[PATCH v4] kasan: Don't call find_vm_area() in a PREEMPT_RT kernel

[Waiman Long <longman@redhat.com>]

The following bug report was found when running a PREEMPT_RT debug kernel.

 BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48
 in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 140605, name: kunit_try_catch
 preempt_count: 1, expected: 0

 Call trace:
  rt_spin_lock+0x70/0x140
  find_vmap_area+0x84/0x168
  find_vm_area+0x1c/0x50
  print_address_description.constprop.0+0x2a0/0x320
  print_report+0x108/0x1f8
  kasan_report+0x90/0xc8

Since commit e30a0361b851 ("kasan: make report_lock a raw spinlock"),
report_lock was changed to raw_spinlock_t to fix another similar
PREEMPT_RT problem. That alone isn't enough to cover other corner cases.

print_address_description() is always invoked under the
report_lock. The context under this lock is always atomic even on
PREEMPT_RT. find_vm_area() acquires vmap_node::busy.lock which is a
spinlock_t, becoming a sleeping lock on PREEMPT_RT and must not be
acquired in atomic context.

Don't invoke find_vm_area() on PREEMPT_RT and just print the address.
Non-PREEMPT_RT builds remain unchanged. Add a DEFINE_WAIT_OVERRIDE_MAP()
macro to tell lockdep that this lock nesting is allowed because the
PREEMPT_RT part (which is invalid) has been taken care of. This macro
was first introduced in commit 0cce06ba859a ("debugobjects,locking:
Annotate debug_object_fill_pool() wait type violation").

Fixes: e30a0361b851 ("kasan: make report_lock a raw spinlock")
Suggested-by: Andrey Konovalov <andreyknvl@gmail.com>
Signed-off-by: Waiman Long <longman@redhat.com>
---
 mm/kasan/report.c | 34 +++++++++++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

 [v4] Use Andrey's suggestion of a kasan_find_vm_area() helper and
 update comment and commit log as suggested by Andrey and Sebastian.

diff --git a/mm/kasan/report.c b/mm/kasan/report.c
index 3fe77a360f1c..8357e1a33699 100644
--- a/mm/kasan/report.c
+++ b/mm/kasan/report.c
@@ -370,6 +370,36 @@ static inline bool init_task_stack_addr(const void *addr)
 			sizeof(init_thread_union.stack));
 }
 
+/*
+ * This function is invoked with report_lock (a raw_spinlock) held. A
+ * PREEMPT_RT kernel cannot call find_vm_area() as it will acquire a sleeping
+ * rt_spinlock.
+ *
+ * For !RT kernel, the PROVE_RAW_LOCK_NESTING config option will print a
+ * lockdep warning for this raw_spinlock -> spinlock dependency. This config
+ * option is enabled by default to ensure better test coverage to expose this
+ * kind of RT kernel problem. This lockdep splat, however, can be suppressed
+ * by using DEFINE_WAIT_OVERRIDE_MAP() if it serves a useful purpose and the
+ * invalid PREEMPT_RT case has been taken care of.
+ */
+static inline struct vm_struct *kasan_find_vm_area(void *addr)
+{
+	static DEFINE_WAIT_OVERRIDE_MAP(vmalloc_map, LD_WAIT_SLEEP);
+	struct vm_struct *va;
+
+	if (IS_ENABLED(CONFIG_PREEMPT_RT))
+		return NULL;
+
+	/*
+	 * Suppress lockdep warning and fetch vmalloc area of the
+	 * offending address.
+	 */
+	lock_map_acquire_try(&vmalloc_map);
+	va = find_vm_area(addr);
+	lock_map_release(&vmalloc_map);
+	return va;
+}
+
 static void print_address_description(void *addr, u8 tag,
 				      struct kasan_report_info *info)
 {
@@ -399,7 +429,7 @@ static void print_address_description(void *addr, u8 tag,
 	}
 
 	if (is_vmalloc_addr(addr)) {
-		struct vm_struct *va = find_vm_area(addr);
+		struct vm_struct *va = kasan_find_vm_area(addr);
 
 		if (va) {
 			pr_err("The buggy address belongs to the virtual mapping at\n"
@@ -409,6 +439,8 @@ static void print_address_description(void *addr, u8 tag,
 			pr_err("\n");
 
 			page = vmalloc_to_page(addr);
+		} else {
+			pr_err("The buggy address %px belongs to a vmalloc virtual mapping\n", addr);
 		}
 	}
 
-- 
2.48.1