ARM: 9485/1: mm: acquire mmap write lock around show_pte() for user faultsgrafted

When CONFIG_DEBUG_USER=y, and cmdline "user_debug=31" is set,
a user fault may trigger show_pte() without any lock.
If another thread in the same process concurrently calls munmap(),
the page table pages may be freed while show_pte() is still traversing
them, causing a use-after-free in show_pte().

If CONFIG_ARM_LPAE=y, this may cause a kernel panic if the pages table
of PMD are freed when show_pte() is running.

Acquire mmap_write_lock() around show_pte() for user faults to fix the
contention.

For user faults, additionally restrict that show_pte() is called only
when the addr is a user-space address (addr < TASK_SIZE). This is because
the lock of tsk->mm only protects the virtual memory of user address space,
furthermore, dumping the page tables of a kernel-space address for user
faults is unnecessary and may have security implications.

Keep everything unchanged for kernel faults, because the kernel is
already in the "oops" state, acquiring a lock may risk a deadlock.

Co-developed-by: Qi Xi <xiqi2@huawei.com>

Fixes: 6d021b724481 ("ARM: dump pgd, pmd and pte states on unhandled data abort faults")
Link: https://lore.kernel.org/20260716014022.2823-1-xieyuanbin1@huawei.com
Acked-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
Reviewed-by: Linus Walleij <linusw@kernel.org>
Signed-off-by: Qi Xi <xiqi2@huawei.com>
Signed-off-by: Xie Yuanbin <xieyuanbin1@huawei.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>

1 files changed, 116 insertions, 0 deletions

diff --git a/include/vdso/helpers.h b/include/vdso/helpers.h
new file mode 100644
index 000000000..a3bf4f1c0
--- /dev/null
+++ b/include/vdso/helpers.h
@@ -0,0 +1,116 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __VDSO_HELPERS_H
+#define __VDSO_HELPERS_H
+
+#ifndef __ASSEMBLY__
+
+#include <asm/barrier.h>
+#include <vdso/datapage.h>
+#include <vdso/processor.h>
+#include <vdso/clocksource.h>
+
+static __always_inline bool vdso_is_timens_clock(const struct vdso_clock *vc)
+{
+	return IS_ENABLED(CONFIG_TIME_NS) && vc->clock_mode == VDSO_CLOCKMODE_TIMENS;
+}
+
+static __always_inline u32 vdso_read_begin(const struct vdso_clock *vc)
+{
+	u32 seq;
+
+	while (unlikely((seq = READ_ONCE(vc->seq)) & 1))
+		cpu_relax();
+
+	smp_rmb();
+	return seq;
+}
+
+/*
+ * Variant of vdso_read_begin() to handle VDSO_CLOCKMODE_TIMENS.
+ *
+ * Time namespace enabled tasks have a special VVAR page installed which has
+ * vc->seq set to 1 and vc->clock_mode set to VDSO_CLOCKMODE_TIMENS. For non
+ * time namespace affected tasks this does not affect performance because if
+ * vc->seq is odd, i.e. a concurrent update is in progress the extra check for
+ * vc->clock_mode is just a few extra instructions while spin waiting for
+ * vc->seq to become even again.
+ */
+static __always_inline bool vdso_read_begin_timens(const struct vdso_clock *vc, u32 *seq)
+{
+	while (unlikely((*seq = READ_ONCE(vc->seq)) & 1)) {
+		if (vdso_is_timens_clock(vc))
+			return true;
+		cpu_relax();
+	}
+	smp_rmb();
+
+	return false;
+}
+
+static __always_inline u32 vdso_read_retry(const struct vdso_clock *vc,
+					   u32 start)
+{
+	u32 seq;
+
+	smp_rmb();
+	seq = READ_ONCE(vc->seq);
+	return unlikely(seq != start);
+}
+
+static __always_inline void vdso_write_seq_begin(struct vdso_clock *vc)
+{
+	/*
+	 * WRITE_ONCE() is required otherwise the compiler can validly tear
+	 * updates to vc->seq and it is possible that the value seen by the
+	 * reader is inconsistent.
+	 */
+	WRITE_ONCE(vc->seq, vc->seq + 1);
+}
+
+static __always_inline void vdso_write_seq_end(struct vdso_clock *vc)
+{
+	/*
+	 * WRITE_ONCE() is required otherwise the compiler can validly tear
+	 * updates to vc->seq and it is possible that the value seen by the
+	 * reader is inconsistent.
+	 */
+	WRITE_ONCE(vc->seq, vc->seq + 1);
+}
+
+static __always_inline void vdso_write_begin_clock(struct vdso_clock *vc)
+{
+	vdso_write_seq_begin(vc);
+	/* Ensure the sequence invalidation is visible before data is modified */
+	smp_wmb();
+}
+
+static __always_inline void vdso_write_end_clock(struct vdso_clock *vc)
+{
+	/* Ensure the data update is visible before the sequence is set valid again */
+	smp_wmb();
+	vdso_write_seq_end(vc);
+}
+
+static __always_inline void vdso_write_begin(struct vdso_time_data *vd)
+{
+	struct vdso_clock *vc = vd->clock_data;
+
+	vdso_write_seq_begin(&vc[CS_HRES_COARSE]);
+	vdso_write_seq_begin(&vc[CS_RAW]);
+	/* Ensure the sequence invalidation is visible before data is modified */
+	smp_wmb();
+}
+
+static __always_inline void vdso_write_end(struct vdso_time_data *vd)
+{
+	struct vdso_clock *vc = vd->clock_data;
+
+	/* Ensure the data update is visible before the sequence is set valid again */
+	smp_wmb();
+	vdso_write_seq_end(&vc[CS_HRES_COARSE]);
+	vdso_write_seq_end(&vc[CS_RAW]);
+}
+
+#endif /* !__ASSEMBLY__ */
+
+#endif /* __VDSO_HELPERS_H */