Hi I am working on an application where I have a single thread per core (the application is based on top of SPDK - https://spdk.io/). Ideally, I would like to have a set of buffers allocated per core and since there is a single thread it can write the trace event to this buffer directly with no synchronization. Once the buffer is filled up, the buffer can be given to LTTng to take care of the persistence. I'm looking at the LTTng code to figure out what kind of changes would need to be made to support something like this. I imagine it would need to change the ring buffer because the configuration for that while tunable does not have any mode which does what I want. Am I looking in the right place? If you can provide me to pointers on how I can prototype this with lttng that would be most helpful. Once I have it prototyped, I would be happy to work with you on making these changes back into LTTng if you feel that this would be of value to other users as well. Regards Aprameya