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, 184 insertions, 0 deletions

diff --git a/drivers/of/of_numa.c b/drivers/of/of_numa.c
new file mode 100644
index 000000000..230d5f628
--- /dev/null
+++ b/drivers/of/of_numa.c
@@ -0,0 +1,184 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * OF NUMA Parsing support.
+ *
+ * Copyright (C) 2015 - 2016 Cavium Inc.
+ */
+
+#define pr_fmt(fmt) "OF: NUMA: " fmt
+
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/nodemask.h>
+#include <linux/numa_memblks.h>
+
+#include <asm/numa.h>
+
+/*
+ * Even though we connect cpus to numa domains later in SMP
+ * init, we need to know the node ids now for all cpus.
+*/
+static void __init of_numa_parse_cpu_nodes(void)
+{
+	u32 nid;
+	int r;
+	struct device_node *np;
+
+	for_each_of_cpu_node(np) {
+		r = of_property_read_u32(np, "numa-node-id", &nid);
+		if (r)
+			continue;
+
+		pr_debug("CPU on %u\n", nid);
+		if (nid >= MAX_NUMNODES)
+			pr_warn("Node id %u exceeds maximum value\n", nid);
+		else
+			node_set(nid, numa_nodes_parsed);
+	}
+}
+
+static int __init of_numa_parse_memory_nodes(void)
+{
+	struct device_node *np = NULL;
+	struct resource rsrc;
+	u32 nid;
+	int i, r = -EINVAL;
+
+	for_each_node_by_type(np, "memory") {
+		r = of_property_read_u32(np, "numa-node-id", &nid);
+		if (r == -EINVAL)
+			/*
+			 * property doesn't exist if -EINVAL, continue
+			 * looking for more memory nodes with
+			 * "numa-node-id" property
+			 */
+			continue;
+
+		if (nid >= MAX_NUMNODES) {
+			pr_warn("Node id %u exceeds maximum value\n", nid);
+			r = -EINVAL;
+		}
+
+		for (i = 0; !r && !of_address_to_resource(np, i, &rsrc); i++)
+			r = numa_add_memblk(nid, rsrc.start, rsrc.end + 1);
+
+		if (!i || r) {
+			of_node_put(np);
+			pr_err("bad property in memory node\n");
+			return r ? : -EINVAL;
+		}
+	}
+
+	return r;
+}
+
+static int __init of_numa_parse_distance_map_v1(struct device_node *map)
+{
+	const __be32 *matrix;
+	int entry_count;
+	int i;
+
+	pr_info("parsing numa-distance-map-v1\n");
+
+	matrix = of_get_property(map, "distance-matrix", NULL);
+	if (!matrix) {
+		pr_err("No distance-matrix property in distance-map\n");
+		return -EINVAL;
+	}
+
+	entry_count = of_property_count_u32_elems(map, "distance-matrix");
+	if (entry_count <= 0) {
+		pr_err("Invalid distance-matrix\n");
+		return -EINVAL;
+	}
+
+	for (i = 0; i + 2 < entry_count; i += 3) {
+		u32 nodea, nodeb, distance;
+
+		nodea = of_read_number(matrix, 1);
+		matrix++;
+		nodeb = of_read_number(matrix, 1);
+		matrix++;
+		distance = of_read_number(matrix, 1);
+		matrix++;
+
+		if ((nodea == nodeb && distance != LOCAL_DISTANCE) ||
+		    (nodea != nodeb && distance <= LOCAL_DISTANCE)) {
+			pr_err("Invalid distance[node%d -> node%d] = %d\n",
+			       nodea, nodeb, distance);
+			return -EINVAL;
+		}
+
+		node_set(nodea, numa_nodes_parsed);
+
+		numa_set_distance(nodea, nodeb, distance);
+
+		/* Set default distance of node B->A same as A->B */
+		if (nodeb > nodea)
+			numa_set_distance(nodeb, nodea, distance);
+	}
+
+	return 0;
+}
+
+static int __init of_numa_parse_distance_map(void)
+{
+	int ret = 0;
+	struct device_node *np;
+
+	np = of_find_compatible_node(NULL, NULL,
+				     "numa-distance-map-v1");
+	if (np)
+		ret = of_numa_parse_distance_map_v1(np);
+
+	of_node_put(np);
+	return ret;
+}
+
+int of_node_to_nid(struct device_node *device)
+{
+	struct device_node *np;
+	u32 nid;
+	int r = -ENODATA;
+
+	np = of_node_get(device);
+
+	while (np) {
+		r = of_property_read_u32(np, "numa-node-id", &nid);
+		/*
+		 * -EINVAL indicates the property was not found, and
+		 *  we walk up the tree trying to find a parent with a
+		 *  "numa-node-id".  Any other type of error indicates
+		 *  a bad device tree and we give up.
+		 */
+		if (r != -EINVAL)
+			break;
+
+		np = of_get_next_parent(np);
+	}
+	if (np && r)
+		pr_warn("Invalid \"numa-node-id\" property in node %pOFn\n",
+			np);
+	of_node_put(np);
+
+	/*
+	 * If numa=off passed on command line, or with a defective
+	 * device tree, the nid may not be in the set of possible
+	 * nodes.  Check for this case and return NUMA_NO_NODE.
+	 */
+	if (!r && nid < MAX_NUMNODES && node_possible(nid))
+		return nid;
+
+	return NUMA_NO_NODE;
+}
+
+int __init of_numa_init(void)
+{
+	int r;
+
+	of_numa_parse_cpu_nodes();
+	r = of_numa_parse_memory_nodes();
+	if (r)
+		return r;
+	return of_numa_parse_distance_map();
+}