On 10/17/25 2:08 AM, Steven Rostedt wrote:

On Fri, 17 Oct 2025 00:07:53 -0700
Randy Dunlap [off-list ref] wrote:

quoted

Format the kernel-doc for RINGBUF_TYPE_DATA_TYPE_LEN_MAX correctly
to prevent a kernel-doc warning:

Warning: include/linux/ring_buffer.h:61 Enum value
 'RINGBUF_TYPE_DATA_TYPE_LEN_MAX' not described in enum 'ring_buffer_type'

Then this needs to be updated differently, because that "<=" is stating
what happens when the value is <= RINGBUF_TYPE_DATA_TYPE_LEN_MAX.

It wasn't random characters.

Basically, the enum describes the event type.

enum ring_buffer_type {
        RINGBUF_TYPE_DATA_TYPE_LEN_MAX = 28,
        RINGBUF_TYPE_PADDING,
        RINGBUF_TYPE_TIME_EXTEND,
        RINGBUF_TYPE_TIME_STAMP,
};

When the type is > 28 it is either a padding, time-extend or
time-stamp. But if it is less than or equal to
RINGBUF_TYPE_DATA_TYPE_LEN_MAX then it is the length of a data event.

Perhaps we should have it be:

 * 0:
 *                              Data record
 *                                array[0] holds the actual length
 *                                array[1..(length+3)/4] holds data
 *                                size = 4 + length (bytes)
 *
 * 1 - @RINGBUF_TYPE_DATA_TYPE_LEN_MAX:
 *                              Data record
 *                                length = type_len << 2
 *                                array[0..(length+3)/4-1] holds data
 *                                size = 4 + length (bytes)

  ?

This data is more important that making kerneldoc work.

For sure it is. Thanks for the explanation.
I'll see if I can come up with an alternative - or not.

quoted

Fixes: 334d4169a659 ("ring_buffer: compressed event header")
Signed-off-by: Randy Dunlap <redacted>
---
Cc: Lai Jiangshan <redacted>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: linux-trace-kernel@vger.kernel.org
---
 include/linux/ring_buffer.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- linux-next-20251016.orig/include/linux/ring_buffer.h
+++ linux-next-20251016/include/linux/ring_buffer.h

@@ -43,7 +43,7 @@ struct ring_buffer_event {
  *				 array[0] = top (28 .. 59) bits
  *				 size = 8 bytes
  *
- * <= @RINGBUF_TYPE_DATA_TYPE_LEN_MAX:
+ * @RINGBUF_TYPE_DATA_TYPE_LEN_MAX:
  *				Data record
  *				 If type_len is zero:
  *				  array[0] holds the actual length

-- 
~Randy