On Thursday 15 February 2007 14:02, Michael Wu wrote: > If we're going to switch to a library approach, let's do it all at once. I > don't want some features requiring flags to toggle and others requiring a > function to call. > To be specific, I don't like the library approach when we can simply set a flag and let the stack handle it. It may clutter the TX path, but it reduces complexity on the driver end. Also, this particular test can be moved from the TX path to device registration with some changes to the way the TX handlers are handled. I will need to check if there are other TX handlers that would benefit from this to make it worthwhile. For things like encryption, I think library functions or callbacks will really be needed to do things right, but for software sequence numbers, I'd rather stick with a simple flag to enable/disable it. It also allows us to report if a particular device supports software sequence numbers which the use of a library function would not specify so directly. -Michael Wu
Attachment:
pgpBGgFSbYpbB.pgp
Description: PGP signature