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, 161 insertions, 0 deletions

diff --git a/fs/fuse/acl.c b/fs/fuse/acl.c
new file mode 100644
index 000000000..31fb50e16
--- /dev/null
+++ b/fs/fuse/acl.c
@@ -0,0 +1,161 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * FUSE: Filesystem in Userspace
+ * Copyright (C) 2016 Canonical Ltd. <seth.forshee@canonical.com>
+ */
+
+#include "fuse_i.h"
+
+#include <linux/posix_acl.h>
+#include <linux/posix_acl_xattr.h>
+
+static struct posix_acl *__fuse_get_acl(struct fuse_conn *fc,
+					struct inode *inode, int type, bool rcu)
+{
+	int size;
+	const char *name;
+	void *value = NULL;
+	struct posix_acl *acl;
+
+	if (rcu)
+		return ERR_PTR(-ECHILD);
+
+	if (fuse_is_bad(inode))
+		return ERR_PTR(-EIO);
+
+	if (fc->no_getxattr)
+		return NULL;
+
+	if (type == ACL_TYPE_ACCESS)
+		name = XATTR_NAME_POSIX_ACL_ACCESS;
+	else if (type == ACL_TYPE_DEFAULT)
+		name = XATTR_NAME_POSIX_ACL_DEFAULT;
+	else
+		return ERR_PTR(-EOPNOTSUPP);
+
+	value = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!value)
+		return ERR_PTR(-ENOMEM);
+	size = fuse_getxattr(inode, name, value, PAGE_SIZE);
+	if (size > 0)
+		acl = posix_acl_from_xattr(fc->user_ns, value, size);
+	else if ((size == 0) || (size == -ENODATA) ||
+		 (size == -EOPNOTSUPP && fc->no_getxattr))
+		acl = NULL;
+	else if (size == -ERANGE)
+		acl = ERR_PTR(-E2BIG);
+	else
+		acl = ERR_PTR(size);
+
+	kfree(value);
+	return acl;
+}
+
+static inline bool fuse_no_acl(const struct fuse_conn *fc,
+			       const struct inode *inode)
+{
+	/*
+	 * Refuse interacting with POSIX ACLs for daemons that
+	 * don't support FUSE_POSIX_ACL and are not mounted on
+	 * the host to retain backwards compatibility.
+	 */
+	return !fc->posix_acl && (i_user_ns(inode) != &init_user_ns);
+}
+
+struct posix_acl *fuse_get_acl(struct mnt_idmap *idmap,
+			       struct dentry *dentry, int type)
+{
+	struct inode *inode = d_inode(dentry);
+	struct fuse_conn *fc = get_fuse_conn(inode);
+
+	if (fuse_no_acl(fc, inode))
+		return ERR_PTR(-EOPNOTSUPP);
+
+	return __fuse_get_acl(fc, inode, type, false);
+}
+
+struct posix_acl *fuse_get_inode_acl(struct inode *inode, int type, bool rcu)
+{
+	struct fuse_conn *fc = get_fuse_conn(inode);
+
+	/*
+	 * FUSE daemons before FUSE_POSIX_ACL was introduced could get and set
+	 * POSIX ACLs without them being used for permission checking by the
+	 * vfs. Retain that behavior for backwards compatibility as there are
+	 * filesystems that do all permission checking for acls in the daemon
+	 * and not in the kernel.
+	 */
+	if (!fc->posix_acl)
+		return NULL;
+	return __fuse_get_acl(fc,  inode, type, rcu);
+}
+
+int fuse_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
+		 struct posix_acl *acl, int type)
+{
+	struct inode *inode = d_inode(dentry);
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	const char *name;
+	int ret;
+
+	if (fuse_is_bad(inode))
+		return -EIO;
+
+	if (fc->no_setxattr || fuse_no_acl(fc, inode))
+		return -EOPNOTSUPP;
+
+	if (type == ACL_TYPE_ACCESS)
+		name = XATTR_NAME_POSIX_ACL_ACCESS;
+	else if (type == ACL_TYPE_DEFAULT)
+		name = XATTR_NAME_POSIX_ACL_DEFAULT;
+	else
+		return -EINVAL;
+
+	if (acl) {
+		unsigned int extra_flags = 0;
+		/*
+		 * Fuse userspace is responsible for updating access
+		 * permissions in the inode, if needed. fuse_setxattr
+		 * invalidates the inode attributes, which will force
+		 * them to be refreshed the next time they are used,
+		 * and it also updates i_ctime.
+		 */
+		size_t size;
+		void *value;
+
+		value = posix_acl_to_xattr(fc->user_ns, acl, &size, GFP_KERNEL);
+		if (!value)
+			return -ENOMEM;
+
+		if (size > PAGE_SIZE) {
+			kfree(value);
+			return -E2BIG;
+		}
+
+		/*
+		 * Fuse daemons without FUSE_POSIX_ACL never changed the passed
+		 * through POSIX ACLs. Such daemons don't expect setgid bits to
+		 * be stripped.
+		 */
+		if (fc->posix_acl &&
+		    !in_group_or_capable(idmap, inode,
+					 i_gid_into_vfsgid(idmap, inode)))
+			extra_flags |= FUSE_SETXATTR_ACL_KILL_SGID;
+
+		ret = fuse_setxattr(inode, name, value, size, 0, extra_flags);
+		kfree(value);
+	} else {
+		ret = fuse_removexattr(inode, name);
+	}
+
+	if (fc->posix_acl) {
+		/*
+		 * Fuse daemons without FUSE_POSIX_ACL never cached POSIX ACLs
+		 * and didn't invalidate attributes. Retain that behavior.
+		 */
+		forget_all_cached_acls(inode);
+		fuse_invalidate_attr(inode);
+	}
+
+	return ret;
+}