x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

1 files changed, 720 insertions, 0 deletions

diff --git a/fs/hfsplus/bnode.c b/fs/hfsplus/bnode.c
new file mode 100644
index 000000000..3a1a1fa6f
--- /dev/null
+++ b/fs/hfsplus/bnode.c
@@ -0,0 +1,720 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  linux/fs/hfsplus/bnode.c
+ *
+ * Copyright (C) 2001
+ * Brad Boyer (flar@allandria.com)
+ * (C) 2003 Ardis Technologies <roman@ardistech.com>
+ *
+ * Handle basic btree node operations
+ */
+
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/pagemap.h>
+#include <linux/fs.h>
+#include <linux/swap.h>
+
+#include "hfsplus_fs.h"
+#include "hfsplus_raw.h"
+
+
+/* Copy a specified range of bytes from the raw data of a node */
+void hfs_bnode_read(struct hfs_bnode *node, void *buf, u32 off, u32 len)
+{
+	struct page **pagep;
+	u32 l;
+
+	memset(buf, 0, len);
+
+	if (!is_bnode_offset_valid(node, off))
+		return;
+
+	if (len == 0) {
+		pr_err("requested zero length: "
+		       "NODE: id %u, type %#x, height %u, "
+		       "node_size %u, offset %u, len %u\n",
+		       node->this, node->type, node->height,
+		       node->tree->node_size, off, len);
+		return;
+	}
+
+	len = check_and_correct_requested_length(node, off, len);
+
+	off += node->page_offset;
+	pagep = node->page + (off >> PAGE_SHIFT);
+	off &= ~PAGE_MASK;
+
+	l = min_t(u32, len, PAGE_SIZE - off);
+	memcpy_from_page(buf, *pagep, off, l);
+
+	while ((len -= l) != 0) {
+		buf += l;
+		l = min_t(u32, len, PAGE_SIZE);
+		memcpy_from_page(buf, *++pagep, 0, l);
+	}
+}
+
+u16 hfs_bnode_read_u16(struct hfs_bnode *node, u32 off)
+{
+	__be16 data;
+	/* TODO: optimize later... */
+	hfs_bnode_read(node, &data, off, 2);
+	return be16_to_cpu(data);
+}
+
+u8 hfs_bnode_read_u8(struct hfs_bnode *node, u32 off)
+{
+	u8 data;
+	/* TODO: optimize later... */
+	hfs_bnode_read(node, &data, off, 1);
+	return data;
+}
+
+void hfs_bnode_read_key(struct hfs_bnode *node, void *key, u32 off)
+{
+	struct hfs_btree *tree;
+	u32 key_len;
+
+	tree = node->tree;
+	if (node->type == HFS_NODE_LEAF ||
+	    tree->attributes & HFS_TREE_VARIDXKEYS ||
+	    node->tree->cnid == HFSPLUS_ATTR_CNID)
+		key_len = hfs_bnode_read_u16(node, off) + 2;
+	else
+		key_len = tree->max_key_len + 2;
+
+	if (key_len > sizeof(hfsplus_btree_key) || key_len < 1) {
+		memset(key, 0, sizeof(hfsplus_btree_key));
+		pr_err("hfsplus: Invalid key length: %u\n", key_len);
+		return;
+	}
+
+	hfs_bnode_read(node, key, off, key_len);
+}
+
+void hfs_bnode_write(struct hfs_bnode *node, void *buf, u32 off, u32 len)
+{
+	struct page **pagep;
+	u32 l;
+
+	if (!is_bnode_offset_valid(node, off))
+		return;
+
+	if (len == 0) {
+		pr_err("requested zero length: "
+		       "NODE: id %u, type %#x, height %u, "
+		       "node_size %u, offset %u, len %u\n",
+		       node->this, node->type, node->height,
+		       node->tree->node_size, off, len);
+		return;
+	}
+
+	len = check_and_correct_requested_length(node, off, len);
+
+	off += node->page_offset;
+	pagep = node->page + (off >> PAGE_SHIFT);
+	off &= ~PAGE_MASK;
+
+	l = min_t(u32, len, PAGE_SIZE - off);
+	memcpy_to_page(*pagep, off, buf, l);
+	set_page_dirty(*pagep);
+
+	while ((len -= l) != 0) {
+		buf += l;
+		l = min_t(u32, len, PAGE_SIZE);
+		memcpy_to_page(*++pagep, 0, buf, l);
+		set_page_dirty(*pagep);
+	}
+}
+
+void hfs_bnode_write_u16(struct hfs_bnode *node, u32 off, u16 data)
+{
+	__be16 v = cpu_to_be16(data);
+	/* TODO: optimize later... */
+	hfs_bnode_write(node, &v, off, 2);
+}
+
+void hfs_bnode_clear(struct hfs_bnode *node, u32 off, u32 len)
+{
+	struct page **pagep;
+	u32 l;
+
+	if (!is_bnode_offset_valid(node, off))
+		return;
+
+	if (len == 0) {
+		pr_err("requested zero length: "
+		       "NODE: id %u, type %#x, height %u, "
+		       "node_size %u, offset %u, len %u\n",
+		       node->this, node->type, node->height,
+		       node->tree->node_size, off, len);
+		return;
+	}
+
+	len = check_and_correct_requested_length(node, off, len);
+
+	off += node->page_offset;
+	pagep = node->page + (off >> PAGE_SHIFT);
+	off &= ~PAGE_MASK;
+
+	l = min_t(u32, len, PAGE_SIZE - off);
+	memzero_page(*pagep, off, l);
+	set_page_dirty(*pagep);
+
+	while ((len -= l) != 0) {
+		l = min_t(u32, len, PAGE_SIZE);
+		memzero_page(*++pagep, 0, l);
+		set_page_dirty(*pagep);
+	}
+}
+
+void hfs_bnode_copy(struct hfs_bnode *dst_node, u32 dst,
+		    struct hfs_bnode *src_node, u32 src, u32 len)
+{
+	struct page **src_page, **dst_page;
+	u32 l;
+
+	hfs_dbg("dst %u, src %u, len %u\n", dst, src, len);
+	if (!len)
+		return;
+
+	len = check_and_correct_requested_length(src_node, src, len);
+	len = check_and_correct_requested_length(dst_node, dst, len);
+
+	src += src_node->page_offset;
+	dst += dst_node->page_offset;
+	src_page = src_node->page + (src >> PAGE_SHIFT);
+	src &= ~PAGE_MASK;
+	dst_page = dst_node->page + (dst >> PAGE_SHIFT);
+	dst &= ~PAGE_MASK;
+
+	if (src == dst) {
+		l = min_t(u32, len, PAGE_SIZE - src);
+		memcpy_page(*dst_page, src, *src_page, src, l);
+		set_page_dirty(*dst_page);
+
+		while ((len -= l) != 0) {
+			l = min_t(u32, len, PAGE_SIZE);
+			memcpy_page(*++dst_page, 0, *++src_page, 0, l);
+			set_page_dirty(*dst_page);
+		}
+	} else {
+		void *src_ptr, *dst_ptr;
+
+		do {
+			dst_ptr = kmap_local_page(*dst_page) + dst;
+			src_ptr = kmap_local_page(*src_page) + src;
+			if (PAGE_SIZE - src < PAGE_SIZE - dst) {
+				l = PAGE_SIZE - src;
+				src = 0;
+				dst += l;
+			} else {
+				l = PAGE_SIZE - dst;
+				src += l;
+				dst = 0;
+			}
+			l = min(len, l);
+			memcpy(dst_ptr, src_ptr, l);
+			kunmap_local(src_ptr);
+			set_page_dirty(*dst_page);
+			kunmap_local(dst_ptr);
+			if (!dst)
+				dst_page++;
+			else
+				src_page++;
+		} while ((len -= l));
+	}
+}
+
+void hfs_bnode_move(struct hfs_bnode *node, u32 dst, u32 src, u32 len)
+{
+	struct page **src_page, **dst_page;
+	void *src_ptr, *dst_ptr;
+	u32 l;
+
+	hfs_dbg("dst %u, src %u, len %u\n", dst, src, len);
+	if (!len)
+		return;
+
+	len = check_and_correct_requested_length(node, src, len);
+	len = check_and_correct_requested_length(node, dst, len);
+
+	src += node->page_offset;
+	dst += node->page_offset;
+	if (dst > src) {
+		src += len - 1;
+		src_page = node->page + (src >> PAGE_SHIFT);
+		src = (src & ~PAGE_MASK) + 1;
+		dst += len - 1;
+		dst_page = node->page + (dst >> PAGE_SHIFT);
+		dst = (dst & ~PAGE_MASK) + 1;
+
+		if (src == dst) {
+			while (src < len) {
+				dst_ptr = kmap_local_page(*dst_page);
+				src_ptr = kmap_local_page(*src_page);
+				memmove(dst_ptr, src_ptr, src);
+				kunmap_local(src_ptr);
+				set_page_dirty(*dst_page);
+				kunmap_local(dst_ptr);
+				len -= src;
+				src = PAGE_SIZE;
+				src_page--;
+				dst_page--;
+			}
+			src -= len;
+			dst_ptr = kmap_local_page(*dst_page);
+			src_ptr = kmap_local_page(*src_page);
+			memmove(dst_ptr + src, src_ptr + src, len);
+			kunmap_local(src_ptr);
+			set_page_dirty(*dst_page);
+			kunmap_local(dst_ptr);
+		} else {
+			do {
+				dst_ptr = kmap_local_page(*dst_page) + dst;
+				src_ptr = kmap_local_page(*src_page) + src;
+				if (src < dst) {
+					l = src;
+					src = PAGE_SIZE;
+					dst -= l;
+				} else {
+					l = dst;
+					src -= l;
+					dst = PAGE_SIZE;
+				}
+				l = min(len, l);
+				memmove(dst_ptr - l, src_ptr - l, l);
+				kunmap_local(src_ptr);
+				set_page_dirty(*dst_page);
+				kunmap_local(dst_ptr);
+				if (dst == PAGE_SIZE)
+					dst_page--;
+				else
+					src_page--;
+			} while ((len -= l));
+		}
+	} else {
+		src_page = node->page + (src >> PAGE_SHIFT);
+		src &= ~PAGE_MASK;
+		dst_page = node->page + (dst >> PAGE_SHIFT);
+		dst &= ~PAGE_MASK;
+
+		if (src == dst) {
+			l = min_t(u32, len, PAGE_SIZE - src);
+
+			dst_ptr = kmap_local_page(*dst_page) + src;
+			src_ptr = kmap_local_page(*src_page) + src;
+			memmove(dst_ptr, src_ptr, l);
+			kunmap_local(src_ptr);
+			set_page_dirty(*dst_page);
+			kunmap_local(dst_ptr);
+
+			while ((len -= l) != 0) {
+				l = min_t(u32, len, PAGE_SIZE);
+				dst_ptr = kmap_local_page(*++dst_page);
+				src_ptr = kmap_local_page(*++src_page);
+				memmove(dst_ptr, src_ptr, l);
+				kunmap_local(src_ptr);
+				set_page_dirty(*dst_page);
+				kunmap_local(dst_ptr);
+			}
+		} else {
+			do {
+				dst_ptr = kmap_local_page(*dst_page) + dst;
+				src_ptr = kmap_local_page(*src_page) + src;
+				if (PAGE_SIZE - src <
+						PAGE_SIZE - dst) {
+					l = PAGE_SIZE - src;
+					src = 0;
+					dst += l;
+				} else {
+					l = PAGE_SIZE - dst;
+					src += l;
+					dst = 0;
+				}
+				l = min(len, l);
+				memmove(dst_ptr, src_ptr, l);
+				kunmap_local(src_ptr);
+				set_page_dirty(*dst_page);
+				kunmap_local(dst_ptr);
+				if (!dst)
+					dst_page++;
+				else
+					src_page++;
+			} while ((len -= l));
+		}
+	}
+}
+
+void hfs_bnode_dump(struct hfs_bnode *node)
+{
+	struct hfs_bnode_desc desc;
+	__be32 cnid;
+	int i, off, key_off;
+	u16 num_recs;
+
+	hfs_dbg("node %d\n", node->this);
+	hfs_bnode_read(node, &desc, 0, sizeof(desc));
+	num_recs = node->num_recs;
+	hfs_dbg("next %d, prev %d, type %d, height %d, num_recs %d\n",
+		be32_to_cpu(desc.next), be32_to_cpu(desc.prev),
+		desc.type, desc.height, be16_to_cpu(desc.num_recs));
+
+	if (hfs_bnode_num_recs_invalid(node)) {
+		hfs_dbg("invalid num_recs %u\n", num_recs);
+		return;
+	}
+
+	off = node->tree->node_size - 2;
+	for (i = num_recs; i >= 0; off -= 2, i--) {
+		key_off = hfs_bnode_read_u16(node, off);
+		hfs_dbg(" key_off %d", key_off);
+		if (i && node->type == HFS_NODE_INDEX) {
+			int tmp;
+
+			if (node->tree->attributes & HFS_TREE_VARIDXKEYS ||
+					node->tree->cnid == HFSPLUS_ATTR_CNID)
+				tmp = hfs_bnode_read_u16(node, key_off) + 2;
+			else
+				tmp = node->tree->max_key_len + 2;
+			hfs_dbg(" (%d", tmp);
+			hfs_bnode_read(node, &cnid, key_off + tmp, 4);
+			hfs_dbg(", cnid %d)", be32_to_cpu(cnid));
+		} else if (i && node->type == HFS_NODE_LEAF) {
+			int tmp;
+
+			tmp = hfs_bnode_read_u16(node, key_off);
+			hfs_dbg(" (%d)", tmp);
+		}
+	}
+	hfs_dbg("\n");
+}
+
+void hfs_bnode_unlink(struct hfs_bnode *node)
+{
+	struct hfs_btree *tree;
+	struct hfs_bnode *tmp;
+	__be32 cnid;
+
+	tree = node->tree;
+	if (node->prev) {
+		tmp = hfs_bnode_find(tree, node->prev);
+		if (IS_ERR(tmp))
+			return;
+		tmp->next = node->next;
+		cnid = cpu_to_be32(tmp->next);
+		hfs_bnode_write(tmp, &cnid,
+			offsetof(struct hfs_bnode_desc, next), 4);
+		hfs_bnode_put(tmp);
+	} else if (node->type == HFS_NODE_LEAF)
+		tree->leaf_head = node->next;
+
+	if (node->next) {
+		tmp = hfs_bnode_find(tree, node->next);
+		if (IS_ERR(tmp))
+			return;
+		tmp->prev = node->prev;
+		cnid = cpu_to_be32(tmp->prev);
+		hfs_bnode_write(tmp, &cnid,
+			offsetof(struct hfs_bnode_desc, prev), 4);
+		hfs_bnode_put(tmp);
+	} else if (node->type == HFS_NODE_LEAF)
+		tree->leaf_tail = node->prev;
+
+	/* move down? */
+	if (!node->prev && !node->next)
+		hfs_dbg("btree delete level\n");
+	if (!node->parent) {
+		tree->root = 0;
+		tree->depth = 0;
+	}
+
+	spin_lock(&tree->hash_lock);
+	set_bit(HFS_BNODE_DELETED, &node->flags);
+	spin_unlock(&tree->hash_lock);
+}
+
+static inline int hfs_bnode_hash(u32 num)
+{
+	num = (num >> 16) + num;
+	num += num >> 8;
+	return num & (NODE_HASH_SIZE - 1);
+}
+
+struct hfs_bnode *hfs_bnode_findhash(struct hfs_btree *tree, u32 cnid)
+{
+	struct hfs_bnode *node;
+
+	if (cnid >= tree->node_count) {
+		pr_err("request for non-existent node %d in B*Tree\n",
+		       cnid);
+		return NULL;
+	}
+
+	for (node = tree->node_hash[hfs_bnode_hash(cnid)];
+			node; node = node->next_hash)
+		if (node->this == cnid)
+			return node;
+	return NULL;
+}
+
+static struct hfs_bnode *__hfs_bnode_create(struct hfs_btree *tree, u32 cnid)
+{
+	struct hfs_bnode *node, *node2;
+	struct address_space *mapping;
+	struct page *page;
+	int block, i, hash;
+	loff_t off;
+
+	if (cnid >= tree->node_count) {
+		pr_err("request for non-existent node %d in B*Tree\n",
+		       cnid);
+		return NULL;
+	}
+
+	node = kzalloc_flex(*node, page, tree->pages_per_bnode);
+	if (!node)
+		return NULL;
+	node->tree = tree;
+	node->this = cnid;
+	set_bit(HFS_BNODE_NEW, &node->flags);
+	atomic_set(&node->refcnt, 1);
+	hfs_dbg("cnid %d, node %d, refcnt 1\n",
+		node->tree->cnid, node->this);
+	init_waitqueue_head(&node->lock_wq);
+	spin_lock(&tree->hash_lock);
+	node2 = hfs_bnode_findhash(tree, cnid);
+	if (!node2) {
+		hash = hfs_bnode_hash(cnid);
+		node->next_hash = tree->node_hash[hash];
+		tree->node_hash[hash] = node;
+		tree->node_hash_cnt++;
+	} else {
+		hfs_bnode_get(node2);
+		spin_unlock(&tree->hash_lock);
+		kfree(node);
+		wait_event(node2->lock_wq,
+			!test_bit(HFS_BNODE_NEW, &node2->flags));
+		return node2;
+	}
+	spin_unlock(&tree->hash_lock);
+
+	mapping = tree->inode->i_mapping;
+	off = (loff_t)cnid << tree->node_size_shift;
+	block = off >> PAGE_SHIFT;
+	node->page_offset = off & ~PAGE_MASK;
+	for (i = 0; i < tree->pages_per_bnode; block++, i++) {
+		page = read_mapping_page(mapping, block, NULL);
+		if (IS_ERR(page))
+			goto fail;
+		node->page[i] = page;
+	}
+
+	return node;
+fail:
+	set_bit(HFS_BNODE_ERROR, &node->flags);
+	return node;
+}
+
+void hfs_bnode_unhash(struct hfs_bnode *node)
+{
+	struct hfs_bnode **p;
+
+	hfs_dbg("cnid %d, node %d, refcnt %d\n",
+		node->tree->cnid, node->this, atomic_read(&node->refcnt));
+	for (p = &node->tree->node_hash[hfs_bnode_hash(node->this)];
+	     *p && *p != node; p = &(*p)->next_hash)
+		;
+	BUG_ON(!*p);
+	*p = node->next_hash;
+	node->tree->node_hash_cnt--;
+}
+
+/* Load a particular node out of a tree */
+struct hfs_bnode *hfs_bnode_find(struct hfs_btree *tree, u32 num)
+{
+	struct hfs_bnode *node;
+	struct hfs_bnode_desc *desc;
+	int i, rec_off, off, next_off;
+	int entry_size, key_size;
+
+	spin_lock(&tree->hash_lock);
+	node = hfs_bnode_findhash(tree, num);
+	if (node) {
+		hfs_bnode_get(node);
+		spin_unlock(&tree->hash_lock);
+		wait_event(node->lock_wq,
+			!test_bit(HFS_BNODE_NEW, &node->flags));
+		if (test_bit(HFS_BNODE_ERROR, &node->flags))
+			goto node_error;
+		return node;
+	}
+	spin_unlock(&tree->hash_lock);
+	node = __hfs_bnode_create(tree, num);
+	if (!node)
+		return ERR_PTR(-ENOMEM);
+	if (test_bit(HFS_BNODE_ERROR, &node->flags))
+		goto node_error;
+	if (!test_bit(HFS_BNODE_NEW, &node->flags))
+		return node;
+
+	desc = (struct hfs_bnode_desc *)(kmap_local_page(node->page[0]) +
+							 node->page_offset);
+	node->prev = be32_to_cpu(desc->prev);
+	node->next = be32_to_cpu(desc->next);
+	node->num_recs = be16_to_cpu(desc->num_recs);
+	node->type = desc->type;
+	node->height = desc->height;
+	kunmap_local(desc);
+
+	if (hfs_bnode_num_recs_invalid(node))
+		goto node_error;
+
+	switch (node->type) {
+	case HFS_NODE_HEADER:
+	case HFS_NODE_MAP:
+		if (node->height != 0)
+			goto node_error;
+		break;
+	case HFS_NODE_LEAF:
+		if (node->height != 1)
+			goto node_error;
+		break;
+	case HFS_NODE_INDEX:
+		if (node->height <= 1 || node->height > tree->depth)
+			goto node_error;
+		break;
+	default:
+		goto node_error;
+	}
+
+	rec_off = tree->node_size - 2;
+	off = hfs_bnode_read_u16(node, rec_off);
+	if (off != sizeof(struct hfs_bnode_desc))
+		goto node_error;
+	for (i = 1; i <= node->num_recs; off = next_off, i++) {
+		rec_off -= 2;
+		next_off = hfs_bnode_read_u16(node, rec_off);
+		if (hfs_brec_offsets_invalid(node, off, next_off))
+			goto node_error;
+		entry_size = next_off - off;
+		if (node->type != HFS_NODE_INDEX &&
+		    node->type != HFS_NODE_LEAF)
+			continue;
+		key_size = hfs_bnode_read_u16(node, off) + 2;
+		if (key_size >= entry_size || key_size & 1)
+			goto node_error;
+	}
+	clear_bit(HFS_BNODE_NEW, &node->flags);
+	wake_up(&node->lock_wq);
+	return node;
+
+node_error:
+	set_bit(HFS_BNODE_ERROR, &node->flags);
+	clear_bit(HFS_BNODE_NEW, &node->flags);
+	wake_up(&node->lock_wq);
+	hfs_bnode_put(node);
+	return ERR_PTR(-EIO);
+}
+
+void hfs_bnode_free(struct hfs_bnode *node)
+{
+	int i;
+
+	for (i = 0; i < node->tree->pages_per_bnode; i++)
+		if (node->page[i])
+			put_page(node->page[i]);
+	kfree(node);
+}
+
+struct hfs_bnode *hfs_bnode_create(struct hfs_btree *tree, u32 num)
+{
+	struct hfs_bnode *node;
+	struct page **pagep;
+	int i;
+
+	spin_lock(&tree->hash_lock);
+	node = hfs_bnode_findhash(tree, num);
+	spin_unlock(&tree->hash_lock);
+	if (node) {
+		pr_crit("new node %u already hashed?\n", num);
+		WARN_ON(1);
+		return ERR_PTR(-EEXIST);
+	}
+	node = __hfs_bnode_create(tree, num);
+	if (!node)
+		return ERR_PTR(-ENOMEM);
+	if (test_bit(HFS_BNODE_ERROR, &node->flags)) {
+		hfs_bnode_put(node);
+		return ERR_PTR(-EIO);
+	}
+
+	pagep = node->page;
+	memzero_page(*pagep, node->page_offset,
+		     min_t(int, PAGE_SIZE, tree->node_size));
+	set_page_dirty(*pagep);
+	for (i = 1; i < tree->pages_per_bnode; i++) {
+		memzero_page(*++pagep, 0, PAGE_SIZE);
+		set_page_dirty(*pagep);
+	}
+	clear_bit(HFS_BNODE_NEW, &node->flags);
+	wake_up(&node->lock_wq);
+
+	return node;
+}
+
+void hfs_bnode_get(struct hfs_bnode *node)
+{
+	if (node) {
+		atomic_inc(&node->refcnt);
+		hfs_dbg("cnid %d, node %d, refcnt %d\n",
+			node->tree->cnid, node->this,
+			atomic_read(&node->refcnt));
+	}
+}
+
+/* Dispose of resources used by a node */
+void hfs_bnode_put(struct hfs_bnode *node)
+{
+	if (node) {
+		struct hfs_btree *tree = node->tree;
+		int i;
+
+		hfs_dbg("cnid %d, node %d, refcnt %d\n",
+			node->tree->cnid, node->this,
+			atomic_read(&node->refcnt));
+		BUG_ON(!atomic_read(&node->refcnt));
+		if (!atomic_dec_and_lock(&node->refcnt, &tree->hash_lock))
+			return;
+		for (i = 0; i < tree->pages_per_bnode; i++) {
+			if (!node->page[i])
+				continue;
+			mark_page_accessed(node->page[i]);
+		}
+
+		if (test_bit(HFS_BNODE_DELETED, &node->flags)) {
+			hfs_bnode_unhash(node);
+			spin_unlock(&tree->hash_lock);
+			if (hfs_bnode_need_zeroout(tree))
+				hfs_bnode_clear(node, 0, tree->node_size);
+			hfs_bmap_free(node);
+			hfs_bnode_free(node);
+			return;
+		}
+		spin_unlock(&tree->hash_lock);
+	}
+}
+
+/*
+ * Unused nodes have to be zeroed if this is the catalog tree and
+ * a corresponding flag in the volume header is set.
+ */
+bool hfs_bnode_need_zeroout(struct hfs_btree *tree)
+{
+	struct super_block *sb = tree->inode->i_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
+	const u32 volume_attr = be32_to_cpu(sbi->s_vhdr->attributes);
+
+	return volume_attr & HFSPLUS_VOL_UNUSED_NODE_FIX;
+}