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, 563 insertions, 0 deletions

diff --git a/fs/hfsplus/brec.c b/fs/hfsplus/brec.c
new file mode 100644
index 000000000..f416e562d
--- /dev/null
+++ b/fs/hfsplus/brec.c
@@ -0,0 +1,563 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  linux/fs/hfsplus/brec.c
+ *
+ * Copyright (C) 2001
+ * Brad Boyer (flar@allandria.com)
+ * (C) 2003 Ardis Technologies <roman@ardistech.com>
+ *
+ * Handle individual btree records
+ */
+
+#include <linux/limits.h>
+
+#include "hfsplus_fs.h"
+#include "hfsplus_raw.h"
+
+static struct hfs_bnode *hfs_bnode_split(struct hfs_find_data *fd);
+static int hfs_brec_update_parent(struct hfs_find_data *fd);
+static int hfs_btree_inc_height(struct hfs_btree *);
+
+/* Get the length and offset of the given record in the given node */
+u16 hfs_brec_lenoff(struct hfs_bnode *node, u16 rec, u16 *off)
+{
+	__be16 retval[2];
+	u16 data_off;
+	u16 next_off;
+
+	if (hfs_brec_record_invalid(node, rec)) {
+		*off = U16_MAX;
+		return U16_MAX;
+	}
+
+	data_off = node->tree->node_size - (rec + 2) * 2;
+	hfs_bnode_read(node, retval, data_off, 4);
+	*off = be16_to_cpu(retval[1]);
+	next_off = be16_to_cpu(retval[0]);
+	if (hfs_brec_offsets_invalid(node, *off, next_off)) {
+		*off = U16_MAX;
+		return U16_MAX;
+	}
+	return next_off - *off;
+}
+
+/* Get the length of the key from a keyed record */
+u16 hfs_brec_keylen(struct hfs_bnode *node, u16 rec)
+{
+	u16 retval, recoff, len;
+
+	if (node->type != HFS_NODE_INDEX && node->type != HFS_NODE_LEAF)
+		return 0;
+	if (hfs_brec_record_invalid(node, rec))
+		return U16_MAX;
+
+	if ((node->type == HFS_NODE_INDEX) &&
+	   !(node->tree->attributes & HFS_TREE_VARIDXKEYS) &&
+	   (node->tree->cnid != HFSPLUS_ATTR_CNID)) {
+		retval = node->tree->max_key_len + 2;
+	} else {
+		len = hfs_brec_lenoff(node, rec, &recoff);
+		if (hfs_brec_len_invalid(node, len))
+			return len;
+
+		retval = hfs_bnode_read_u16(node, recoff) + 2;
+		if (retval > node->tree->max_key_len + 2) {
+			pr_err("keylen %d too large\n",
+				retval);
+			retval = U16_MAX;
+		}
+	}
+	return retval;
+}
+
+int hfs_brec_insert(struct hfs_find_data *fd, void *entry, u32 entry_len)
+{
+	struct hfs_btree *tree;
+	struct hfs_bnode *node, *new_node;
+	int size, key_len, rec;
+	int data_off, end_off;
+	int idx_rec_off, data_rec_off, end_rec_off;
+	__be32 cnid;
+
+	tree = fd->tree;
+	if (!fd->bnode) {
+		if (!tree->root)
+			hfs_btree_inc_height(tree);
+		node = hfs_bnode_find(tree, tree->leaf_head);
+		if (IS_ERR(node))
+			return PTR_ERR(node);
+		fd->bnode = node;
+		fd->record = -1;
+	}
+	new_node = NULL;
+	key_len = be16_to_cpu(fd->search_key->key_len) + 2;
+again:
+	/* new record idx and complete record size */
+	rec = fd->record + 1;
+	size = key_len + entry_len;
+
+	node = fd->bnode;
+	hfs_bnode_dump(node);
+	/* get last offset */
+	end_rec_off = tree->node_size - (node->num_recs + 1) * 2;
+	end_off = hfs_bnode_read_u16(node, end_rec_off);
+	end_rec_off -= 2;
+	hfs_dbg("rec %d, size %d, end_off %d, end_rec_off %d\n",
+		rec, size, end_off, end_rec_off);
+	if (size > end_rec_off - end_off) {
+		if (new_node)
+			panic("not enough room!\n");
+		new_node = hfs_bnode_split(fd);
+		if (IS_ERR(new_node))
+			return PTR_ERR(new_node);
+		goto again;
+	}
+	if (node->type == HFS_NODE_LEAF) {
+		tree->leaf_count++;
+		mark_inode_dirty(tree->inode);
+	}
+	node->num_recs++;
+	/* write new last offset */
+	hfs_bnode_write_u16(node,
+		offsetof(struct hfs_bnode_desc, num_recs),
+		node->num_recs);
+	hfs_bnode_write_u16(node, end_rec_off, end_off + size);
+	data_off = end_off;
+	data_rec_off = end_rec_off + 2;
+	idx_rec_off = tree->node_size - (rec + 1) * 2;
+	if (idx_rec_off == data_rec_off)
+		goto skip;
+	/* move all following entries */
+	do {
+		data_off = hfs_bnode_read_u16(node, data_rec_off + 2);
+		hfs_bnode_write_u16(node, data_rec_off, data_off + size);
+		data_rec_off += 2;
+	} while (data_rec_off < idx_rec_off);
+
+	/* move data away */
+	hfs_bnode_move(node, data_off + size, data_off,
+		       end_off - data_off);
+
+skip:
+	hfs_bnode_write(node, fd->search_key, data_off, key_len);
+	hfs_bnode_write(node, entry, data_off + key_len, entry_len);
+	hfs_bnode_dump(node);
+
+	/*
+	 * update parent key if we inserted a key
+	 * at the start of the node and it is not the new node
+	 */
+	if (!rec && new_node != node) {
+		hfs_bnode_read_key(node, fd->search_key, data_off + size);
+		hfs_brec_update_parent(fd);
+	}
+
+	if (new_node) {
+		hfs_bnode_put(fd->bnode);
+		if (!new_node->parent) {
+			hfs_btree_inc_height(tree);
+			new_node->parent = tree->root;
+		}
+		fd->bnode = hfs_bnode_find(tree, new_node->parent);
+
+		/* create index data entry */
+		cnid = cpu_to_be32(new_node->this);
+		entry = &cnid;
+		entry_len = sizeof(cnid);
+
+		/* get index key */
+		hfs_bnode_read_key(new_node, fd->search_key, 14);
+		__hfs_brec_find(fd->bnode, fd, hfs_find_rec_by_key);
+
+		hfs_bnode_put(new_node);
+		new_node = NULL;
+
+		if ((tree->attributes & HFS_TREE_VARIDXKEYS) ||
+				(tree->cnid == HFSPLUS_ATTR_CNID))
+			key_len = be16_to_cpu(fd->search_key->key_len) + 2;
+		else {
+			fd->search_key->key_len =
+				cpu_to_be16(tree->max_key_len);
+			key_len = tree->max_key_len + 2;
+		}
+		goto again;
+	}
+
+	return 0;
+}
+
+int hfs_brec_remove(struct hfs_find_data *fd)
+{
+	struct hfs_btree *tree;
+	struct hfs_bnode *node, *parent;
+	int end_off, rec_off, data_off, size;
+	int res;
+
+	tree = fd->tree;
+	node = fd->bnode;
+again:
+	if (hfs_brec_record_invalid(node, fd->record))
+		return -EINVAL;
+
+	rec_off = tree->node_size - (fd->record + 2) * 2;
+	end_off = tree->node_size - (node->num_recs + 1) * 2;
+
+	if (node->type == HFS_NODE_LEAF) {
+		if (tree->leaf_count == 0)
+			return -EINVAL;
+		tree->leaf_count--;
+		mark_inode_dirty(tree->inode);
+	}
+	hfs_bnode_dump(node);
+	hfs_dbg("rec %d, len %d\n",
+		fd->record, fd->keylength + fd->entrylength);
+	if (!--node->num_recs) {
+		hfs_bnode_unlink(node);
+		if (!node->parent)
+			return 0;
+		parent = hfs_bnode_find(tree, node->parent);
+		if (IS_ERR(parent))
+			return PTR_ERR(parent);
+		hfs_bnode_put(node);
+		node = fd->bnode = parent;
+
+		res = __hfs_brec_find(node, fd, hfs_find_rec_by_key);
+		if (res && res != -ENOENT)
+			return res;
+		goto again;
+	}
+	hfs_bnode_write_u16(node,
+		offsetof(struct hfs_bnode_desc, num_recs),
+		node->num_recs);
+
+	if (rec_off == end_off)
+		goto skip;
+	size = fd->keylength + fd->entrylength;
+
+	do {
+		data_off = hfs_bnode_read_u16(node, rec_off);
+		hfs_bnode_write_u16(node, rec_off + 2, data_off - size);
+		rec_off -= 2;
+	} while (rec_off >= end_off);
+
+	/* fill hole */
+	hfs_bnode_move(node, fd->keyoffset, fd->keyoffset + size,
+		       data_off - fd->keyoffset - size);
+skip:
+	hfs_bnode_dump(node);
+	if (!fd->record)
+		hfs_brec_update_parent(fd);
+	return 0;
+}
+
+static struct hfs_bnode *hfs_bnode_split(struct hfs_find_data *fd)
+{
+	struct hfs_btree *tree;
+	struct hfs_bnode *node, *new_node, *next_node;
+	struct hfs_bnode_desc node_desc;
+	int num_recs, new_rec_off, new_off, old_rec_off;
+	int data_start, data_end, size;
+	size_t rec_off_tbl_size;
+	size_t node_desc_size = sizeof(struct hfs_bnode_desc);
+	size_t rec_size = sizeof(__be16);
+
+	tree = fd->tree;
+	node = fd->bnode;
+	new_node = hfs_bmap_alloc(tree);
+	if (IS_ERR(new_node))
+		return new_node;
+	hfs_bnode_get(node);
+	hfs_dbg("this %d - new %d - next %d\n",
+		node->this, new_node->this, node->next);
+	new_node->next = node->next;
+	new_node->prev = node->this;
+	new_node->parent = node->parent;
+	new_node->type = node->type;
+	new_node->height = node->height;
+
+	if (node->next)
+		next_node = hfs_bnode_find(tree, node->next);
+	else
+		next_node = NULL;
+
+	if (IS_ERR(next_node)) {
+		hfs_bnode_put(node);
+		hfs_bnode_put(new_node);
+		return next_node;
+	}
+
+	rec_off_tbl_size = node->num_recs * rec_size;
+	size = tree->node_size / 2;
+	size -= node_desc_size;
+	size -= rec_off_tbl_size;
+	old_rec_off = tree->node_size - (2 * rec_size);
+
+	num_recs = 1;
+	for (;;) {
+		data_start = hfs_bnode_read_u16(node, old_rec_off);
+		if (data_start > size)
+			break;
+		old_rec_off -= rec_size;
+		if (++num_recs < node->num_recs)
+			continue;
+		hfs_bnode_put(node);
+		hfs_bnode_unlink(new_node);
+		hfs_bnode_put(new_node);
+		if (next_node)
+			hfs_bnode_put(next_node);
+		return ERR_PTR(-ENOSPC);
+	}
+
+	if (fd->record + 1 < num_recs) {
+		/* new record is in the lower half,
+		 * so leave some more space there
+		 */
+		old_rec_off += rec_size;
+		num_recs--;
+		data_start = hfs_bnode_read_u16(node, old_rec_off);
+	} else {
+		hfs_bnode_put(node);
+		hfs_bnode_get(new_node);
+		fd->bnode = new_node;
+		fd->record -= num_recs;
+		fd->keyoffset -= data_start - node_desc_size;
+		fd->entryoffset -= data_start - node_desc_size;
+	}
+	new_node->num_recs = node->num_recs - num_recs;
+	node->num_recs = num_recs;
+
+	new_rec_off = tree->node_size - rec_size;
+	new_off = node_desc_size;
+	size = data_start - new_off;
+	num_recs = new_node->num_recs;
+	data_end = data_start;
+	while (num_recs) {
+		hfs_bnode_write_u16(new_node, new_rec_off, new_off);
+		old_rec_off -= rec_size;
+		new_rec_off -= rec_size;
+		data_end = hfs_bnode_read_u16(node, old_rec_off);
+		new_off = data_end - size;
+		num_recs--;
+	}
+	hfs_bnode_write_u16(new_node, new_rec_off, new_off);
+	hfs_bnode_copy(new_node, node_desc_size,
+			node, data_start, data_end - data_start);
+
+	/* update new bnode header */
+	node_desc.next = cpu_to_be32(new_node->next);
+	node_desc.prev = cpu_to_be32(new_node->prev);
+	node_desc.type = new_node->type;
+	node_desc.height = new_node->height;
+	node_desc.num_recs = cpu_to_be16(new_node->num_recs);
+	node_desc.reserved = 0;
+	hfs_bnode_write(new_node, &node_desc, 0, sizeof(node_desc));
+
+	/* update previous bnode header */
+	node->next = new_node->this;
+	hfs_bnode_read(node, &node_desc, 0, sizeof(node_desc));
+	node_desc.next = cpu_to_be32(node->next);
+	node_desc.num_recs = cpu_to_be16(node->num_recs);
+	hfs_bnode_write(node, &node_desc, 0, sizeof(node_desc));
+
+	/* update next bnode header */
+	if (next_node) {
+		next_node->prev = new_node->this;
+		hfs_bnode_read(next_node, &node_desc, 0, sizeof(node_desc));
+		node_desc.prev = cpu_to_be32(next_node->prev);
+		hfs_bnode_write(next_node, &node_desc, 0, sizeof(node_desc));
+		hfs_bnode_put(next_node);
+	} else if (node->this == tree->leaf_tail) {
+		/* if there is no next node, this might be the new tail */
+		tree->leaf_tail = new_node->this;
+		mark_inode_dirty(tree->inode);
+	}
+
+	hfs_bnode_dump(node);
+	hfs_bnode_dump(new_node);
+	hfs_bnode_put(node);
+
+	return new_node;
+}
+
+static int hfs_brec_update_parent(struct hfs_find_data *fd)
+{
+	struct hfs_btree *tree;
+	struct hfs_bnode *node, *new_node, *parent;
+	int newkeylen, diff;
+	int rec, rec_off, end_rec_off;
+	int start_off, end_off;
+	int res;
+
+	tree = fd->tree;
+	node = fd->bnode;
+	new_node = NULL;
+	if (!node->parent)
+		return 0;
+
+again:
+	parent = hfs_bnode_find(tree, node->parent);
+	if (IS_ERR(parent))
+		return PTR_ERR(parent);
+	res = __hfs_brec_find(parent, fd, hfs_find_rec_by_key);
+	if (res && res != -ENOENT)
+		return res;
+	if (fd->record < 0)
+		return -ENOENT;
+	hfs_bnode_dump(parent);
+	rec = fd->record;
+
+	/* size difference between old and new key */
+	if ((tree->attributes & HFS_TREE_VARIDXKEYS) ||
+				(tree->cnid == HFSPLUS_ATTR_CNID))
+		newkeylen = hfs_bnode_read_u16(node, 14) + 2;
+	else
+		fd->keylength = newkeylen = tree->max_key_len + 2;
+	hfs_dbg("rec %d, keylength %d, newkeylen %d\n",
+		rec, fd->keylength, newkeylen);
+
+	rec_off = tree->node_size - (rec + 2) * 2;
+	end_rec_off = tree->node_size - (parent->num_recs + 1) * 2;
+	diff = newkeylen - fd->keylength;
+	if (!diff)
+		goto skip;
+	if (diff > 0) {
+		end_off = hfs_bnode_read_u16(parent, end_rec_off);
+		if (end_rec_off - end_off < diff) {
+
+			hfs_dbg("splitting index node\n");
+			fd->bnode = parent;
+			new_node = hfs_bnode_split(fd);
+			if (IS_ERR(new_node))
+				return PTR_ERR(new_node);
+			parent = fd->bnode;
+			rec = fd->record;
+			rec_off = tree->node_size - (rec + 2) * 2;
+			end_rec_off = tree->node_size -
+				(parent->num_recs + 1) * 2;
+		}
+	}
+
+	end_off = start_off = hfs_bnode_read_u16(parent, rec_off);
+	hfs_bnode_write_u16(parent, rec_off, start_off + diff);
+	start_off -= 4;	/* move previous cnid too */
+
+	while (rec_off > end_rec_off) {
+		rec_off -= 2;
+		end_off = hfs_bnode_read_u16(parent, rec_off);
+		hfs_bnode_write_u16(parent, rec_off, end_off + diff);
+	}
+	hfs_bnode_move(parent, start_off + diff, start_off,
+		       end_off - start_off);
+skip:
+	hfs_bnode_copy(parent, fd->keyoffset, node, 14, newkeylen);
+	hfs_bnode_dump(parent);
+
+	hfs_bnode_put(node);
+	node = parent;
+
+	if (new_node) {
+		__be32 cnid;
+
+		if (!new_node->parent) {
+			hfs_btree_inc_height(tree);
+			new_node->parent = tree->root;
+		}
+		fd->bnode = hfs_bnode_find(tree, new_node->parent);
+		/* create index key and entry */
+		hfs_bnode_read_key(new_node, fd->search_key, 14);
+		cnid = cpu_to_be32(new_node->this);
+
+		__hfs_brec_find(fd->bnode, fd, hfs_find_rec_by_key);
+		hfs_brec_insert(fd, &cnid, sizeof(cnid));
+		hfs_bnode_put(fd->bnode);
+		hfs_bnode_put(new_node);
+
+		if (!rec) {
+			if (new_node == node)
+				goto out;
+			/* restore search_key */
+			hfs_bnode_read_key(node, fd->search_key, 14);
+		}
+		new_node = NULL;
+	}
+
+	if (!rec && node->parent)
+		goto again;
+out:
+	fd->bnode = node;
+	return 0;
+}
+
+static int hfs_btree_inc_height(struct hfs_btree *tree)
+{
+	struct hfs_bnode *node, *new_node;
+	struct hfs_bnode_desc node_desc;
+	int key_size, rec;
+	__be32 cnid;
+
+	node = NULL;
+	if (tree->root) {
+		node = hfs_bnode_find(tree, tree->root);
+		if (IS_ERR(node))
+			return PTR_ERR(node);
+	}
+	new_node = hfs_bmap_alloc(tree);
+	if (IS_ERR(new_node)) {
+		hfs_bnode_put(node);
+		return PTR_ERR(new_node);
+	}
+
+	tree->root = new_node->this;
+	if (!tree->depth) {
+		tree->leaf_head = tree->leaf_tail = new_node->this;
+		new_node->type = HFS_NODE_LEAF;
+		new_node->num_recs = 0;
+	} else {
+		new_node->type = HFS_NODE_INDEX;
+		new_node->num_recs = 1;
+	}
+	new_node->parent = 0;
+	new_node->next = 0;
+	new_node->prev = 0;
+	new_node->height = ++tree->depth;
+
+	node_desc.next = cpu_to_be32(new_node->next);
+	node_desc.prev = cpu_to_be32(new_node->prev);
+	node_desc.type = new_node->type;
+	node_desc.height = new_node->height;
+	node_desc.num_recs = cpu_to_be16(new_node->num_recs);
+	node_desc.reserved = 0;
+	hfs_bnode_write(new_node, &node_desc, 0, sizeof(node_desc));
+
+	rec = tree->node_size - 2;
+	hfs_bnode_write_u16(new_node, rec, 14);
+
+	if (node) {
+		/* insert old root idx into new root */
+		node->parent = tree->root;
+		if (node->type == HFS_NODE_LEAF ||
+				tree->attributes & HFS_TREE_VARIDXKEYS ||
+				tree->cnid == HFSPLUS_ATTR_CNID)
+			key_size = hfs_bnode_read_u16(node, 14) + 2;
+		else
+			key_size = tree->max_key_len + 2;
+		hfs_bnode_copy(new_node, 14, node, 14, key_size);
+
+		if (!(tree->attributes & HFS_TREE_VARIDXKEYS) &&
+				(tree->cnid != HFSPLUS_ATTR_CNID)) {
+			key_size = tree->max_key_len + 2;
+			hfs_bnode_write_u16(new_node, 14, tree->max_key_len);
+		}
+		cnid = cpu_to_be32(node->this);
+		hfs_bnode_write(new_node, &cnid, 14 + key_size, 4);
+
+		rec -= 2;
+		hfs_bnode_write_u16(new_node, rec, 14 + key_size + 4);
+
+		hfs_bnode_put(node);
+	}
+	hfs_bnode_put(new_node);
+	mark_inode_dirty(tree->inode);
+
+	return 0;
+}