Re: Profiling native-compilation

[Andrea Corallo]

Eli Zaretskii <eliz@gnu.org> writes:

> I tried to profile the code involved in native-compilation using
> profiler.el, but the results seem to omit most of the functions
> involved in that: the percents shown in the resulting profile are very
> low, which seems to hint that most of the CPU time is spent in places
> that profiler.el cannot identify.
>
> Is it possible to use profiler.el to produce a meaningful profile of
> native-compilation?


> If yes, how does one go about setting up such
> profiling?  And if this is not currently possible, why isn't it, and
> can we do something to make the situation better?  For example, for at
> least two C functions that are important to see in the profile -- GC
> and redisplay -- we did arrange to be shown in the profile.  So if the
> problem is that most of the time is spent inside C code, perhaps we
> could do something similar in comp.c?

Hi Eli,

AFAIK profiling syncronous native compilation with profile.el should be
possible but with one caveat that is: we run the final pass in a
subprocess so we'll loose visibility on that, OTOH we can consider final
being essentially all libgccjit work.

Maybe that's the reason why you see low percentages? Have you profiled a
`native-compile' invocation?

> I think this is important because native-compilation is quite slow,
> and at least I don't have a clear idea where most of that time is
> spent.  Andrea, you said once that most of the time is spent in
> libgccjit functions, which I find strange, since those are basically
> parts of GCC code, and I don't think I ever saw such slow compilation
> times from GCC.

Yes I mentioned that but since we changed the layout of generated code
for storing immediates GCC got way faster and now the Emacs side is
about equally responsible for the time spent in compilation.

One simple approach to get an idea of where time is spent in the
compiler is to bind `comp-log-time-report' to get a report pass by pass.

Running:

===
(require 'comp)
(let ((comp-log-time-report t))
  (native-compile ".../emacs2/lisp/emacs-lisp/byte-opt.el"))
===

I get:

Done compiling 
.../.emacs.d/eln-cache/28.0.50-7f3f9b99/byte-opt-9c5f25f5-d2005e8f.eln
Pass comp-spill-lap took: 0.095720s.
Pass comp-limplify took: 0.423704s.
Pass comp-fwprop took: 3.334509s.
Pass comp-call-optim took: 0.023363s.
Pass comp-ipa-pure took: 0.000046s.
Pass comp-add-cstrs took: 0.222719s.
Pass comp-fwprop took: 2.549254s.
Pass comp-tco took: 0.000044s.
Pass comp-fwprop took: 0.256151s.
Pass comp-remove-type-hints took: 0.025859s.
Pass comp-final took: 4.268238s.

> So I guess there's more here than meets the eye, and
> I think we should work on getting a clear idea where the time is
> spent, so that we could improve it.  For example, if indeed libgccjit
> is the main sink, we should ask the respective GCC developers to speed
> it up.

Agree there is certainly room for ATM unexplored improvements, IIRC
consing and GC have a considerable factor in the Emacs side of the
compilation.

I think would be interesting to sort by priority what I think are the
three main ares to work on the native compiler on the post merge phase:
compile-time, generated code performance, hardening, documentation.

Thanks

  Andrea