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, 280 insertions, 0 deletions

diff --git a/net/sctp/stream_sched.c b/net/sctp/stream_sched.c
new file mode 100644
index 000000000..50f8b5240
--- /dev/null
+++ b/net/sctp/stream_sched.c
@@ -0,0 +1,280 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* SCTP kernel implementation
+ * (C) Copyright Red Hat Inc. 2017
+ *
+ * This file is part of the SCTP kernel implementation
+ *
+ * These functions manipulate sctp stream queue/scheduling.
+ *
+ * Please send any bug reports or fixes you make to the
+ * email addresched(es):
+ *    lksctp developers <linux-sctp@vger.kernel.org>
+ *
+ * Written or modified by:
+ *    Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
+ */
+
+#include <linux/list.h>
+#include <net/sctp/sctp.h>
+#include <net/sctp/sm.h>
+#include <net/sctp/stream_sched.h>
+
+/* First Come First Serve (a.k.a. FIFO)
+ * RFC DRAFT ndata Section 3.1
+ */
+static int sctp_sched_fcfs_set(struct sctp_stream *stream, __u16 sid,
+			       __u16 value, gfp_t gfp)
+{
+	return 0;
+}
+
+static int sctp_sched_fcfs_get(struct sctp_stream *stream, __u16 sid,
+			       __u16 *value)
+{
+	*value = 0;
+	return 0;
+}
+
+static int sctp_sched_fcfs_init(struct sctp_stream *stream)
+{
+	return 0;
+}
+
+static int sctp_sched_fcfs_init_sid(struct sctp_stream *stream, __u16 sid,
+				    gfp_t gfp)
+{
+	return 0;
+}
+
+static void sctp_sched_fcfs_free_sid(struct sctp_stream *stream, __u16 sid)
+{
+}
+
+static void sctp_sched_fcfs_enqueue(struct sctp_outq *q,
+				    struct sctp_datamsg *msg)
+{
+}
+
+static struct sctp_chunk *sctp_sched_fcfs_dequeue(struct sctp_outq *q)
+{
+	struct sctp_stream *stream = &q->asoc->stream;
+	struct sctp_chunk *ch = NULL;
+	struct list_head *entry;
+
+	if (list_empty(&q->out_chunk_list))
+		goto out;
+
+	if (stream->out_curr) {
+		ch = list_entry(stream->out_curr->ext->outq.next,
+				struct sctp_chunk, stream_list);
+	} else {
+		entry = q->out_chunk_list.next;
+		ch = list_entry(entry, struct sctp_chunk, list);
+	}
+
+	sctp_sched_dequeue_common(q, ch);
+
+out:
+	return ch;
+}
+
+static void sctp_sched_fcfs_dequeue_done(struct sctp_outq *q,
+					 struct sctp_chunk *chunk)
+{
+}
+
+static void sctp_sched_fcfs_sched_all(struct sctp_stream *stream)
+{
+}
+
+static void sctp_sched_fcfs_unsched_all(struct sctp_stream *stream)
+{
+}
+
+static const struct sctp_sched_ops sctp_sched_fcfs = {
+	.set = sctp_sched_fcfs_set,
+	.get = sctp_sched_fcfs_get,
+	.init = sctp_sched_fcfs_init,
+	.init_sid = sctp_sched_fcfs_init_sid,
+	.free_sid = sctp_sched_fcfs_free_sid,
+	.enqueue = sctp_sched_fcfs_enqueue,
+	.dequeue = sctp_sched_fcfs_dequeue,
+	.dequeue_done = sctp_sched_fcfs_dequeue_done,
+	.sched_all = sctp_sched_fcfs_sched_all,
+	.unsched_all = sctp_sched_fcfs_unsched_all,
+};
+
+static void sctp_sched_ops_fcfs_init(void)
+{
+	sctp_sched_ops_register(SCTP_SS_FCFS, &sctp_sched_fcfs);
+}
+
+/* API to other parts of the stack */
+
+static const struct sctp_sched_ops *sctp_sched_ops[SCTP_SS_MAX + 1];
+
+void sctp_sched_ops_register(enum sctp_sched_type sched,
+			     const struct sctp_sched_ops *sched_ops)
+{
+	sctp_sched_ops[sched] = sched_ops;
+}
+
+void sctp_sched_ops_init(void)
+{
+	sctp_sched_ops_fcfs_init();
+	sctp_sched_ops_prio_init();
+	sctp_sched_ops_rr_init();
+	sctp_sched_ops_fc_init();
+	sctp_sched_ops_wfq_init();
+}
+
+static void sctp_sched_free_sched(struct sctp_stream *stream)
+{
+	const struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream);
+	struct sctp_stream_out_ext *soute;
+	int i;
+
+	sched->unsched_all(stream);
+	for (i = 0; i < stream->outcnt; i++) {
+		soute = SCTP_SO(stream, i)->ext;
+		if (!soute)
+			continue;
+		sched->free_sid(stream, i);
+		/* Give the next scheduler a clean slate. */
+		memset_after(soute, 0, outq);
+	}
+}
+
+int sctp_sched_set_sched(struct sctp_association *asoc,
+			 enum sctp_sched_type sched)
+{
+	const struct sctp_sched_ops *old = asoc->outqueue.sched;
+	struct sctp_datamsg *msg = NULL;
+	const struct sctp_sched_ops *n;
+	struct sctp_chunk *ch;
+	int i, ret = 0;
+
+	if (sched > SCTP_SS_MAX)
+		return -EINVAL;
+
+	n = sctp_sched_ops[sched];
+	if (old == n)
+		return ret;
+
+	if (old)
+		sctp_sched_free_sched(&asoc->stream);
+
+	asoc->outqueue.sched = n;
+	n->init(&asoc->stream);
+	for (i = 0; i < asoc->stream.outcnt; i++) {
+		if (!SCTP_SO(&asoc->stream, i)->ext)
+			continue;
+
+		ret = n->init_sid(&asoc->stream, i, GFP_ATOMIC);
+		if (ret)
+			goto err;
+	}
+
+	/* We have to requeue all chunks already queued. */
+	list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list) {
+		if (ch->msg == msg)
+			continue;
+		msg = ch->msg;
+		n->enqueue(&asoc->outqueue, msg);
+	}
+
+	return ret;
+
+err:
+	sctp_sched_free_sched(&asoc->stream);
+	asoc->outqueue.sched = &sctp_sched_fcfs; /* Always safe */
+
+	return ret;
+}
+
+int sctp_sched_get_sched(struct sctp_association *asoc)
+{
+	int i;
+
+	for (i = 0; i <= SCTP_SS_MAX; i++)
+		if (asoc->outqueue.sched == sctp_sched_ops[i])
+			return i;
+
+	return 0;
+}
+
+int sctp_sched_set_value(struct sctp_association *asoc, __u16 sid,
+			 __u16 value, gfp_t gfp)
+{
+	if (sid >= asoc->stream.outcnt)
+		return -EINVAL;
+
+	if (!SCTP_SO(&asoc->stream, sid)->ext) {
+		int ret;
+
+		ret = sctp_stream_init_ext(&asoc->stream, sid);
+		if (ret)
+			return ret;
+	}
+
+	return asoc->outqueue.sched->set(&asoc->stream, sid, value, gfp);
+}
+
+int sctp_sched_get_value(struct sctp_association *asoc, __u16 sid,
+			 __u16 *value)
+{
+	if (sid >= asoc->stream.outcnt)
+		return -EINVAL;
+
+	if (!SCTP_SO(&asoc->stream, sid)->ext)
+		return 0;
+
+	return asoc->outqueue.sched->get(&asoc->stream, sid, value);
+}
+
+void sctp_sched_dequeue_done(struct sctp_outq *q, struct sctp_chunk *ch)
+{
+	if (!list_is_last(&ch->frag_list, &ch->msg->chunks) &&
+	    !q->asoc->peer.intl_capable) {
+		struct sctp_stream_out *sout;
+		__u16 sid;
+
+		/* datamsg is not finish, so save it as current one,
+		 * in case application switch scheduler or a higher
+		 * priority stream comes in.
+		 */
+		sid = sctp_chunk_stream_no(ch);
+		sout = SCTP_SO(&q->asoc->stream, sid);
+		q->asoc->stream.out_curr = sout;
+		return;
+	}
+
+	q->asoc->stream.out_curr = NULL;
+	q->sched->dequeue_done(q, ch);
+}
+
+/* Auxiliary functions for the schedulers */
+void sctp_sched_dequeue_common(struct sctp_outq *q, struct sctp_chunk *ch)
+{
+	list_del_init(&ch->list);
+	list_del_init(&ch->stream_list);
+	q->out_qlen -= ch->skb->len;
+}
+
+int sctp_sched_init_sid(struct sctp_stream *stream, __u16 sid, gfp_t gfp)
+{
+	const struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream);
+	struct sctp_stream_out_ext *ext = SCTP_SO(stream, sid)->ext;
+
+	INIT_LIST_HEAD(&ext->outq);
+	return sched->init_sid(stream, sid, gfp);
+}
+
+const struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream)
+{
+	struct sctp_association *asoc;
+
+	asoc = container_of(stream, struct sctp_association, stream);
+
+	return asoc->outqueue.sched;
+}