kcov: prefault the kcov_area

On many architectures the vmalloc area is lazily faulted in upon first
access. This is problematic for KCOV, as __sanitizer_cov_trace_pc
accesses the (vmalloc'd) kcov_area, and fault handling code may be
instrumented. If an access to kcov_area faults, this will result in
mutual recursion through the fault handling code and
__sanitizer_cov_trace_pc(), eventually leading to stack corruption
and/or overflow.

We can avoid this by faulting in the kcov_area before
__sanitizer_cov_trace_pc() is permitted to access it. Once it has been
faulted in, it will remain present in the process page tables, and will
not fault again.

[[e-mail]: code cleanup]
[[e-mail]: add comment explaining kcov_fault_in_area()]
[[e-mail]: fancier code comment from Mark]
Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Mark Rutland <[e-mail]>
Acked-by: Andrey Ryabinin <[e-mail]>
Cc: Dmitry Vyukov <[e-mail]>
Cc: Ingo Molnar <[e-mail]>
Cc: Peter Zijlstra <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>
Signed-off-by: Linus Torvalds <[e-mail]>

kernel/kcov.c

···
 	return 0;
 }
 
+/*
+ * Fault in a lazily-faulted vmalloc area before it can be used by
+ * __santizer_cov_trace_pc(), to avoid recursion issues if any code on the
+ * vmalloc fault handling path is instrumented.
+ */
+static void kcov_fault_in_area(struct kcov *kcov)
+{
+	unsigned long stride = PAGE_SIZE / sizeof(unsigned long);
+	unsigned long *area = kcov->area;
+	unsigned long offset;
+
+	for (offset = 0; offset < kcov->size; offset += stride)
+		READ_ONCE(area[offset]);
+}
+
 static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,
 			     unsigned long arg)
 {
···
 #endif
 		else
 			return -EINVAL;
+		kcov_fault_in_area(kcov);
 		/* Cache in task struct for performance. */
 		t->kcov_size = kcov->size;
 		t->kcov_area = kcov->area;