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, 173 insertions, 0 deletions

diff --git a/io_uring/eventfd.c b/io_uring/eventfd.c
new file mode 100644
index 000000000..63fe6e5d7
--- /dev/null
+++ b/io_uring/eventfd.c
@@ -0,0 +1,173 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/eventfd.h>
+#include <linux/eventpoll.h>
+#include <linux/io_uring.h>
+#include <linux/io_uring_types.h>
+
+#include "io-wq.h"
+#include "eventfd.h"
+
+struct io_ev_fd {
+	struct eventfd_ctx	*cq_ev_fd;
+	unsigned int		eventfd_async;
+	/* protected by ->completion_lock */
+	unsigned		last_cq_tail;
+	refcount_t		refs;
+	atomic_t		ops;
+	struct rcu_head		rcu;
+};
+
+enum {
+	IO_EVENTFD_OP_SIGNAL_BIT,
+};
+
+static void io_eventfd_free(struct rcu_head *rcu)
+{
+	struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu);
+
+	eventfd_ctx_put(ev_fd->cq_ev_fd);
+	kfree(ev_fd);
+}
+
+static void io_eventfd_put(struct io_ev_fd *ev_fd)
+{
+	if (refcount_dec_and_test(&ev_fd->refs))
+		call_rcu(&ev_fd->rcu, io_eventfd_free);
+}
+
+static void io_eventfd_do_signal(struct rcu_head *rcu)
+{
+	struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu);
+
+	atomic_andnot(BIT(IO_EVENTFD_OP_SIGNAL_BIT), &ev_fd->ops);
+	eventfd_signal_mask(ev_fd->cq_ev_fd, EPOLL_URING_WAKE);
+	io_eventfd_put(ev_fd);
+}
+
+/*
+ * Returns true if the caller should put the ev_fd reference, false if not.
+ */
+static bool __io_eventfd_signal(struct io_ev_fd *ev_fd, bool defer)
+{
+	if (!defer && eventfd_signal_allowed()) {
+		eventfd_signal_mask(ev_fd->cq_ev_fd, EPOLL_URING_WAKE);
+		return true;
+	}
+	if (!atomic_fetch_or(BIT(IO_EVENTFD_OP_SIGNAL_BIT), &ev_fd->ops)) {
+		call_rcu_hurry(&ev_fd->rcu, io_eventfd_do_signal);
+		return false;
+	}
+	return true;
+}
+
+/*
+ * Trigger if eventfd_async isn't set, or if it's set and the caller is
+ * an async worker.
+ */
+static bool io_eventfd_trigger(struct io_ev_fd *ev_fd)
+{
+	return !ev_fd->eventfd_async || io_wq_current_is_worker();
+}
+
+void io_eventfd_signal(struct io_ring_ctx *ctx, bool cqe_event, bool defer)
+{
+	bool skip = false;
+	struct io_ev_fd *ev_fd;
+	struct io_rings *rings;
+
+	guard(rcu)();
+
+	rings = rcu_dereference(ctx->rings_rcu);
+	if (!rings)
+		return;
+	if (READ_ONCE(rings->cq_flags) & IORING_CQ_EVENTFD_DISABLED)
+		return;
+	ev_fd = rcu_dereference(ctx->io_ev_fd);
+	/*
+	 * Check again if ev_fd exists in case an io_eventfd_unregister call
+	 * completed between the NULL check of ctx->io_ev_fd at the start of
+	 * the function and rcu_read_lock.
+	 */
+	if (!ev_fd)
+		return;
+	if (!io_eventfd_trigger(ev_fd) || !refcount_inc_not_zero(&ev_fd->refs))
+		return;
+
+	if (cqe_event) {
+		/*
+		 * Eventfd should only get triggered when at least one event
+		 * has been posted. Some applications rely on the eventfd
+		 * notification count only changing IFF a new CQE has been
+		 * added to the CQ ring. There's no dependency on 1:1
+		 * relationship between how many times this function is called
+		 * (and hence the eventfd count) and number of CQEs posted to
+		 * the CQ ring.
+		 */
+		spin_lock(&ctx->completion_lock);
+		skip = ctx->cached_cq_tail == ev_fd->last_cq_tail;
+		ev_fd->last_cq_tail = ctx->cached_cq_tail;
+		spin_unlock(&ctx->completion_lock);
+	}
+
+	if (skip || __io_eventfd_signal(ev_fd, defer))
+		io_eventfd_put(ev_fd);
+}
+
+int io_eventfd_register(struct io_ring_ctx *ctx, void __user *arg,
+			unsigned int eventfd_async)
+{
+	struct io_ev_fd *ev_fd;
+	__s32 __user *fds = arg;
+	int fd;
+
+	ev_fd = rcu_dereference_protected(ctx->io_ev_fd,
+					lockdep_is_held(&ctx->uring_lock));
+	if (ev_fd)
+		return -EBUSY;
+
+	if (copy_from_user(&fd, fds, sizeof(*fds)))
+		return -EFAULT;
+
+	ev_fd = kmalloc_obj(*ev_fd);
+	if (!ev_fd)
+		return -ENOMEM;
+
+	ev_fd->cq_ev_fd = eventfd_ctx_fdget(fd);
+	if (IS_ERR(ev_fd->cq_ev_fd)) {
+		int ret = PTR_ERR(ev_fd->cq_ev_fd);
+
+		kfree(ev_fd);
+		return ret;
+	}
+
+	spin_lock(&ctx->completion_lock);
+	ev_fd->last_cq_tail = ctx->cached_cq_tail;
+	spin_unlock(&ctx->completion_lock);
+
+	ev_fd->eventfd_async = eventfd_async;
+	ctx->int_flags |= IO_RING_F_HAS_EVFD;
+	refcount_set(&ev_fd->refs, 1);
+	atomic_set(&ev_fd->ops, 0);
+	rcu_assign_pointer(ctx->io_ev_fd, ev_fd);
+	return 0;
+}
+
+int io_eventfd_unregister(struct io_ring_ctx *ctx)
+{
+	struct io_ev_fd *ev_fd;
+
+	ev_fd = rcu_dereference_protected(ctx->io_ev_fd,
+					lockdep_is_held(&ctx->uring_lock));
+	if (ev_fd) {
+		ctx->int_flags &= ~IO_RING_F_HAS_EVFD;
+		rcu_assign_pointer(ctx->io_ev_fd, NULL);
+		io_eventfd_put(ev_fd);
+		return 0;
+	}
+
+	return -ENXIO;
+}