Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: anton@mips.complang.tuwien.ac.at (Anton Ertl)
Newsgroups: comp.lang.forth
Subject: Re: Stack vs stackless operation
Date: Thu, 27 Feb 2025 22:03:55 GMT
Organization: Institut fuer Computersprachen, Technische Universitaet Wien
Lines: 72
Message-ID: <2025Feb27.230355@mips.complang.tuwien.ac.at>
References: <591e7bf58ebb1f90bd34fba20c730b83@www.novabbs.com> <34df278ef0a52d0eab9d035f45795389@www.novabbs.com> <a6c9d8a0aa7e1046af7948093e07cff0@www.novabbs.com> <2025Feb26.153250@mips.complang.tuwien.ac.at> <2025Feb26.184613@mips.complang.tuwien.ac.at> <875xkwo5io.fsf@nightsong.com> <2025Feb27.082944@mips.complang.tuwien.ac.at> <871pvjnnlc.fsf@nightsong.com>
Injection-Date: Thu, 27 Feb 2025 23:30:13 +0100 (CET)
Injection-Info: dont-email.me; posting-host="139269432581838c150e739c2a2a64f6";
	logging-data="3477830"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19Ek/aGSceiRuR/2HF5/xFf"
Cancel-Lock: sha1:rd+J2oqglsGnDqswxc6PKE8N0Lo=
X-newsreader: xrn 10.11
Bytes: 4328

Paul Rubin <no.email@nospam.invalid> writes:
>anton@mips.complang.tuwien.ac.at (Anton Ertl) writes:
>> Results (on Zen4):
>> gforth-fast (development): ...
>
>It's interesting how little difference there is with gforth-fast.  Could
>you also do gforth-itc?

gforth-itc (development):
      :=:        exchange          ex         ex-locals      exchange2
 7_527_256_553 5_224_615_325  6_825_283_178  9_238_357_501  7_036_128_309 c.
13_127_503_990 9_326_561_471 12_927_054_153 16_927_820_825 12_027_146_677 i.

For comparison: gforth-fast (development):
      :=:        exchange         ex        ex-locals      exchange2
  814_881_277   879_389_133   928_825_521   875_574_895   808_543_975 cyc.
3_908_874_164 3_708_891_336 4_508_966_770 4_209_778_557 3_708_865_505 inst.

>exchange2 is a big win with VFX, suggesting its
>optimizer could do better with some of the other versions.

On VFX exchange2 takes the same speed and the same number of
instructions as :=:.  EX is slower because VFX does not analyse the
return stack, unlike the data stack.  EX-LOCALS is slow because VFX's
locals implementation is not particularly good.

To see what a better analysis can do, let's look at lxf:

      :=:           ex         ex-locals     exchange2
  502_740_029   502_189_567   502_134_842   502_043_217 cycles
1_701_663_782 1_701_657_866 1_701_677_273 1_701_684_186 instructions

The cycles and instructions are worse (except for ex-locals) than with
VFX, but that's due to inlining (which VFX does and lxf does not).

E.g., here's lxf's code for EX-LOCALS:

 869204C  804FCE2  23  88C8000   5 normal  EX-LOCALS

 804FCE2 8B4500                 mov     eax , [ebp]
 804FCE5 8B00                   mov     eax , [eax]
 804FCE7 8BCB                   mov     ecx , ebx
 804FCE9 8B09                   mov     ecx , [ecx]
 804FCEB 8B5500                 mov     edx , [ebp]
 804FCEE 890A                   mov     [edx] , ecx
 804FCF0 8903                   mov     [ebx] , eax
 804FCF2 8B5D04                 mov     ebx , [ebp+4h]
 804FCF5 8D6D08                 lea     ebp , [ebp+8h]
 804FCF8 C3                     ret     near

It's the same code as lxf produces for :=:.

The code lxf produces for EX and EXCHANGE2 is:

 804FCF9 8BC3                   mov     eax , ebx
 804FCFB 8B00                   mov     eax , [eax]
 804FCFD 8B4D00                 mov     ecx , [ebp]
 804FD00 8B09                   mov     ecx , [ecx]
 804FD02 890B                   mov     [ebx] , ecx
 804FD04 8B5D00                 mov     ebx , [ebp]
 804FD07 8903                   mov     [ebx] , eax
 804FD09 8B5D04                 mov     ebx , [ebp+4h]
 804FD0C 8D6D08                 lea     ebp , [ebp+8h]
 804FD0F C3                     ret     near

- anton
--
M. Anton Ertl  http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
     New standard: https://forth-standard.org/
EuroForth 2023 proceedings: http://www.euroforth.org/ef23/papers/
EuroForth 2024 proceedings: http://www.euroforth.org/ef24/papers/