On (08/30/07 20:42), Peter Memishian wrote: > > I didn't realize that was the justification for the NOTR() variants. > With Jim's new design, it doesn't seem like it would be hard to support > this case and remove the NOTR() stuff. Am I missing something? >
I believe that the entire refcount debugging model was built around ensuring that a thread that takes the ref also releases the ref before it exits. This cannot be done for long-lived refs e.g., the ones taken for ire->ire_nce (hence we have ugly macros like NCE_REFHOLD_TO_REFHOLD_NOTR). I'm not sure how Jim's design changes that basic assumption behind the trace_refs. --sowmini _______________________________________________ networking-discuss mailing list [email protected]
