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, 472 insertions, 0 deletions

diff --git a/fs/afs/volume.c b/fs/afs/volume.c
new file mode 100644
index 000000000..4f79d25ec
--- /dev/null
+++ b/fs/afs/volume.c
@@ -0,0 +1,472 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* AFS volume management
+ *
+ * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include "internal.h"
+
+static unsigned __read_mostly afs_volume_record_life = 60 * 60;
+static atomic_t afs_volume_debug_id;
+
+static void afs_destroy_volume(struct work_struct *work);
+
+/*
+ * Insert a volume into a cell.  If there's an existing volume record, that is
+ * returned instead with a ref held.
+ */
+static struct afs_volume *afs_insert_volume_into_cell(struct afs_cell *cell,
+						      struct afs_volume *volume)
+{
+	struct afs_volume *p;
+	struct rb_node *parent = NULL, **pp;
+
+	write_seqlock(&cell->volume_lock);
+
+	pp = &cell->volumes.rb_node;
+	while (*pp) {
+		parent = *pp;
+		p = rb_entry(parent, struct afs_volume, cell_node);
+		if (p->vid < volume->vid) {
+			pp = &(*pp)->rb_left;
+		} else if (p->vid > volume->vid) {
+			pp = &(*pp)->rb_right;
+		} else {
+			if (afs_try_get_volume(p, afs_volume_trace_get_cell_insert)) {
+				volume = p;
+				goto found;
+			}
+
+			set_bit(AFS_VOLUME_RM_TREE, &p->flags);
+			rb_replace_node_rcu(&p->cell_node, &volume->cell_node, &cell->volumes);
+		}
+	}
+
+	rb_link_node_rcu(&volume->cell_node, parent, pp);
+	rb_insert_color(&volume->cell_node, &cell->volumes);
+	hlist_add_head_rcu(&volume->proc_link, &cell->proc_volumes);
+
+found:
+	write_sequnlock(&cell->volume_lock);
+	return volume;
+
+}
+
+static void afs_remove_volume_from_cell(struct afs_volume *volume)
+{
+	struct afs_cell *cell = volume->cell;
+
+	if (!hlist_unhashed(&volume->proc_link)) {
+		trace_afs_volume(volume->debug_id, volume->vid, refcount_read(&volume->ref),
+				 afs_volume_trace_remove);
+		write_seqlock(&cell->volume_lock);
+		hlist_del_rcu(&volume->proc_link);
+		if (!test_and_set_bit(AFS_VOLUME_RM_TREE, &volume->flags))
+			rb_erase(&volume->cell_node, &cell->volumes);
+		write_sequnlock(&cell->volume_lock);
+	}
+}
+
+/*
+ * Allocate a volume record and load it up from a vldb record.
+ */
+static struct afs_volume *afs_alloc_volume(struct afs_fs_context *params,
+					   struct afs_vldb_entry *vldb,
+					   struct afs_server_list **_slist)
+{
+	struct afs_server_list *slist;
+	struct afs_volume *volume;
+	int ret = -ENOMEM, i;
+
+	volume = kzalloc_obj(struct afs_volume);
+	if (!volume)
+		goto error_0;
+
+	volume->debug_id	= atomic_inc_return(&afs_volume_debug_id);
+	volume->vid		= vldb->vid[params->type];
+	volume->update_at	= ktime_get_real_seconds() + afs_volume_record_life;
+	volume->cell		= afs_get_cell(params->cell, afs_cell_trace_get_vol);
+	volume->type		= params->type;
+	volume->type_force	= params->force;
+	volume->name_len	= vldb->name_len;
+	volume->creation_time	= TIME64_MIN;
+	volume->update_time	= TIME64_MIN;
+
+	refcount_set(&volume->ref, 1);
+	INIT_HLIST_NODE(&volume->proc_link);
+	INIT_WORK(&volume->destructor, afs_destroy_volume);
+	rwlock_init(&volume->servers_lock);
+	mutex_init(&volume->volsync_lock);
+	mutex_init(&volume->cb_check_lock);
+	rwlock_init(&volume->cb_v_break_lock);
+	INIT_LIST_HEAD(&volume->open_mmaps);
+	init_rwsem(&volume->open_mmaps_lock);
+	memcpy(volume->name, vldb->name, vldb->name_len + 1);
+
+	for (i = 0; i < AFS_MAXTYPES; i++)
+		volume->vids[i] = vldb->vid[i];
+
+	slist = afs_alloc_server_list(volume, params->key, vldb);
+	if (IS_ERR(slist)) {
+		ret = PTR_ERR(slist);
+		goto error_1;
+	}
+
+	*_slist = slist;
+	rcu_assign_pointer(volume->servers, slist);
+	trace_afs_volume(volume->debug_id, volume->vid, 1, afs_volume_trace_alloc);
+	return volume;
+
+error_1:
+	afs_put_cell(volume->cell, afs_cell_trace_put_vol);
+	kfree(volume);
+error_0:
+	return ERR_PTR(ret);
+}
+
+/*
+ * Look up or allocate a volume record.
+ */
+static struct afs_volume *afs_lookup_volume(struct afs_fs_context *params,
+					    struct afs_vldb_entry *vldb)
+{
+	struct afs_server_list *slist;
+	struct afs_volume *candidate, *volume;
+
+	candidate = afs_alloc_volume(params, vldb, &slist);
+	if (IS_ERR(candidate))
+		return candidate;
+
+	volume = afs_insert_volume_into_cell(params->cell, candidate);
+	if (volume == candidate)
+		afs_attach_volume_to_servers(volume, slist);
+	else
+		afs_put_volume(candidate, afs_volume_trace_put_cell_dup);
+	return volume;
+}
+
+/*
+ * Look up a VLDB record for a volume.
+ */
+static struct afs_vldb_entry *afs_vl_lookup_vldb(struct afs_cell *cell,
+						 struct key *key,
+						 const char *volname,
+						 size_t volnamesz)
+{
+	struct afs_vldb_entry *vldb = ERR_PTR(-EDESTADDRREQ);
+	struct afs_vl_cursor vc;
+	int ret;
+
+	if (!afs_begin_vlserver_operation(&vc, cell, key))
+		return ERR_PTR(-ERESTARTSYS);
+
+	while (afs_select_vlserver(&vc)) {
+		vldb = afs_vl_get_entry_by_name_u(&vc, volname, volnamesz);
+	}
+
+	ret = afs_end_vlserver_operation(&vc);
+	return ret < 0 ? ERR_PTR(ret) : vldb;
+}
+
+/*
+ * Look up a volume in the VL server and create a candidate volume record for
+ * it.
+ *
+ * The volume name can be one of the following:
+ *	"%[cell:]volume[.]"		R/W volume
+ *	"#[cell:]volume[.]"		R/O or R/W volume (rwparent=0),
+ *					 or R/W (rwparent=1) volume
+ *	"%[cell:]volume.readonly"	R/O volume
+ *	"#[cell:]volume.readonly"	R/O volume
+ *	"%[cell:]volume.backup"		Backup volume
+ *	"#[cell:]volume.backup"		Backup volume
+ *
+ * The cell name is optional, and defaults to the current cell.
+ *
+ * See "The Rules of Mount Point Traversal" in Chapter 5 of the AFS SysAdmin
+ * Guide
+ * - Rule 1: Explicit type suffix forces access of that type or nothing
+ *           (no suffix, then use Rule 2 & 3)
+ * - Rule 2: If parent volume is R/O, then mount R/O volume by preference, R/W
+ *           if not available
+ * - Rule 3: If parent volume is R/W, then only mount R/W volume unless
+ *           explicitly told otherwise
+ */
+struct afs_volume *afs_create_volume(struct afs_fs_context *params)
+{
+	struct afs_vldb_entry *vldb;
+	struct afs_volume *volume;
+	unsigned long type_mask = 1UL << params->type;
+
+	vldb = afs_vl_lookup_vldb(params->cell, params->key,
+				  params->volname, params->volnamesz);
+	if (IS_ERR(vldb))
+		return ERR_CAST(vldb);
+
+	if (test_bit(AFS_VLDB_QUERY_ERROR, &vldb->flags)) {
+		volume = ERR_PTR(vldb->error);
+		goto error;
+	}
+
+	/* Make the final decision on the type we want */
+	volume = ERR_PTR(-ENOMEDIUM);
+	if (params->force) {
+		if (!(vldb->flags & type_mask))
+			goto error;
+	} else if (test_bit(AFS_VLDB_HAS_RO, &vldb->flags)) {
+		params->type = AFSVL_ROVOL;
+	} else if (test_bit(AFS_VLDB_HAS_RW, &vldb->flags)) {
+		params->type = AFSVL_RWVOL;
+	} else {
+		goto error;
+	}
+
+	volume = afs_lookup_volume(params, vldb);
+
+error:
+	kfree(vldb);
+	return volume;
+}
+
+/*
+ * Destroy a volume record
+ */
+static void afs_destroy_volume(struct work_struct *work)
+{
+	struct afs_volume *volume = container_of(work, struct afs_volume, destructor);
+	struct afs_server_list *slist = rcu_access_pointer(volume->servers);
+
+	_enter("%p", volume);
+
+#ifdef CONFIG_AFS_FSCACHE
+	ASSERTCMP(volume->cache, ==, NULL);
+#endif
+
+	afs_detach_volume_from_servers(volume, slist);
+	afs_remove_volume_from_cell(volume);
+	afs_put_serverlist(volume->cell->net, slist);
+	afs_put_cell(volume->cell, afs_cell_trace_put_vol);
+	trace_afs_volume(volume->debug_id, volume->vid, refcount_read(&volume->ref),
+			 afs_volume_trace_free);
+	kfree_rcu(volume, rcu);
+
+	_leave(" [destroyed]");
+}
+
+/*
+ * Try to get a reference on a volume record.
+ */
+bool afs_try_get_volume(struct afs_volume *volume, enum afs_volume_trace reason)
+{
+	int r;
+
+	if (__refcount_inc_not_zero(&volume->ref, &r)) {
+		trace_afs_volume(volume->debug_id, volume->vid, r + 1, reason);
+		return true;
+	}
+	return false;
+}
+
+/*
+ * Get a reference on a volume record.
+ */
+struct afs_volume *afs_get_volume(struct afs_volume *volume,
+				  enum afs_volume_trace reason)
+{
+	if (volume) {
+		int r;
+
+		__refcount_inc(&volume->ref, &r);
+		trace_afs_volume(volume->debug_id, volume->vid, r + 1, reason);
+	}
+	return volume;
+}
+
+
+/*
+ * Drop a reference on a volume record.
+ */
+void afs_put_volume(struct afs_volume *volume, enum afs_volume_trace reason)
+{
+	if (volume) {
+		unsigned int debug_id = volume->debug_id;
+		afs_volid_t vid = volume->vid;
+		bool zero;
+		int r;
+
+		zero = __refcount_dec_and_test(&volume->ref, &r);
+		trace_afs_volume(debug_id, vid, r - 1, reason);
+		if (zero)
+			schedule_work(&volume->destructor);
+	}
+}
+
+/*
+ * Activate a volume.
+ */
+int afs_activate_volume(struct afs_volume *volume)
+{
+#ifdef CONFIG_AFS_FSCACHE
+	struct fscache_volume *vcookie;
+	char *name;
+
+	name = kasprintf(GFP_KERNEL, "afs,%s,%llx",
+			 volume->cell->name, volume->vid);
+	if (!name)
+		return -ENOMEM;
+
+	vcookie = fscache_acquire_volume(name, NULL, NULL, 0);
+	if (IS_ERR(vcookie)) {
+		if (vcookie != ERR_PTR(-EBUSY)) {
+			kfree(name);
+			return PTR_ERR(vcookie);
+		}
+		pr_err("AFS: Cache volume key already in use (%s)\n", name);
+		vcookie = NULL;
+	}
+	volume->cache = vcookie;
+	kfree(name);
+#endif
+	return 0;
+}
+
+/*
+ * Deactivate a volume.
+ */
+void afs_deactivate_volume(struct afs_volume *volume)
+{
+	_enter("%s", volume->name);
+
+#ifdef CONFIG_AFS_FSCACHE
+	fscache_relinquish_volume(volume->cache, NULL,
+				  test_bit(AFS_VOLUME_DELETED, &volume->flags));
+	volume->cache = NULL;
+#endif
+
+	_leave("");
+}
+
+/*
+ * Query the VL service to update the volume status.
+ */
+static int afs_update_volume_status(struct afs_volume *volume, struct key *key)
+{
+	struct afs_server_list *new, *old, *discard;
+	struct afs_vldb_entry *vldb;
+	char idbuf[24];
+	int ret, idsz;
+
+	_enter("");
+
+	/* We look up an ID by passing it as a decimal string in the
+	 * operation's name parameter.
+	 */
+	idsz = snprintf(idbuf, sizeof(idbuf), "%llu", volume->vid);
+
+	vldb = afs_vl_lookup_vldb(volume->cell, key, idbuf, idsz);
+	if (IS_ERR(vldb)) {
+		ret = PTR_ERR(vldb);
+		goto error;
+	}
+
+	/* See if the volume got renamed. */
+	if (vldb->name_len != volume->name_len ||
+	    memcmp(vldb->name, volume->name, vldb->name_len) != 0) {
+		/* TODO: Use RCU'd string. */
+		memcpy(volume->name, vldb->name, AFS_MAXVOLNAME);
+		volume->name_len = vldb->name_len;
+	}
+
+	/* See if the volume's server list got updated. */
+	new = afs_alloc_server_list(volume, key, vldb);
+	if (IS_ERR(new)) {
+		ret = PTR_ERR(new);
+		goto error_vldb;
+	}
+
+	write_lock(&volume->servers_lock);
+
+	discard = new;
+	old = rcu_dereference_protected(volume->servers,
+					lockdep_is_held(&volume->servers_lock));
+	if (afs_annotate_server_list(new, old)) {
+		new->seq = volume->servers_seq + 1;
+		rcu_assign_pointer(volume->servers, new);
+		smp_wmb();
+		volume->servers_seq++;
+		discard = old;
+	}
+
+	/* Check more often if replication is ongoing. */
+	if (new->ro_replicating)
+		volume->update_at = ktime_get_real_seconds() + 10 * 60;
+	else
+		volume->update_at = ktime_get_real_seconds() + afs_volume_record_life;
+	write_unlock(&volume->servers_lock);
+
+	if (discard == old)
+		afs_reattach_volume_to_servers(volume, new, old);
+	afs_put_serverlist(volume->cell->net, discard);
+	ret = 0;
+error_vldb:
+	kfree(vldb);
+error:
+	_leave(" = %d", ret);
+	return ret;
+}
+
+/*
+ * Make sure the volume record is up to date.
+ */
+int afs_check_volume_status(struct afs_volume *volume, struct afs_operation *op)
+{
+	int ret, retries = 0;
+
+	_enter("");
+
+retry:
+	if (test_bit(AFS_VOLUME_WAIT, &volume->flags))
+		goto wait;
+	if (volume->update_at <= ktime_get_real_seconds() ||
+	    test_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags))
+		goto update;
+	_leave(" = 0");
+	return 0;
+
+update:
+	if (!test_and_set_bit_lock(AFS_VOLUME_UPDATING, &volume->flags)) {
+		clear_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags);
+		ret = afs_update_volume_status(volume, op->key);
+		if (ret < 0)
+			set_bit(AFS_VOLUME_NEEDS_UPDATE, &volume->flags);
+		clear_bit_unlock(AFS_VOLUME_WAIT, &volume->flags);
+		clear_bit_unlock(AFS_VOLUME_UPDATING, &volume->flags);
+		wake_up_bit(&volume->flags, AFS_VOLUME_WAIT);
+		_leave(" = %d", ret);
+		return ret;
+	}
+
+wait:
+	if (!test_bit(AFS_VOLUME_WAIT, &volume->flags)) {
+		_leave(" = 0 [no wait]");
+		return 0;
+	}
+
+	ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT,
+			  (op->flags & AFS_OPERATION_UNINTR) ?
+			  TASK_UNINTERRUPTIBLE : TASK_INTERRUPTIBLE);
+	if (ret == -ERESTARTSYS) {
+		_leave(" = %d", ret);
+		return ret;
+	}
+
+	retries++;
+	if (retries == 4) {
+		_leave(" = -ESTALE");
+		return -ESTALE;
+	}
+	goto retry;
+}