On 2023-02-13 12:43, Namhyung Kim wrote:
Hi Steve,
On Sat, Feb 11, 2023 at 8:07 PM Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
On Sun, 12 Feb 2023 11:38:52 +0800
Yafang Shao <laoar.shao@xxxxxxxxx> wrote:
Actually, there are cases that this needs to be a number, as b3bc8547d3be6
("tracing: Have TRACE_DEFINE_ENUM affect trace event types as well") made
it update fields as well as the printk fmt.
It seems that TRACE_DEFINE_ENUM(TASK_COMM_LEN) in the trace events
header files would be a better fix.
NACK! I much prefer the proper fix that adds the length.
Can we just have both enum and macro at the same time?
I guess the enum would fill the BTF and the macro would provide
backward compatibility.
This is no need to keep the define. The root cause of the issue is
addressed by this fix:
https://lore.kernel.org/lkml/20230212154620.4d8fe033@xxxxxxxxxxxxxxxxxx/
Thanks,
Mathieu
Thanks,
Namhyung
--
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com
