On Mon, Nov 12, 2018 at 09:31:51PM -0800, Joe Perches wrote: > On Tue, 2018-11-13 at 16:26 +1100, Dave Chinner wrote: > > On Mon, Nov 12, 2018 at 08:23:42PM -0800, Joe Perches wrote: > > > On Tue, 2018-11-13 at 14:09 +1100, Dave Chinner wrote: > > > > On Mon, Nov 12, 2018 at 08:54:10PM -0500, Theodore Y. Ts'o wrote: > > > > > On Tue, Nov 13, 2018 at 12:18:05PM +1100, Dave Chinner wrote: > > > > > > I'm not interested in making code fast if distro support engineers > > > > > > can't debug problems on user systems easily. Optimising for > > > > > > performance over debuggability is a horrible trade off for us to > > > > > > make because it means users and distros end up much more reliant on > > > > > > single points of expertise for debugging problems. And that means > > > > > > the majority of the load of problem triage falls directly on very > > > > > > limited resources - the core XFS development team. A little bit of > > > > > > thought about how to make code easier to triage and debug goes a > > > > > > long, long way.... > > > > > > > > > > So at least in my experience, if the kernels are compiled with > > > > > CONFIG_DEBUG_INFO and/or CONFIG_DEBUG_INFO_REDUCED, > > > > > scripts/decode_stracktrace.sh seems to do a very nice job with inlined > > > > > > > > That doesn't help with kernel profiling and other such things that > > > > are based on callgraphs... > > > > > > If that's really the case: > > > > > > I rather suspect the xfs static v STATIC function marking is not > > > particularly curated and the marking is somewhat arbitrary. I disagree. I've added plenty of code over the past couple of years. Short functions with few or no branches (e.g. converters) are 'static'; longer functions (loops, iterators, "decide what to do with this" functions, etc.) with many branches are STATIC to make it easier for me to ftrace their decisions over a given dataset. > > That's a common opinion for an outsider to form when they come > > across something unfamiliar they don't really understand. "I don't > > understand this, so I must rewrite it" is an unfortunate habit that > > programmers have. > > Silly. Yet frequent. > > > So perhaps given the large number of static, but not STATIC > > > functions, perhaps a sed of s/static/STATIC/ should be done > > > when it's not inline for all xfs functions. > > > > That's just as bad as removing them all, if not worse. > > Why? > > > If you are writing new code or reworking existing code, then we'll > > consider the usage of STATIC/static in the context of that work. > > Otherwise, we leave it alone. > > If your statement is as described above, and > the STATIC use to enable call stack tracing i > useful, why shouldn't it be systemic? > > > It if ain't broke, don't fix it. > > A generically lazy statement. Please everyone let's take a breather from this thread for a few hours. A 3.7% reduction in code size is not worth getting worked up over, IMO. --D > >
