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, 455 insertions, 0 deletions

diff --git a/fs/f2fs/acl.c b/fs/f2fs/acl.c
new file mode 100644
index 000000000..d32535491
--- /dev/null
+++ b/fs/f2fs/acl.c
@@ -0,0 +1,455 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * fs/f2fs/acl.c
+ *
+ * Copyright (c) 2012 Samsung Electronics Co., Ltd.
+ *             http://www.samsung.com/
+ *
+ * Portions of this code from linux/fs/ext2/acl.c
+ *
+ * Copyright (C) 2001-2003 Andreas Gruenbacher, <agruen@suse.de>
+ */
+#include <linux/fs_struct.h>
+#include <linux/f2fs_fs.h>
+#include "f2fs.h"
+#include "xattr.h"
+#include "acl.h"
+
+static inline size_t f2fs_acl_size(int count)
+{
+	if (count <= 4) {
+		return sizeof(struct f2fs_acl_header) +
+			count * sizeof(struct f2fs_acl_entry_short);
+	} else {
+		return sizeof(struct f2fs_acl_header) +
+			4 * sizeof(struct f2fs_acl_entry_short) +
+			(count - 4) * sizeof(struct f2fs_acl_entry);
+	}
+}
+
+static inline int f2fs_acl_count(size_t size)
+{
+	ssize_t s;
+
+	size -= sizeof(struct f2fs_acl_header);
+	s = size - 4 * sizeof(struct f2fs_acl_entry_short);
+	if (s < 0) {
+		if (size % sizeof(struct f2fs_acl_entry_short))
+			return -1;
+		return size / sizeof(struct f2fs_acl_entry_short);
+	} else {
+		if (s % sizeof(struct f2fs_acl_entry))
+			return -1;
+		return s / sizeof(struct f2fs_acl_entry) + 4;
+	}
+}
+
+static struct posix_acl *f2fs_acl_from_disk(const char *value, size_t size)
+{
+	int i, count;
+	int err = -EINVAL;
+	struct posix_acl *acl;
+	struct f2fs_acl_header *hdr = (struct f2fs_acl_header *)value;
+	struct f2fs_acl_entry *entry = (struct f2fs_acl_entry *)(hdr + 1);
+	const char *end = value + size;
+
+	if (size < sizeof(struct f2fs_acl_header))
+		return ERR_PTR(-EINVAL);
+
+	if (hdr->a_version != cpu_to_le32(F2FS_ACL_VERSION))
+		return ERR_PTR(-EINVAL);
+
+	count = f2fs_acl_count(size);
+	if (count < 0)
+		return ERR_PTR(-EINVAL);
+	if (count == 0)
+		return NULL;
+
+	acl = posix_acl_alloc(count, GFP_NOFS);
+	if (!acl)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < count; i++) {
+
+		if (unlikely((char *)entry +
+				sizeof(struct f2fs_acl_entry_short) > end)) {
+			err = -EFSCORRUPTED;
+			goto fail;
+		}
+
+		acl->a_entries[i].e_tag  = le16_to_cpu(entry->e_tag);
+		acl->a_entries[i].e_perm = le16_to_cpu(entry->e_perm);
+
+		switch (acl->a_entries[i].e_tag) {
+		case ACL_USER_OBJ:
+		case ACL_GROUP_OBJ:
+		case ACL_MASK:
+		case ACL_OTHER:
+			entry = (struct f2fs_acl_entry *)((char *)entry +
+					sizeof(struct f2fs_acl_entry_short));
+			break;
+
+		case ACL_USER:
+			if (unlikely((char *)entry +
+					sizeof(struct f2fs_acl_entry) > end)) {
+				err = -EFSCORRUPTED;
+				goto fail;
+			}
+			acl->a_entries[i].e_uid =
+				make_kuid(&init_user_ns,
+						le32_to_cpu(entry->e_id));
+			entry = (struct f2fs_acl_entry *)((char *)entry +
+					sizeof(struct f2fs_acl_entry));
+			break;
+		case ACL_GROUP:
+			if (unlikely((char *)entry +
+					sizeof(struct f2fs_acl_entry) > end)) {
+				err = -EFSCORRUPTED;
+				goto fail;
+			}
+			acl->a_entries[i].e_gid =
+				make_kgid(&init_user_ns,
+						le32_to_cpu(entry->e_id));
+			entry = (struct f2fs_acl_entry *)((char *)entry +
+					sizeof(struct f2fs_acl_entry));
+			break;
+		default:
+			goto fail;
+		}
+	}
+	if ((char *)entry != end)
+		goto fail;
+	return acl;
+fail:
+	posix_acl_release(acl);
+	return ERR_PTR(err);
+}
+
+static void *f2fs_acl_to_disk(struct f2fs_sb_info *sbi,
+				const struct posix_acl *acl, size_t *size)
+{
+	struct f2fs_acl_header *f2fs_acl;
+	struct f2fs_acl_entry *entry;
+	int i;
+
+	f2fs_acl = f2fs_kmalloc(sbi, sizeof(struct f2fs_acl_header) +
+			acl->a_count * sizeof(struct f2fs_acl_entry),
+			GFP_NOFS);
+	if (!f2fs_acl)
+		return ERR_PTR(-ENOMEM);
+
+	f2fs_acl->a_version = cpu_to_le32(F2FS_ACL_VERSION);
+	entry = (struct f2fs_acl_entry *)(f2fs_acl + 1);
+
+	for (i = 0; i < acl->a_count; i++) {
+
+		entry->e_tag  = cpu_to_le16(acl->a_entries[i].e_tag);
+		entry->e_perm = cpu_to_le16(acl->a_entries[i].e_perm);
+
+		switch (acl->a_entries[i].e_tag) {
+		case ACL_USER:
+			entry->e_id = cpu_to_le32(
+					from_kuid(&init_user_ns,
+						acl->a_entries[i].e_uid));
+			entry = (struct f2fs_acl_entry *)((char *)entry +
+					sizeof(struct f2fs_acl_entry));
+			break;
+		case ACL_GROUP:
+			entry->e_id = cpu_to_le32(
+					from_kgid(&init_user_ns,
+						acl->a_entries[i].e_gid));
+			entry = (struct f2fs_acl_entry *)((char *)entry +
+					sizeof(struct f2fs_acl_entry));
+			break;
+		case ACL_USER_OBJ:
+		case ACL_GROUP_OBJ:
+		case ACL_MASK:
+		case ACL_OTHER:
+			entry = (struct f2fs_acl_entry *)((char *)entry +
+					sizeof(struct f2fs_acl_entry_short));
+			break;
+		default:
+			goto fail;
+		}
+	}
+	*size = f2fs_acl_size(acl->a_count);
+	return (void *)f2fs_acl;
+
+fail:
+	kfree(f2fs_acl);
+	return ERR_PTR(-EINVAL);
+}
+
+static struct posix_acl *__f2fs_get_acl(struct inode *inode, int type,
+						struct folio *dfolio)
+{
+	int name_index = F2FS_XATTR_INDEX_POSIX_ACL_DEFAULT;
+	void *value = NULL;
+	struct posix_acl *acl;
+	int retval;
+
+	if (type == ACL_TYPE_ACCESS)
+		name_index = F2FS_XATTR_INDEX_POSIX_ACL_ACCESS;
+
+	retval = f2fs_getxattr(inode, name_index, "", NULL, 0, dfolio);
+	if (retval > 0) {
+		value = f2fs_kmalloc(F2FS_I_SB(inode), retval, GFP_F2FS_ZERO);
+		if (!value)
+			return ERR_PTR(-ENOMEM);
+		retval = f2fs_getxattr(inode, name_index, "", value,
+							retval, dfolio);
+	}
+
+	if (retval > 0)
+		acl = f2fs_acl_from_disk(value, retval);
+	else if (retval == -ENODATA)
+		acl = NULL;
+	else
+		acl = ERR_PTR(retval);
+	kfree(value);
+
+	return acl;
+}
+
+struct posix_acl *f2fs_get_acl(struct inode *inode, int type, bool rcu)
+{
+	if (rcu)
+		return ERR_PTR(-ECHILD);
+
+	return __f2fs_get_acl(inode, type, NULL);
+}
+
+static int f2fs_acl_update_mode(struct mnt_idmap *idmap,
+				struct inode *inode, umode_t *mode_p,
+				struct posix_acl **acl)
+{
+	umode_t mode = inode->i_mode;
+	int error;
+
+	if (is_inode_flag_set(inode, FI_ACL_MODE))
+		mode = F2FS_I(inode)->i_acl_mode;
+
+	error = posix_acl_equiv_mode(*acl, &mode);
+	if (error < 0)
+		return error;
+	if (error == 0)
+		*acl = NULL;
+	if (!in_group_or_capable(idmap, inode, i_gid_into_vfsgid(idmap, inode)))
+		mode &= ~S_ISGID;
+	*mode_p = mode;
+	return 0;
+}
+
+static int __f2fs_set_acl(struct mnt_idmap *idmap,
+			struct inode *inode, int type,
+			struct posix_acl *acl, struct folio *ifolio)
+{
+	int name_index;
+	void *value = NULL;
+	size_t size = 0;
+	int error;
+	umode_t mode = inode->i_mode;
+
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		name_index = F2FS_XATTR_INDEX_POSIX_ACL_ACCESS;
+		if (acl && !ifolio) {
+			error = f2fs_acl_update_mode(idmap, inode, &mode, &acl);
+			if (error)
+				return error;
+			set_acl_inode(inode, mode);
+		}
+		break;
+
+	case ACL_TYPE_DEFAULT:
+		name_index = F2FS_XATTR_INDEX_POSIX_ACL_DEFAULT;
+		if (!S_ISDIR(inode->i_mode))
+			return acl ? -EACCES : 0;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	if (acl) {
+		value = f2fs_acl_to_disk(F2FS_I_SB(inode), acl, &size);
+		if (IS_ERR(value)) {
+			clear_inode_flag(inode, FI_ACL_MODE);
+			return PTR_ERR(value);
+		}
+	}
+
+	error = f2fs_setxattr(inode, name_index, "", value, size, ifolio, 0);
+
+	kfree(value);
+	if (!error)
+		set_cached_acl(inode, type, acl);
+
+	clear_inode_flag(inode, FI_ACL_MODE);
+	return error;
+}
+
+int f2fs_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
+		 struct posix_acl *acl, int type)
+{
+	struct inode *inode = d_inode(dentry);
+
+	if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
+		return -EIO;
+
+	return __f2fs_set_acl(idmap, inode, type, acl, NULL);
+}
+
+/*
+ * Most part of f2fs_acl_clone, f2fs_acl_create_masq, f2fs_acl_create
+ * are copied from posix_acl.c
+ */
+static struct posix_acl *f2fs_acl_clone(const struct posix_acl *acl,
+							gfp_t flags)
+{
+	struct posix_acl *clone = NULL;
+
+	if (acl) {
+		clone = kmemdup(acl, struct_size(acl, a_entries, acl->a_count),
+				flags);
+		if (clone)
+			refcount_set(&clone->a_refcount, 1);
+	}
+	return clone;
+}
+
+static int f2fs_acl_create_masq(struct posix_acl *acl, umode_t *mode_p)
+{
+	struct posix_acl_entry *pa, *pe;
+	struct posix_acl_entry *group_obj = NULL, *mask_obj = NULL;
+	umode_t mode = *mode_p;
+	int not_equiv = 0;
+
+	/* assert(atomic_read(acl->a_refcount) == 1); */
+
+	FOREACH_ACL_ENTRY(pa, acl, pe) {
+		switch (pa->e_tag) {
+		case ACL_USER_OBJ:
+			pa->e_perm &= (mode >> 6) | ~S_IRWXO;
+			mode &= (pa->e_perm << 6) | ~S_IRWXU;
+			break;
+
+		case ACL_USER:
+		case ACL_GROUP:
+			not_equiv = 1;
+			break;
+
+		case ACL_GROUP_OBJ:
+			group_obj = pa;
+			break;
+
+		case ACL_OTHER:
+			pa->e_perm &= mode | ~S_IRWXO;
+			mode &= pa->e_perm | ~S_IRWXO;
+			break;
+
+		case ACL_MASK:
+			mask_obj = pa;
+			not_equiv = 1;
+			break;
+
+		default:
+			return -EIO;
+		}
+	}
+
+	if (mask_obj) {
+		mask_obj->e_perm &= (mode >> 3) | ~S_IRWXO;
+		mode &= (mask_obj->e_perm << 3) | ~S_IRWXG;
+	} else {
+		if (!group_obj)
+			return -EIO;
+		group_obj->e_perm &= (mode >> 3) | ~S_IRWXO;
+		mode &= (group_obj->e_perm << 3) | ~S_IRWXG;
+	}
+
+	*mode_p = (*mode_p & ~S_IRWXUGO) | mode;
+	return not_equiv;
+}
+
+static int f2fs_acl_create(struct inode *dir, umode_t *mode,
+		struct posix_acl **default_acl, struct posix_acl **acl,
+		struct folio *dfolio)
+{
+	struct posix_acl *p;
+	struct posix_acl *clone;
+	int ret;
+
+	*acl = NULL;
+	*default_acl = NULL;
+
+	if (S_ISLNK(*mode) || !IS_POSIXACL(dir))
+		return 0;
+
+	p = __f2fs_get_acl(dir, ACL_TYPE_DEFAULT, dfolio);
+	if (!p || p == ERR_PTR(-EOPNOTSUPP)) {
+		*mode &= ~current_umask();
+		return 0;
+	}
+	if (IS_ERR(p))
+		return PTR_ERR(p);
+
+	clone = f2fs_acl_clone(p, GFP_NOFS);
+	if (!clone) {
+		ret = -ENOMEM;
+		goto release_acl;
+	}
+
+	ret = f2fs_acl_create_masq(clone, mode);
+	if (ret < 0)
+		goto release_clone;
+
+	if (ret == 0)
+		posix_acl_release(clone);
+	else
+		*acl = clone;
+
+	if (!S_ISDIR(*mode))
+		posix_acl_release(p);
+	else
+		*default_acl = p;
+
+	return 0;
+
+release_clone:
+	posix_acl_release(clone);
+release_acl:
+	posix_acl_release(p);
+	return ret;
+}
+
+int f2fs_init_acl(struct inode *inode, struct inode *dir, struct folio *ifolio,
+							struct folio *dfolio)
+{
+	struct posix_acl *default_acl = NULL, *acl = NULL;
+	int error;
+
+	error = f2fs_acl_create(dir, &inode->i_mode, &default_acl, &acl, dfolio);
+	if (error)
+		return error;
+
+	f2fs_mark_inode_dirty_sync(inode, true);
+
+	if (default_acl) {
+		error = __f2fs_set_acl(NULL, inode, ACL_TYPE_DEFAULT,
+				default_acl, ifolio);
+		posix_acl_release(default_acl);
+	} else {
+		inode->i_default_acl = NULL;
+	}
+	if (acl) {
+		if (!error)
+			error = __f2fs_set_acl(NULL, inode, ACL_TYPE_ACCESS,
+					acl, ifolio);
+		posix_acl_release(acl);
+	} else {
+		inode->i_acl = NULL;
+	}
+
+	return error;
+}