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, 160 insertions, 0 deletions

diff --git a/drivers/gpu/host1x/intr.c b/drivers/gpu/host1x/intr.c
new file mode 100644
index 000000000..6a383e0c6
--- /dev/null
+++ b/drivers/gpu/host1x/intr.c
@@ -0,0 +1,160 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Tegra host1x Interrupt Management
+ *
+ * Copyright (c) 2010-2021, NVIDIA Corporation.
+ */
+
+#include <linux/clk.h>
+#include <linux/interrupt.h>
+#include <linux/overflow.h>
+#include "dev.h"
+#include "fence.h"
+#include "intr.h"
+
+static void host1x_intr_add_fence_to_list(struct host1x_fence_list *list,
+					  struct host1x_syncpt_fence *fence)
+{
+	struct host1x_syncpt_fence *fence_in_list;
+
+	list_for_each_entry_reverse(fence_in_list, &list->list, list) {
+		if ((s32)wrapping_sub(u32, fence_in_list->threshold, fence->threshold) <= 0) {
+			/* Fence in list is before us, we can insert here */
+			list_add(&fence->list, &fence_in_list->list);
+			return;
+		}
+	}
+
+	/* Add as first in list */
+	list_add(&fence->list, &list->list);
+}
+
+static void host1x_intr_update_hw_state(struct host1x *host, struct host1x_syncpt *sp)
+{
+	struct host1x_syncpt_fence *fence;
+
+	if (!list_empty(&sp->fences.list)) {
+		fence = list_first_entry(&sp->fences.list, struct host1x_syncpt_fence, list);
+
+		host1x_hw_intr_set_syncpt_threshold(host, sp->id, fence->threshold);
+		host1x_hw_intr_enable_syncpt_intr(host, sp->id);
+	} else {
+		host1x_hw_intr_disable_syncpt_intr(host, sp->id);
+	}
+}
+
+void host1x_intr_add_fence_locked(struct host1x *host, struct host1x_syncpt_fence *fence)
+{
+	struct host1x_fence_list *fence_list = &fence->sp->fences;
+
+	INIT_LIST_HEAD(&fence->list);
+
+	host1x_intr_add_fence_to_list(fence_list, fence);
+	host1x_intr_update_hw_state(host, fence->sp);
+}
+
+bool host1x_intr_remove_fence(struct host1x *host, struct host1x_syncpt_fence *fence)
+{
+	struct host1x_fence_list *fence_list = &fence->sp->fences;
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&fence_list->lock, irqflags);
+
+	if (list_empty(&fence->list)) {
+		spin_unlock_irqrestore(&fence_list->lock, irqflags);
+		return false;
+	}
+
+	list_del_init(&fence->list);
+	host1x_intr_update_hw_state(host, fence->sp);
+
+	spin_unlock_irqrestore(&fence_list->lock, irqflags);
+
+	return true;
+}
+
+void host1x_intr_handle_interrupt(struct host1x *host, unsigned int id)
+{
+	struct host1x_syncpt *sp = &host->syncpt[id];
+	struct host1x_syncpt_fence *fence, *tmp;
+	unsigned int value;
+
+	value = host1x_syncpt_load(sp);
+
+	spin_lock(&sp->fences.lock);
+
+	list_for_each_entry_safe(fence, tmp, &sp->fences.list, list) {
+		if ((wrapping_sub(u32, value, fence->threshold) & 0x80000000U) != 0U) {
+			/* Fence is not yet expired, we are done */
+			break;
+		}
+
+		list_del_init(&fence->list);
+		host1x_fence_signal(fence);
+	}
+
+	/*
+	 * Re-enable interrupt if necessary. The ISR already disabled the interrupt,
+	 * so if no fences remain, no update is needed.
+	 */
+	if (!list_empty(&sp->fences.list))
+		host1x_intr_update_hw_state(host, sp);
+
+	spin_unlock(&sp->fences.lock);
+}
+
+int host1x_intr_init(struct host1x *host)
+{
+	struct host1x_intr_irq_data *irq_data;
+	unsigned int id;
+	int i, err;
+
+	for (id = 0; id < host1x_syncpt_nb_pts(host); ++id) {
+		struct host1x_syncpt *syncpt = &host->syncpt[id];
+
+		spin_lock_init(&syncpt->fences.lock);
+		INIT_LIST_HEAD(&syncpt->fences.list);
+	}
+
+	irq_data = devm_kcalloc(host->dev, host->num_syncpt_irqs, sizeof(irq_data[0]), GFP_KERNEL);
+	if (!irq_data)
+		return -ENOMEM;
+
+	host1x_hw_intr_disable_all_syncpt_intrs(host);
+
+	for (i = 0; i < host->num_syncpt_irqs; i++) {
+		irq_data[i].host = host;
+		irq_data[i].offset = i;
+
+		err = devm_request_irq(host->dev, host->syncpt_irqs[i],
+				       host->intr_op->isr, IRQF_SHARED,
+				       "host1x_syncpt", &irq_data[i]);
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
+
+void host1x_intr_deinit(struct host1x *host)
+{
+}
+
+void host1x_intr_start(struct host1x *host)
+{
+	u32 hz = clk_get_rate(host->clk);
+	int err;
+
+	mutex_lock(&host->intr_mutex);
+	err = host1x_hw_intr_init_host_sync(host, DIV_ROUND_UP(hz, 1000000));
+	if (err) {
+		mutex_unlock(&host->intr_mutex);
+		return;
+	}
+	mutex_unlock(&host->intr_mutex);
+}
+
+void host1x_intr_stop(struct host1x *host)
+{
+	host1x_hw_intr_disable_all_syncpt_intrs(host);
+}