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, 230 insertions, 0 deletions

diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c
new file mode 100644
index 000000000..79536b2e3
--- /dev/null
+++ b/fs/ubifs/ioctl.c
@@ -0,0 +1,230 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * This file is part of UBIFS.
+ *
+ * Copyright (C) 2006-2008 Nokia Corporation.
+ * Copyright (C) 2006, 2007 University of Szeged, Hungary
+ *
+ * Authors: Zoltan Sogor
+ *          Artem Bityutskiy (Битюцкий Артём)
+ *          Adrian Hunter
+ */
+
+/* This file implements EXT2-compatible extended attribute ioctl() calls */
+
+#include <linux/compat.h>
+#include <linux/mount.h>
+#include <linux/fileattr.h>
+#include "ubifs.h"
+
+/* Need to be kept consistent with checked flags in ioctl2ubifs() */
+#define UBIFS_SETTABLE_IOCTL_FLAGS \
+	(FS_COMPR_FL | FS_SYNC_FL | FS_APPEND_FL | \
+	 FS_IMMUTABLE_FL | FS_DIRSYNC_FL)
+
+/* Need to be kept consistent with checked flags in ubifs2ioctl() */
+#define UBIFS_GETTABLE_IOCTL_FLAGS \
+	(UBIFS_SETTABLE_IOCTL_FLAGS | FS_ENCRYPT_FL)
+
+/**
+ * ubifs_set_inode_flags - set VFS inode flags.
+ * @inode: VFS inode to set flags for
+ *
+ * This function propagates flags from UBIFS inode object to VFS inode object.
+ */
+void ubifs_set_inode_flags(struct inode *inode)
+{
+	unsigned int flags = ubifs_inode(inode)->flags;
+
+	inode->i_flags &= ~(S_SYNC | S_APPEND | S_IMMUTABLE | S_DIRSYNC |
+			    S_ENCRYPTED);
+	if (flags & UBIFS_SYNC_FL)
+		inode->i_flags |= S_SYNC;
+	if (flags & UBIFS_APPEND_FL)
+		inode->i_flags |= S_APPEND;
+	if (flags & UBIFS_IMMUTABLE_FL)
+		inode->i_flags |= S_IMMUTABLE;
+	if (flags & UBIFS_DIRSYNC_FL)
+		inode->i_flags |= S_DIRSYNC;
+	if (flags & UBIFS_CRYPT_FL)
+		inode->i_flags |= S_ENCRYPTED;
+}
+
+/*
+ * ioctl2ubifs - convert ioctl inode flags to UBIFS inode flags.
+ * @ioctl_flags: flags to convert
+ *
+ * This function converts ioctl flags (@FS_COMPR_FL, etc) to UBIFS inode flags
+ * (@UBIFS_COMPR_FL, etc).
+ */
+static int ioctl2ubifs(int ioctl_flags)
+{
+	int ubifs_flags = 0;
+
+	if (ioctl_flags & FS_COMPR_FL)
+		ubifs_flags |= UBIFS_COMPR_FL;
+	if (ioctl_flags & FS_SYNC_FL)
+		ubifs_flags |= UBIFS_SYNC_FL;
+	if (ioctl_flags & FS_APPEND_FL)
+		ubifs_flags |= UBIFS_APPEND_FL;
+	if (ioctl_flags & FS_IMMUTABLE_FL)
+		ubifs_flags |= UBIFS_IMMUTABLE_FL;
+	if (ioctl_flags & FS_DIRSYNC_FL)
+		ubifs_flags |= UBIFS_DIRSYNC_FL;
+
+	return ubifs_flags;
+}
+
+/*
+ * ubifs2ioctl - convert UBIFS inode flags to ioctl inode flags.
+ * @ubifs_flags: flags to convert
+ *
+ * This function converts UBIFS inode flags (@UBIFS_COMPR_FL, etc) to ioctl
+ * flags (@FS_COMPR_FL, etc).
+ */
+static int ubifs2ioctl(int ubifs_flags)
+{
+	int ioctl_flags = 0;
+
+	if (ubifs_flags & UBIFS_COMPR_FL)
+		ioctl_flags |= FS_COMPR_FL;
+	if (ubifs_flags & UBIFS_SYNC_FL)
+		ioctl_flags |= FS_SYNC_FL;
+	if (ubifs_flags & UBIFS_APPEND_FL)
+		ioctl_flags |= FS_APPEND_FL;
+	if (ubifs_flags & UBIFS_IMMUTABLE_FL)
+		ioctl_flags |= FS_IMMUTABLE_FL;
+	if (ubifs_flags & UBIFS_DIRSYNC_FL)
+		ioctl_flags |= FS_DIRSYNC_FL;
+	if (ubifs_flags & UBIFS_CRYPT_FL)
+		ioctl_flags |= FS_ENCRYPT_FL;
+
+	return ioctl_flags;
+}
+
+static int setflags(struct inode *inode, int flags)
+{
+	int err, release;
+	struct ubifs_inode *ui = ubifs_inode(inode);
+	struct ubifs_info *c = inode->i_sb->s_fs_info;
+	struct ubifs_budget_req req = { .dirtied_ino = 1,
+			.dirtied_ino_d = ALIGN(ui->data_len, 8) };
+
+	err = ubifs_budget_space(c, &req);
+	if (err)
+		return err;
+
+	mutex_lock(&ui->ui_mutex);
+	ui->flags &= ~ioctl2ubifs(UBIFS_SETTABLE_IOCTL_FLAGS);
+	ui->flags |= ioctl2ubifs(flags);
+	ubifs_set_inode_flags(inode);
+	inode_set_ctime_current(inode);
+	release = ui->dirty;
+	mark_inode_dirty_sync(inode);
+	mutex_unlock(&ui->ui_mutex);
+
+	if (release)
+		ubifs_release_budget(c, &req);
+	if (IS_SYNC(inode))
+		err = write_inode_now(inode, 1);
+	return err;
+}
+
+int ubifs_fileattr_get(struct dentry *dentry, struct file_kattr *fa)
+{
+	struct inode *inode = d_inode(dentry);
+	int flags = ubifs2ioctl(ubifs_inode(inode)->flags);
+
+	if (d_is_special(dentry))
+		return -ENOTTY;
+
+	dbg_gen("get flags: %#x, i_flags %#x", flags, inode->i_flags);
+	fileattr_fill_flags(fa, flags);
+
+	return 0;
+}
+
+int ubifs_fileattr_set(struct mnt_idmap *idmap,
+		       struct dentry *dentry, struct file_kattr *fa)
+{
+	struct inode *inode = d_inode(dentry);
+	int flags = fa->flags;
+
+	if (d_is_special(dentry))
+		return -ENOTTY;
+
+	if (fileattr_has_fsx(fa))
+		return -EOPNOTSUPP;
+
+	if (flags & ~UBIFS_GETTABLE_IOCTL_FLAGS)
+		return -EOPNOTSUPP;
+
+	flags &= UBIFS_SETTABLE_IOCTL_FLAGS;
+
+	if (!S_ISDIR(inode->i_mode))
+		flags &= ~FS_DIRSYNC_FL;
+
+	dbg_gen("set flags: %#x, i_flags %#x", flags, inode->i_flags);
+	return setflags(inode, flags);
+}
+
+long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	int err;
+	struct inode *inode = file_inode(file);
+
+	switch (cmd) {
+	case FS_IOC_SET_ENCRYPTION_POLICY: {
+		struct ubifs_info *c = inode->i_sb->s_fs_info;
+
+		err = ubifs_enable_encryption(c);
+		if (err)
+			return err;
+
+		return fscrypt_ioctl_set_policy(file, (const void __user *)arg);
+	}
+	case FS_IOC_GET_ENCRYPTION_POLICY:
+		return fscrypt_ioctl_get_policy(file, (void __user *)arg);
+
+	case FS_IOC_GET_ENCRYPTION_POLICY_EX:
+		return fscrypt_ioctl_get_policy_ex(file, (void __user *)arg);
+
+	case FS_IOC_ADD_ENCRYPTION_KEY:
+		return fscrypt_ioctl_add_key(file, (void __user *)arg);
+
+	case FS_IOC_REMOVE_ENCRYPTION_KEY:
+		return fscrypt_ioctl_remove_key(file, (void __user *)arg);
+
+	case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS:
+		return fscrypt_ioctl_remove_key_all_users(file,
+							  (void __user *)arg);
+	case FS_IOC_GET_ENCRYPTION_KEY_STATUS:
+		return fscrypt_ioctl_get_key_status(file, (void __user *)arg);
+
+	case FS_IOC_GET_ENCRYPTION_NONCE:
+		return fscrypt_ioctl_get_nonce(file, (void __user *)arg);
+
+	default:
+		return -ENOTTY;
+	}
+}
+
+#ifdef CONFIG_COMPAT
+long ubifs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	switch (cmd) {
+	case FS_IOC_SET_ENCRYPTION_POLICY:
+	case FS_IOC_GET_ENCRYPTION_POLICY:
+	case FS_IOC_GET_ENCRYPTION_POLICY_EX:
+	case FS_IOC_ADD_ENCRYPTION_KEY:
+	case FS_IOC_REMOVE_ENCRYPTION_KEY:
+	case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS:
+	case FS_IOC_GET_ENCRYPTION_KEY_STATUS:
+	case FS_IOC_GET_ENCRYPTION_NONCE:
+		break;
+	default:
+		return -ENOIOCTLCMD;
+	}
+	return ubifs_ioctl(file, cmd, (unsigned long)compat_ptr(arg));
+}
+#endif