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, 271 insertions, 0 deletions

diff --git a/fs/squashfs/xattr.c b/fs/squashfs/xattr.c
new file mode 100644
index 000000000..ce6608cab
--- /dev/null
+++ b/fs/squashfs/xattr.c
@@ -0,0 +1,271 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Squashfs - a compressed read only filesystem for Linux
+ *
+ * Copyright (c) 2010
+ * Phillip Lougher <phillip@squashfs.org.uk>
+ *
+ * xattr.c
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/fs.h>
+#include <linux/vfs.h>
+#include <linux/xattr.h>
+#include <linux/slab.h>
+
+#include "squashfs_fs.h"
+#include "squashfs_fs_sb.h"
+#include "squashfs_fs_i.h"
+#include "squashfs.h"
+
+static const struct xattr_handler *squashfs_xattr_handler(int);
+
+ssize_t squashfs_listxattr(struct dentry *d, char *buffer,
+	size_t buffer_size)
+{
+	struct inode *inode = d_inode(d);
+	struct super_block *sb = inode->i_sb;
+	struct squashfs_sb_info *msblk = sb->s_fs_info;
+	u64 start = SQUASHFS_XATTR_BLK(squashfs_i(inode)->xattr)
+						 + msblk->xattr_table;
+	int offset = SQUASHFS_XATTR_OFFSET(squashfs_i(inode)->xattr);
+	int count = squashfs_i(inode)->xattr_count;
+	size_t rest = buffer_size;
+	int err;
+
+	/* check that the file system has xattrs */
+	if (msblk->xattr_id_table == NULL)
+		return -EOPNOTSUPP;
+
+	/* loop reading each xattr name */
+	while (count--) {
+		struct squashfs_xattr_entry entry;
+		struct squashfs_xattr_val val;
+		const struct xattr_handler *handler;
+		int name_size;
+
+		err = squashfs_read_metadata(sb, &entry, &start, &offset,
+							sizeof(entry));
+		if (err < 0)
+			goto failed;
+
+		name_size = le16_to_cpu(entry.size);
+		handler = squashfs_xattr_handler(le16_to_cpu(entry.type));
+		if (handler && (!handler->list || handler->list(d))) {
+			const char *prefix = handler->prefix ?: handler->name;
+			size_t prefix_size = strlen(prefix);
+
+			if (buffer) {
+				if (prefix_size + name_size + 1 > rest) {
+					err = -ERANGE;
+					goto failed;
+				}
+				memcpy(buffer, prefix, prefix_size);
+				buffer += prefix_size;
+			}
+			err = squashfs_read_metadata(sb, buffer, &start,
+				&offset, name_size);
+			if (err < 0)
+				goto failed;
+			if (buffer) {
+				buffer[name_size] = '\0';
+				buffer += name_size + 1;
+			}
+			rest -= prefix_size + name_size + 1;
+		} else  {
+			/* no handler or insuffficient privileges, so skip */
+			err = squashfs_read_metadata(sb, NULL, &start,
+				&offset, name_size);
+			if (err < 0)
+				goto failed;
+		}
+
+
+		/* skip remaining xattr entry */
+		err = squashfs_read_metadata(sb, &val, &start, &offset,
+						sizeof(val));
+		if (err < 0)
+			goto failed;
+
+		err = squashfs_read_metadata(sb, NULL, &start, &offset,
+						le32_to_cpu(val.vsize));
+		if (err < 0)
+			goto failed;
+	}
+	err = buffer_size - rest;
+
+failed:
+	return err;
+}
+
+
+static int squashfs_xattr_get(struct inode *inode, int name_index,
+	const char *name, void *buffer, size_t buffer_size)
+{
+	struct super_block *sb = inode->i_sb;
+	struct squashfs_sb_info *msblk = sb->s_fs_info;
+	u64 start = SQUASHFS_XATTR_BLK(squashfs_i(inode)->xattr)
+						 + msblk->xattr_table;
+	int offset = SQUASHFS_XATTR_OFFSET(squashfs_i(inode)->xattr);
+	int count = squashfs_i(inode)->xattr_count;
+	int name_len = strlen(name);
+	int err, vsize;
+	char *target = kmalloc(name_len, GFP_KERNEL);
+
+	if (target == NULL)
+		return  -ENOMEM;
+
+	/* loop reading each xattr name */
+	for (; count; count--) {
+		struct squashfs_xattr_entry entry;
+		struct squashfs_xattr_val val;
+		int type, prefix, name_size;
+
+		err = squashfs_read_metadata(sb, &entry, &start, &offset,
+							sizeof(entry));
+		if (err < 0)
+			goto failed;
+
+		name_size = le16_to_cpu(entry.size);
+		type = le16_to_cpu(entry.type);
+		prefix = type & SQUASHFS_XATTR_PREFIX_MASK;
+
+		if (prefix == name_index && name_size == name_len)
+			err = squashfs_read_metadata(sb, target, &start,
+						&offset, name_size);
+		else
+			err = squashfs_read_metadata(sb, NULL, &start,
+						&offset, name_size);
+		if (err < 0)
+			goto failed;
+
+		if (prefix == name_index && name_size == name_len &&
+					strncmp(target, name, name_size) == 0) {
+			/* found xattr */
+			if (type & SQUASHFS_XATTR_VALUE_OOL) {
+				__le64 xattr_val;
+				u64 xattr;
+				/* val is a reference to the real location */
+				err = squashfs_read_metadata(sb, &val, &start,
+						&offset, sizeof(val));
+				if (err < 0)
+					goto failed;
+				err = squashfs_read_metadata(sb, &xattr_val,
+					&start, &offset, sizeof(xattr_val));
+				if (err < 0)
+					goto failed;
+				xattr = le64_to_cpu(xattr_val);
+				start = SQUASHFS_XATTR_BLK(xattr) +
+							msblk->xattr_table;
+				offset = SQUASHFS_XATTR_OFFSET(xattr);
+			}
+			/* read xattr value */
+			err = squashfs_read_metadata(sb, &val, &start, &offset,
+							sizeof(val));
+			if (err < 0)
+				goto failed;
+
+			vsize = le32_to_cpu(val.vsize);
+			if (buffer) {
+				if (vsize > buffer_size) {
+					err = -ERANGE;
+					goto failed;
+				}
+				err = squashfs_read_metadata(sb, buffer, &start,
+					 &offset, vsize);
+				if (err < 0)
+					goto failed;
+			}
+			break;
+		}
+
+		/* no match, skip remaining xattr entry */
+		err = squashfs_read_metadata(sb, &val, &start, &offset,
+							sizeof(val));
+		if (err < 0)
+			goto failed;
+		err = squashfs_read_metadata(sb, NULL, &start, &offset,
+						le32_to_cpu(val.vsize));
+		if (err < 0)
+			goto failed;
+	}
+	err = count ? vsize : -ENODATA;
+
+failed:
+	kfree(target);
+	return err;
+}
+
+
+static int squashfs_xattr_handler_get(const struct xattr_handler *handler,
+				      struct dentry *unused,
+				      struct inode *inode,
+				      const char *name,
+				      void *buffer, size_t size)
+{
+	return squashfs_xattr_get(inode, handler->flags, name,
+		buffer, size);
+}
+
+/*
+ * User namespace support
+ */
+static const struct xattr_handler squashfs_xattr_user_handler = {
+	.prefix	= XATTR_USER_PREFIX,
+	.flags	= SQUASHFS_XATTR_USER,
+	.get	= squashfs_xattr_handler_get
+};
+
+/*
+ * Trusted namespace support
+ */
+static bool squashfs_trusted_xattr_handler_list(struct dentry *d)
+{
+	return capable(CAP_SYS_ADMIN);
+}
+
+static const struct xattr_handler squashfs_xattr_trusted_handler = {
+	.prefix	= XATTR_TRUSTED_PREFIX,
+	.flags	= SQUASHFS_XATTR_TRUSTED,
+	.list	= squashfs_trusted_xattr_handler_list,
+	.get	= squashfs_xattr_handler_get
+};
+
+/*
+ * Security namespace support
+ */
+static const struct xattr_handler squashfs_xattr_security_handler = {
+	.prefix	= XATTR_SECURITY_PREFIX,
+	.flags	= SQUASHFS_XATTR_SECURITY,
+	.get	= squashfs_xattr_handler_get
+};
+
+static const struct xattr_handler *squashfs_xattr_handler(int type)
+{
+	if (type & ~(SQUASHFS_XATTR_PREFIX_MASK | SQUASHFS_XATTR_VALUE_OOL))
+		/* ignore unrecognised type */
+		return NULL;
+
+	switch (type & SQUASHFS_XATTR_PREFIX_MASK) {
+	case SQUASHFS_XATTR_USER:
+		return &squashfs_xattr_user_handler;
+	case SQUASHFS_XATTR_TRUSTED:
+		return &squashfs_xattr_trusted_handler;
+	case SQUASHFS_XATTR_SECURITY:
+		return &squashfs_xattr_security_handler;
+	default:
+		/* ignore unrecognised type */
+		return NULL;
+	}
+}
+
+const struct xattr_handler * const squashfs_xattr_handlers[] = {
+	&squashfs_xattr_user_handler,
+	&squashfs_xattr_trusted_handler,
+	&squashfs_xattr_security_handler,
+	NULL
+};
+