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, 168 insertions, 0 deletions

diff --git a/lib/cpumask.c b/lib/cpumask.c
new file mode 100644
index 000000000..5adb9874f
--- /dev/null
+++ b/lib/cpumask.c
@@ -0,0 +1,168 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+#include <linux/cpumask.h>
+#include <linux/export.h>
+#include <linux/memblock.h>
+#include <linux/numa.h>
+
+/* These are not inline because of header tangles. */
+#ifdef CONFIG_CPUMASK_OFFSTACK
+/**
+ * alloc_cpumask_var_node - allocate a struct cpumask on a given node
+ * @mask: pointer to cpumask_var_t where the cpumask is returned
+ * @flags: GFP_ flags
+ * @node: memory node from which to allocate or %NUMA_NO_NODE
+ *
+ * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
+ * a nop returning a constant 1 (in <linux/cpumask.h>).
+ *
+ * Return: TRUE if memory allocation succeeded, FALSE otherwise.
+ *
+ * In addition, mask will be NULL if this fails.  Note that gcc is
+ * usually smart enough to know that mask can never be NULL if
+ * CONFIG_CPUMASK_OFFSTACK=n, so does code elimination in that case
+ * too.
+ */
+bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node)
+{
+	*mask = kmalloc_node(cpumask_size(), flags, node);
+
+#ifdef CONFIG_DEBUG_PER_CPU_MAPS
+	if (!*mask) {
+		printk(KERN_ERR "=> alloc_cpumask_var: failed!\n");
+		dump_stack();
+	}
+#endif
+
+	return *mask != NULL;
+}
+EXPORT_SYMBOL(alloc_cpumask_var_node);
+
+/**
+ * alloc_bootmem_cpumask_var - allocate a struct cpumask from the bootmem arena.
+ * @mask: pointer to cpumask_var_t where the cpumask is returned
+ *
+ * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
+ * a nop (in <linux/cpumask.h>).
+ * Either returns an allocated (zero-filled) cpumask, or causes the
+ * system to panic.
+ */
+void __init alloc_bootmem_cpumask_var(cpumask_var_t *mask)
+{
+	*mask = memblock_alloc_or_panic(cpumask_size(), SMP_CACHE_BYTES);
+}
+
+/**
+ * free_cpumask_var - frees memory allocated for a struct cpumask.
+ * @mask: cpumask to free
+ *
+ * This is safe on a NULL mask.
+ */
+void free_cpumask_var(cpumask_var_t mask)
+{
+	kfree(mask);
+}
+EXPORT_SYMBOL(free_cpumask_var);
+
+/**
+ * free_bootmem_cpumask_var - frees result of alloc_bootmem_cpumask_var
+ * @mask: cpumask to free
+ */
+void __init free_bootmem_cpumask_var(cpumask_var_t mask)
+{
+	memblock_free(mask, cpumask_size());
+}
+#endif
+
+/**
+ * cpumask_local_spread - select the i'th cpu based on NUMA distances
+ * @i: index number
+ * @node: local numa_node
+ *
+ * Return: online CPU according to a numa aware policy; local cpus are returned
+ * first, followed by non-local ones, then it wraps around.
+ *
+ * For those who wants to enumerate all CPUs based on their NUMA distances,
+ * i.e. call this function in a loop, like:
+ *
+ * for (i = 0; i < num_online_cpus(); i++) {
+ *	cpu = cpumask_local_spread(i, node);
+ *	do_something(cpu);
+ * }
+ *
+ * There's a better alternative based on for_each()-like iterators:
+ *
+ *	for_each_numa_hop_mask(mask, node) {
+ *		for_each_cpu_andnot(cpu, mask, prev)
+ *			do_something(cpu);
+ *		prev = mask;
+ *	}
+ *
+ * It's simpler and more verbose than above. Complexity of iterator-based
+ * enumeration is O(sched_domains_numa_levels * nr_cpu_ids), while
+ * cpumask_local_spread() when called for each cpu is
+ * O(sched_domains_numa_levels * nr_cpu_ids * log(nr_cpu_ids)).
+ */
+unsigned int cpumask_local_spread(unsigned int i, int node)
+{
+	unsigned int cpu;
+
+	/* Wrap: we always want a cpu. */
+	i %= num_online_cpus();
+
+	cpu = sched_numa_find_nth_cpu(cpu_online_mask, i, node);
+
+	WARN_ON(cpu >= nr_cpu_ids);
+	return cpu;
+}
+EXPORT_SYMBOL(cpumask_local_spread);
+
+static DEFINE_PER_CPU(int, distribute_cpu_mask_prev);
+
+/**
+ * cpumask_any_and_distribute - Return an arbitrary cpu within src1p & src2p.
+ * @src1p: first &cpumask for intersection
+ * @src2p: second &cpumask for intersection
+ *
+ * Iterated calls using the same srcp1 and srcp2 will be distributed within
+ * their intersection.
+ *
+ * Return: >= nr_cpu_ids if the intersection is empty.
+ */
+unsigned int cpumask_any_and_distribute(const struct cpumask *src1p,
+			       const struct cpumask *src2p)
+{
+	unsigned int next, prev;
+
+	/* NOTE: our first selection will skip 0. */
+	prev = __this_cpu_read(distribute_cpu_mask_prev);
+
+	next = cpumask_next_and_wrap(prev, src1p, src2p);
+	if (next < nr_cpu_ids)
+		__this_cpu_write(distribute_cpu_mask_prev, next);
+
+	return next;
+}
+EXPORT_SYMBOL(cpumask_any_and_distribute);
+
+/**
+ * cpumask_any_distribute - Return an arbitrary cpu from srcp
+ * @srcp: &cpumask for selection
+ *
+ * Return: >= nr_cpu_ids if the intersection is empty.
+ */
+unsigned int cpumask_any_distribute(const struct cpumask *srcp)
+{
+	unsigned int next, prev;
+
+	/* NOTE: our first selection will skip 0. */
+	prev = __this_cpu_read(distribute_cpu_mask_prev);
+	next = cpumask_next_wrap(prev, srcp);
+	if (next < nr_cpu_ids)
+		__this_cpu_write(distribute_cpu_mask_prev, next);
+
+	return next;
+}
+EXPORT_SYMBOL(cpumask_any_distribute);