Mariusz Tkaczyk wrote: > On Thu, 4 May 2023 16:58:35 -0700 > Dan Williams <dan.j.williams@xxxxxxxxx> wrote: > > > Mariusz Tkaczyk wrote: > > > Remove active, status, fault and locate variables from > > > enclosure_component struct. Return then directly. > > > No functional changes intended. > > > > This looks ok although it's not a clear win on the diffstat. Does this > > make the NPEM implementation easier to remove the indirection through > > "struct enclosure_component" for reading fresh values? That would help > > make the case. > > I did that to familiarize better with this API. I determined that those values > can be just returned. They are refreshed every time on read in get_*led*(). I > believed that it makes implementation simpler for reader. > > It could save me from some questions, "why not to reuse existing active, > fault, status variables" but it is no clear benefit. > It saves some memory because those variable probably won't be used in > NPEM/_DSM implementation (at least draft I left doesn't not use them). > > If you don't see this valuable let me know, I can drop it. I think the problem with it is that it now requires a get_active() implementation where one was not needed before, see my comment on patch3. The attributes to cache the values allows a 'get' method to be skipped when 'set'+caching is sufficient.
