[PATCH v2] perf/ring_buffer: Fix exposing a temporarily decreased data_head.

[Yabin Cui]

In perf_output_put_handle(), an IRQ/NMI can happen in below location and
write records to the same ring buffer:
        ...
        local_dec_and_test(&rb->nest)
        ...                          <-- an IRQ/NMI can happen here
        rb->user_page->data_head = head;
        ...

In this case, a value A is written to data_head in the IRQ, then a value
B is written to data_head after the IRQ. And A > B. As a result,
data_head is temporarily decreased from A to B. And a reader may see
data_head < data_tail if it read the buffer frequently enough, which
creates unexpected behaviors.

This can be fixed by moving dec(&rb->nest) to after updating data_head,
which prevents the IRQ/NMI above from updating data_head.

Signed-off-by: Yabin Cui <yab...@google.com>
---

v1 -> v2: change rb->nest from local_t to unsigned int, and add barriers.

---
 kernel/events/internal.h    |  2 +-
 kernel/events/ring_buffer.c | 24 ++++++++++++++++++------
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/kernel/events/internal.h b/kernel/events/internal.h
index 79c47076700a..0a8c003b9bcf 100644
--- a/kernel/events/internal.h
+++ b/kernel/events/internal.h
@@ -24,7 +24,7 @@ struct ring_buffer {
        atomic_t                        poll;           /* POLL_ for wakeups */
 
        local_t                         head;           /* write position    */
-       local_t                         nest;           /* nested writers    */
+       unsigned int                    nest;           /* nested writers    */
        local_t                         events;         /* event limit       */
        local_t                         wakeup;         /* wakeup stamp      */
        local_t                         lost;           /* nr records lost   */
diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
index 674b35383491..c677beb01fb1 100644
--- a/kernel/events/ring_buffer.c
+++ b/kernel/events/ring_buffer.c
@@ -38,7 +38,8 @@ static void perf_output_get_handle(struct perf_output_handle 
*handle)
        struct ring_buffer *rb = handle->rb;
 
        preempt_disable();
-       local_inc(&rb->nest);
+       rb->nest++;
+       barrier();
        handle->wakeup = local_read(&rb->wakeup);
 }
 
@@ -54,8 +55,10 @@ static void perf_output_put_handle(struct perf_output_handle 
*handle)
         * IRQ/NMI can happen here, which means we can miss a head update.
         */
 
-       if (!local_dec_and_test(&rb->nest))
+       if (rb->nest > 1) {
+               rb->nest--;
                goto out;
+       }
 
        /*
         * Since the mmap() consumer (userspace) can run on a different CPU:
@@ -84,14 +87,23 @@ static void perf_output_put_handle(struct 
perf_output_handle *handle)
         * See perf_output_begin().
         */
        smp_wmb(); /* B, matches C */
-       rb->user_page->data_head = head;
+       WRITE_ONCE(rb->user_page->data_head, head);
+
+       /*
+        * Clear rb->nest after updating data_head. This prevents IRQ/NMI from
+        * updating data_head before us. If that happens, we will expose a
+        * temporarily decreased data_head.
+        */
+       WRITE_ONCE(rb->nest, 0);
 
        /*
-        * Now check if we missed an update -- rely on previous implied
-        * compiler barriers to force a re-read.
+        * Now check if we missed an update -- use barrier() to force a
+        * re-read.
         */
+       barrier();
        if (unlikely(head != local_read(&rb->head))) {
-               local_inc(&rb->nest);
+               rb->nest++;
+               barrier();
                goto again;
        }
 
-- 
2.21.0.1020.gf2820cf01a-goog