Merge tag 'trace-ring-buffer-v7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

Pull ring-buffer updates from Steven Rostedt

 - Do not invalidate entire buffer for invalid sub-buffers

   For the persistent ring buffer, if one sub-buffer is found to be
   invalid, it invalidates the entire per CPU ring buffer. This can lose
   a lot of valuable data if there's some corruption with the writes to
   the buffer not syncing properly on a hard crash. Instead, if a
   sub-buffer is found to be invalid, simply zero it out and mark it for
   "missed events".

   When the persistent ring buffer is read and a sub-buffer that was
   cleared due to being invalid on boot up is discovered, the output
   will show "[LOST EVENTS]" to let the user know that events were
   missing at that location. Displaying the events from valid buffers
   can still be useful.

 - Add a test to be able to test corrupted sub-buffers

   If a persistent ring buffer is created as "ptraingtest" and the new
   config that adds the test is enabled, when a panic happens, the
   kernel will randomly corrupt one of the per CPU ring buffers. On boot
   up, the sub-buffers with the corruption should be cleared and
   flagged. When reading this buffer, the missed events should should
   [LOST EVENTS].

 - Add commit number in the sub-buffer meta debug info

   The commit is used to know the content of a meta page. Add it to the
   buffer_meta file that is shown for each per CPU buffer.

 - Clean up the persistent ring buffer validation code

   Add some helper functions and make variable names more consistent.

* tag 'trace-ring-buffer-v7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Better comment the use of RB_MISSED_EVENTS
  ring-buffer: Show persistent buffer dropped events in trace_pipe file
  ring-buffer: Show persistent buffer dropped events in trace file
  ring-buffer: Have dropped subbuffers be persistent across reboots
  ring-buffer: Cleanup buffer_data_page related code
  ring-buffer: Cleanup persistent ring buffer validation
  ring-buffer: Show commit numbers in buffer_meta file
  ring-buffer: Add persistent ring buffer invalid-page inject test
  ring-buffer: Skip invalid sub-buffers when rewinding persistent ring buffer
  ring-buffer: Skip invalid sub-buffers when validating persistent ring buffer

include/linux/ring_buffer.h

@@ -238,6 +238,7 @@ int ring_buffer_subbuf_size_get(struct trace_buffer *buffer);
 
 enum ring_buffer_flags {
 	RB_FL_OVERWRITE		= 1 << 0,
+	RB_FL_TESTING		= 1 << 1,
 };
 
 #ifdef CONFIG_RING_BUFFER

kernel/trace/Kconfig

@@ -1202,6 +1202,40 @@ config RING_BUFFER_VALIDATE_TIME_DELTAS
 	  Only say Y if you understand what this does, and you
 	  still want it enabled. Otherwise say N
 
+config RING_BUFFER_PERSISTENT_INJECT
+	bool "Enable persistent ring buffer error injection test"
+	depends on RING_BUFFER
+	help
+	  This option will have the kernel check if the persistent ring
+	  buffer is named "ptracingtest". and if so, it will corrupt some
+	  of its pages on a kernel panic. This is used to test if the
+	  persistent ring buffer can recover from some of its sub-buffers
+	  being corrupted.
+	  To use this, boot a kernel with a "ptracingtest" persistent
+	  ring buffer, e.g.
+
+	   reserve_mem=20M:2M:trace trace_instance=ptracingtest@trace panic=1
+
+	  And after the 1st boot, run the following commands:
+
+	   cd /sys/kernel/tracing/instances/ptracingtest
+	   echo 1 > events/enable
+	   echo 1 > tracing_on
+	   sleep 3
+	   echo c > /proc/sysrq-trigger
+
+	  After the panic message, the kernel will reboot and will show
+	  the test results in the console output.
+
+	  Note that events for the test ring buffer needs to be enabled
+	  prior to crashing the kernel so that the ring buffer has content
+	  that the test will corrupt.
+	  As the test will corrupt events in the "ptracingtest" persistent
+	  ring buffer, it should not be used for any other purpose other
+	  than this test.
+
+	  If unsure, say N
+
 config MMIOTRACE_TEST
 	tristate "Test module for mmiotrace"
 	depends on MMIOTRACE && m

kernel/trace/trace.c

@@ -8380,6 +8380,8 @@ static void setup_trace_scratch(struct trace_array *tr,
 	memset(tscratch, 0, size);
 }
 
+#define TRACE_TEST_PTRACING_NAME	"ptracingtest"
+
 int allocate_trace_buffer(struct trace_array *tr, struct array_buffer *buf, int size)
 {
 	enum ring_buffer_flags rb_flags;
@@ -8391,6 +8393,8 @@ int allocate_trace_buffer(struct trace_array *tr, struct array_buffer *buf, int
 	buf->tr = tr;
 
 	if (tr->range_addr_start && tr->range_addr_size) {
+		if (tr->name && !strcmp(tr->name, TRACE_TEST_PTRACING_NAME))
+			rb_flags |= RB_FL_TESTING;
 		/* Add scratch buffer to handle 128 modules */
 		buf->buffer = ring_buffer_alloc_range(size, rb_flags, 0,
 						      tr->range_addr_start,