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, 172 insertions, 0 deletions

diff --git a/drivers/dma/ioat/sysfs.c b/drivers/dma/ioat/sysfs.c
new file mode 100644
index 000000000..e796ddb53
--- /dev/null
+++ b/drivers/dma/ioat/sysfs.c
@@ -0,0 +1,172 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Intel I/OAT DMA Linux driver
+ * Copyright(c) 2004 - 2015 Intel Corporation.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/dmaengine.h>
+#include <linux/pci.h>
+#include "dma.h"
+#include "registers.h"
+#include "hw.h"
+
+#include "../dmaengine.h"
+
+struct ioat_sysfs_entry {
+	struct attribute attr;
+	ssize_t (*show)(struct dma_chan *, char *);
+	ssize_t (*store)(struct dma_chan *, const char *, size_t);
+};
+
+static ssize_t cap_show(struct dma_chan *c, char *page)
+{
+	struct dma_device *dma = c->device;
+
+	return sprintf(page, "copy%s%s%s%s%s\n",
+		       dma_has_cap(DMA_PQ, dma->cap_mask) ? " pq" : "",
+		       dma_has_cap(DMA_PQ_VAL, dma->cap_mask) ? " pq_val" : "",
+		       dma_has_cap(DMA_XOR, dma->cap_mask) ? " xor" : "",
+		       dma_has_cap(DMA_XOR_VAL, dma->cap_mask) ? " xor_val" : "",
+		       dma_has_cap(DMA_INTERRUPT, dma->cap_mask) ? " intr" : "");
+
+}
+static const struct ioat_sysfs_entry ioat_cap_attr = __ATTR_RO(cap);
+
+static ssize_t version_show(struct dma_chan *c, char *page)
+{
+	struct dma_device *dma = c->device;
+	struct ioatdma_device *ioat_dma = to_ioatdma_device(dma);
+
+	return sprintf(page, "%d.%d\n",
+		       ioat_dma->version >> 4, ioat_dma->version & 0xf);
+}
+static const struct ioat_sysfs_entry ioat_version_attr = __ATTR_RO(version);
+
+static ssize_t
+ioat_attr_show(struct kobject *kobj, struct attribute *attr, char *page)
+{
+	const struct ioat_sysfs_entry *entry;
+	struct ioatdma_chan *ioat_chan;
+
+	entry = container_of_const(attr, struct ioat_sysfs_entry, attr);
+	ioat_chan = container_of(kobj, struct ioatdma_chan, kobj);
+
+	if (!entry->show)
+		return -EIO;
+	return entry->show(&ioat_chan->dma_chan, page);
+}
+
+static ssize_t
+ioat_attr_store(struct kobject *kobj, struct attribute *attr,
+const char *page, size_t count)
+{
+	const struct ioat_sysfs_entry *entry;
+	struct ioatdma_chan *ioat_chan;
+
+	entry = container_of_const(attr, struct ioat_sysfs_entry, attr);
+	ioat_chan = container_of(kobj, struct ioatdma_chan, kobj);
+
+	if (!entry->store)
+		return -EIO;
+	return entry->store(&ioat_chan->dma_chan, page, count);
+}
+
+static const struct sysfs_ops ioat_sysfs_ops = {
+	.show	= ioat_attr_show,
+	.store  = ioat_attr_store,
+};
+
+void ioat_kobject_add(struct ioatdma_device *ioat_dma, const struct kobj_type *type)
+{
+	struct dma_device *dma = &ioat_dma->dma_dev;
+	struct dma_chan *c;
+
+	list_for_each_entry(c, &dma->channels, device_node) {
+		struct ioatdma_chan *ioat_chan = to_ioat_chan(c);
+		struct kobject *parent = &c->dev->device.kobj;
+		int err;
+
+		err = kobject_init_and_add(&ioat_chan->kobj, type,
+					   parent, "quickdata");
+		if (err) {
+			dev_warn(to_dev(ioat_chan),
+				 "sysfs init error (%d), continuing...\n", err);
+			kobject_put(&ioat_chan->kobj);
+			set_bit(IOAT_KOBJ_INIT_FAIL, &ioat_chan->state);
+		}
+	}
+}
+
+void ioat_kobject_del(struct ioatdma_device *ioat_dma)
+{
+	struct dma_device *dma = &ioat_dma->dma_dev;
+	struct dma_chan *c;
+
+	list_for_each_entry(c, &dma->channels, device_node) {
+		struct ioatdma_chan *ioat_chan = to_ioat_chan(c);
+
+		if (!test_bit(IOAT_KOBJ_INIT_FAIL, &ioat_chan->state)) {
+			kobject_del(&ioat_chan->kobj);
+			kobject_put(&ioat_chan->kobj);
+		}
+	}
+}
+
+static ssize_t ring_size_show(struct dma_chan *c, char *page)
+{
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(c);
+
+	return sprintf(page, "%d\n", (1 << ioat_chan->alloc_order) & ~1);
+}
+static const struct ioat_sysfs_entry ring_size_attr = __ATTR_RO(ring_size);
+
+static ssize_t ring_active_show(struct dma_chan *c, char *page)
+{
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(c);
+
+	/* ...taken outside the lock, no need to be precise */
+	return sprintf(page, "%d\n", ioat_ring_active(ioat_chan));
+}
+static const struct ioat_sysfs_entry ring_active_attr = __ATTR_RO(ring_active);
+
+static ssize_t intr_coalesce_show(struct dma_chan *c, char *page)
+{
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(c);
+
+	return sprintf(page, "%d\n", ioat_chan->intr_coalesce);
+}
+
+static ssize_t intr_coalesce_store(struct dma_chan *c, const char *page,
+size_t count)
+{
+	int intr_coalesce = 0;
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(c);
+
+	if (sscanf(page, "%du", &intr_coalesce) != -1) {
+		if ((intr_coalesce < 0) ||
+		    (intr_coalesce > IOAT_INTRDELAY_MASK))
+			return -EINVAL;
+		ioat_chan->intr_coalesce = intr_coalesce;
+	}
+
+	return count;
+}
+
+static const struct ioat_sysfs_entry intr_coalesce_attr = __ATTR_RW(intr_coalesce);
+
+static const struct attribute *const ioat_attrs[] = {
+	&ring_size_attr.attr,
+	&ring_active_attr.attr,
+	&ioat_cap_attr.attr,
+	&ioat_version_attr.attr,
+	&intr_coalesce_attr.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(ioat);
+
+const struct kobj_type ioat_ktype = {
+	.sysfs_ops = &ioat_sysfs_ops,
+	.default_groups = ioat_groups,
+};