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, 112 insertions, 0 deletions

diff --git a/fs/cachefiles/security.c b/fs/cachefiles/security.c
new file mode 100644
index 000000000..fc6611886
--- /dev/null
+++ b/fs/cachefiles/security.c
@@ -0,0 +1,112 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* CacheFiles security management
+ *
+ * Copyright (C) 2007, 2021 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include <linux/fs.h>
+#include <linux/cred.h>
+#include "internal.h"
+
+/*
+ * determine the security context within which we access the cache from within
+ * the kernel
+ */
+int cachefiles_get_security_ID(struct cachefiles_cache *cache)
+{
+	struct cred *new;
+	int ret;
+
+	_enter("{%u}", cache->have_secid ? cache->secid : 0);
+
+	new = prepare_kernel_cred(current);
+	if (!new) {
+		ret = -ENOMEM;
+		goto error;
+	}
+
+	if (cache->have_secid) {
+		ret = set_security_override(new, cache->secid);
+		if (ret < 0) {
+			put_cred(new);
+			pr_err("Security denies permission to nominate security context: error %d\n",
+			       ret);
+			goto error;
+		}
+	}
+
+	cache->cache_cred = new;
+	ret = 0;
+error:
+	_leave(" = %d", ret);
+	return ret;
+}
+
+/*
+ * see if mkdir and create can be performed in the root directory
+ */
+static int cachefiles_check_cache_dir(struct cachefiles_cache *cache,
+				      struct dentry *root)
+{
+	int ret;
+
+	ret = security_inode_mkdir(d_backing_inode(root), root, 0);
+	if (ret < 0) {
+		pr_err("Security denies permission to make dirs: error %d",
+		       ret);
+		return ret;
+	}
+
+	ret = security_inode_create(d_backing_inode(root), root, 0);
+	if (ret < 0)
+		pr_err("Security denies permission to create files: error %d",
+		       ret);
+
+	return ret;
+}
+
+/*
+ * check the security details of the on-disk cache
+ * - must be called with security override in force
+ * - must return with a security override in force - even in the case of an
+ *   error
+ */
+int cachefiles_determine_cache_security(struct cachefiles_cache *cache,
+					struct dentry *root,
+					const struct cred **_saved_cred)
+{
+	struct cred *new;
+	int ret;
+
+	_enter("");
+
+	/* duplicate the cache creds for COW (the override is currently in
+	 * force, so we can use prepare_creds() to do this) */
+	new = prepare_creds();
+	if (!new)
+		return -ENOMEM;
+
+	cachefiles_end_secure(cache, *_saved_cred);
+
+	/* use the cache root dir's security context as the basis with
+	 * which create files */
+	ret = set_create_files_as(new, d_backing_inode(root));
+	if (ret < 0) {
+		abort_creds(new);
+		cachefiles_begin_secure(cache, _saved_cred);
+		_leave(" = %d [cfa]", ret);
+		return ret;
+	}
+
+	put_cred(cache->cache_cred);
+	cache->cache_cred = new;
+
+	cachefiles_begin_secure(cache, _saved_cred);
+	ret = cachefiles_check_cache_dir(cache, root);
+
+	if (ret == -EOPNOTSUPP)
+		ret = 0;
+	_leave(" = %d", ret);
+	return ret;
+}