Hello,

I observe that FPC trunk compiler is about 65-70% (factor 1.65-1.7) slower than FPC 3.0.4 compiler.

E.g. building Lazarus IDE takes on my machine:
1:00 with FPC 3.0.4
1:40 with FPC trunk

Do you observe the same? Any hints why? Is anybody worried about the speed? Is anybody working on the speed issues? The compile time increase is huge.

I seem to have the same fpc.cfg and FPC/Lazarus build options.

Ondrej

Building the IDE with custom opts, include -gw ...

3.0.4  60 seconds
trunk 67 seconds   3.1.1 [2018/04/16] not sure what revision trunk 73 seconds   revision 39024 (18 May 2018)

All builds were run twice, and the 2nd run taken. So OS file caching from previous builds should be similar for all of them.

New features? E.g. helpers made fpc a lot slower (or are they already in 3.0.x?) Optimizer improved and made the compiler slower etc.

Is anybody worried about the speed? Is anybody working on the speed issues?

Not really, we could only remove features ;)

The compile time increase is huge.

The new typed constant generator is also much slower than the old "just emit sequences of bytes" approach. On the other hand, it's much less error prone (no more alignment issues), much more portable and higher level. It's been in trunk for four years already though.

Jonas

One could possibly refactor elements of the compiler, and the improvements to the optimizer might also help in places. I wouldn't call it a major priority though, although I will make suggestions if I spot anything.  Granted, the niche that I've crafted for myself is assembly language optimisation, which is not appropriate for cross-platform source code.

Gareth aka. Kit

On Mon 21/05/18 16:55 , Jonas Maebe sent:

Florian Klämpfl wrote

Am 18.05.2018 um 17:32 schrieb Ondrej Pokorny: > > I observe that FPC trunk compiler is about 65-70% (factor 1.65-1.7)

slower than FPC 3.0.4 compiler.

3.0.x?) Optimizer improved

and made the compiler slower etc.

The new typed constant generator is also much slower than the old "just emit sequences of bytes" approach. On the other hand, it's much less error prone (no more alignment issues), much more portable and higher level. It's been in trunk for four years already though.

Jonas

In our previous episode, Florian Kl?mpfl said:

Afaik 3.0.x only has helpers for classes (structed types in general?), and trunk also for other types.

Marco van de Voort wrote

Afaik 3.0.x only has helpers for classes (structed types in general?), and trunk also for other types.

3.0.4 at least has type helpers for integer/string etc.

Bart

Marco van de Voort schrieb am Mo., 21. Mai 2018, 21:05:

Type helpers are already in 3.0.

Regards,
Sven