+			m->pblk = le32_to_cpu(di->di_u.blkaddr);
+		}
+	}
+	return 0;
+}
+
+static unsigned int decode_compactedbits(unsigned int lobits,
+					 u8 *in, unsigned int pos, u8 *type)
+{
+	const unsigned int v = get_unaligned_le32(in + pos / 8) >> (pos & 7);
+	const unsigned int lo = v & ((1 << lobits) - 1);
+
+	*type = (v >> lobits) & 3;
+	return lo;
+}
+
+static int get_compacted_la_distance(unsigned int lobits,
+				     unsigned int encodebits,
+				     unsigned int vcnt, u8 *in, int i)
+{
+	unsigned int lo, d1 = 0;
+	u8 type;
+
+	DBG_BUGON(i >= vcnt);
+
+	do {
+		lo = decode_compactedbits(lobits, 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 z_erofs_load_compact_lcluster(struct z_erofs_maprecorder *m,
+					 u64 lcn, bool lookahead)
+{
+	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;
+	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);
+
+	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)
+		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);
+	lobits = max(lclusterbits, ilog2(Z_EROFS_LI_D0_CBLKCNT) + 1U);
+	encodebits = ((vcnt << amortizedshift) - sizeof(__le32)) * 8 / vcnt;
+	bytes = pos & ((vcnt << amortizedshift) - 1);
+	in -= bytes;
+	i = bytes >> amortizedshift;
+
+	lo = decode_compactedbits(lobits, 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(lobits,
+						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(lobits, 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(lobits, 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(lobits, 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_lcluster_from_disk(struct z_erofs_maprecorder *m,
+					   u64 lcn, bool lookahead)
+{
+	struct erofs_inode *vi = EROFS_I(m->inode);
+	int err;
+
+	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);
+	}
+	if (err)
+		return err;
+
+	if (m->type >= Z_EROFS_LCLUSTER_TYPE_MAX) {
+		erofs_err(m->inode->i_sb, "unknown type %u @ lcn %llu 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);
+	const unsigned int lclusterbits = vi->z_lclusterbits;
+
+	while (m->lcn >= lookback_distance) {
+		u64 lcn = m->lcn - lookback_distance;
+		int err;
+
+		if (!lookback_distance)
+			break;
+
+		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;
+		}
+		m->headtype = m->type;
+		m->map->m_la = (lcn << lclusterbits) | m->clusterofs;
+		return 0;
+	}
+	erofs_err(sb, "bogus lookback distance %u @ lcn %llu 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 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;
+	u64 lcn = m->lcn + 1;
+	int err;
+
+	DBG_BUGON(m->type == Z_EROFS_LCLUSTER_TYPE_NONHEAD);
+	DBG_BUGON(m->type != m->headtype);
+
+	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)
+		if (m->compressedblks < 0)
+			m->compressedblks = 1;
+
+	if (m->compressedblks >= 0)
+		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);
+
+	if (m->type != Z_EROFS_LCLUSTER_TYPE_NONHEAD) {
+		/*
+		 * if the 1st NONHEAD lcluster is actually PLAIN or HEAD type
+		 * rather than CBLKCNT, it's a 1 block-sized pcluster.
+		 */
+		if (m->compressedblks < 0)
+			m->compressedblks = 1;
+	} else if (m->delta[0] != 1 || m->compressedblks < 0) {
+		erofs_err(sb, "bogus CBLKCNT @ lcn %llu of nid %llu", lcn, vi->nid);
+		DBG_BUGON(1);
+		return -EFSCORRUPTED;
+	}
+
+out:
+	if (!m->compressedblks)
+		m->map->m_flags &= ~EROFS_MAP_MAPPED;
+	m->map->m_plen = erofs_pos(sb, m->compressedblks);
+	return 0;
+}
+
+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_lclusterbits;
+	u64 lcn = m->lcn, headlcn = map->m_la >> lclusterbits;
+	int err;
+
+	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_lcluster_from_disk(m, lcn, true);
+		if (err)
+			return err;
+
+		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;	/* ends at the next HEAD lcluster */
+			m->delta[1] = 1;
+		}
+		lcn += m->delta[1];
+	}
+	map->m_llen = (lcn << lclusterbits) + m->clusterofs - map->m_la;
+	return 0;
+}
+
+static int z_erofs_map_blocks_fo(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 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),
+		.compressedblks = -1,
+	};
+	unsigned int endoff;
+	unsigned long initial_lcn;
+	unsigned long long ofs, end;
+	int err;
+
+	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_lcluster_from_disk(&m, initial_lcn, false);
+	if (err)
+		goto unmap_out;
+
+	if ((flags & EROFS_GET_BLOCKS_FINDTAIL) && ztailpacking)
+		vi->z_fragmentoff = m.nextpackoff;
+	map->m_flags = EROFS_MAP_MAPPED | EROFS_MAP_PARTIAL_MAPPED;
+	end = (m.lcn + 1ULL) << lclusterbits;
+
+	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_PARTIAL_MAPPED;
+			m.delta[0] = 1;
+		}
+		/* get the corresponding first chunk */
+		err = z_erofs_extent_lookback(&m, m.delta[0]);
+		if (err)
+			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_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 = erofs_pos(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 ((vi->z_advise & Z_EROFS_ADVISE_INTERLACED_PCLUSTER) &&
+		    !(map->m_flags & EROFS_MAP_META))
+			map->m_algorithmformat = Z_EROFS_COMPRESSION_INTERLACED;
+		else
+			map->m_algorithmformat = Z_EROFS_COMPRESSION_SHIFTED;
+	} else {
+		map->m_algorithmformat =
+			vi->z_algofmt[m.headtype == Z_EROFS_LCLUSTER_TYPE_HEAD2];
+	}
+
+	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_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_PARTIAL_MAPPED;
+	}
+
+unmap_out:
+	erofs_unmap_metabuf(&m.map->buf);
+	return err;
+}
+
+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);
+	unsigned int bmask = sb->s_blocksize - 1;
+	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;
+			fmt = map->m_plen >> Z_EROFS_EXTENT_PLEN_FMT_BIT;
+			if (map->m_plen & Z_EROFS_EXTENT_PLEN_PARTIAL)
+				map->m_flags |= EROFS_MAP_PARTIAL_REF;
+			map->m_plen &= Z_EROFS_EXTENT_PLEN_MASK;
+			if (fmt) {
+				map->m_algorithmformat = --fmt;
+				if (fmt >= Z_EROFS_COMPRESSION_MAX) {
+					erofs_err(sb, "unknown algorithm %d @ pos %llu for nid %llu, please upgrade kernel",
+						  fmt, map->m_la, vi->nid);
+					return -EOPNOTSUPP;
+				}
+			} else if (interlaced && !((map->m_pa | map->m_plen) & bmask))
+				map->m_algorithmformat =
+					Z_EROFS_COMPRESSION_INTERLACED;
+			else
+				map->m_algorithmformat =
+					Z_EROFS_COMPRESSION_SHIFTED;
+		}
+	}
+	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, nr;
+
+	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;
+	}
+
+	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);
+	for (nr = 0; nr < 2; ++nr) {
+		vi->z_algofmt[nr] = (h->h_algorithmtype >> (4 * nr)) & 15;
+		if (vi->z_algofmt[nr] >= Z_EROFS_COMPRESSION_MAX) {
+			erofs_err(sb, "unknown HEAD%u format %u for nid %llu, please upgrade kernel",
+				  nr + 1, vi->z_algofmt[nr], vi->nid);
+			err = -EOPNOTSUPP;
+			goto out_unlock;
+		}
+	}
+
+	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;
+}
+
+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_MAPPED))
+		return 0;
+
+	DBG_BUGON(map->m_algorithmformat >= Z_EROFS_COMPRESSION_RUNTIME_MAX);
+	if (map->m_algorithmformat < Z_EROFS_COMPRESSION_MAX) {
+		if (!(sbi->available_compr_algs & BIT(map->m_algorithmformat))) {
+			erofs_err(inode->i_sb, "inconsistent algorithmtype %u for nid %llu",
+				  map->m_algorithmformat, EROFS_I(inode)->nid);
+			return -EFSCORRUPTED;
+		}
+		if (EROFS_MAP_FULL(map->m_flags) && map->m_llen < map->m_plen &&
+		    map->m_la + map->m_llen < inode->i_size) {
+			erofs_err(inode->i_sb, "too much compressed data @ la %llu of nid %llu",
+				  map->m_la, EROFS_I(inode)->nid);
+			return -EFSCORRUPTED;
+		}
+	} else if (map->m_llen > map->m_plen) {
+		erofs_err(inode->i_sb, "not enough plain data on disk @ la %llu of nid %llu",
+			  map->m_la, 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_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;
+	} 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;
+	}
+	trace_erofs_map_blocks_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_FRAGMENT) {
+		iomap->type = IOMAP_MAPPED;
+		iomap->addr = IOMAP_NULL_ADDR;
+	} else if (map.m_flags & EROFS_MAP_MAPPED) {
+		iomap->type = IOMAP_MAPPED;
+		iomap->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;
+}
+
+static DEFINE_IOMAP_ITER_NEXT(z_erofs_iomap_next_report,
+			      z_erofs_iomap_begin_report);
+
+const struct iomap_ops z_erofs_iomap_report_ops = {
+	.iomap_next = z_erofs_iomap_next_report,
+};