Re: [Patch] C++ inlining heuristics changed (3.0.1)

[Daniel Berlin <dan at cgsoftware dot com>]

Kurt Garloff <kurt@garloff.de> writes:

> Hi Richard,
>
> thanks for your report!
>
> On Thu, Aug 23, 2001 at 03:38:22PM +0200, Richard B. Kreckel wrote:
> > On Wed, 22 Aug 2001 21:42:29 +0200, Kurt Garloff wrote:
>> > I'd like to have this patch included in 3.0.2. (Unless somebody can come
>> > with an inliner for 3.0.2 that does not cut inlining at leaves but at the
>> > trunk of the call tree.)
>> 
>> I just bootstrapped successfully with Kurt's patch applied and tested it
>> on the benchmarks in GiNaC's suite of regression tests.  Run-time
>> performance was somewhere between inline-limit 3000 and 4000 as seen
>> before, i.e. close to maximum, so this looks good.  File-size was actually
>> slightly smaller than any I have seen before.  (But as to compile-time it
>> seems that we are not very sensitive, they varied only about 10%.)
>
> Compared to 3.0.1? It should more or less be the same, yes.
> If I want to summarize the patch: It should give 3.0 performance (or better
> in case of i-cache effects) at 3.0.1 compile speed (or slightly better :-)
>
> But after 2.95.3 experience, I might come up with a slightly better variant.
> http://gcc.gnu.org/ml/gcc/2001-08/msg01110.html
> At this moment, C++ inlining is a bit detuned WRT C inlining, I believe; we
> need something like 500 or 600 to get good C++ performance, as we only allow
> half of this size for single functions. For plain C, this limit might be a
> little high; that's unchanged from 3.0.1 though, so no pessimization.
>
> But, I'll check whether I can do the following:
> * Set the default max size for inlining to a lower value (300), and use
>   it as max size for single functions. (So this stays effectively unchanged
>   from v2 of the patch; but the RTL inliner may be more happy, as it uses
>   the same number AFAICS so C programmers might appreciate this. OTOH, they
>   did not complain too much about 3.0 or 3.0.1, where it was 10k resp. 600,
>   did they?)

This is because the RTL inliner has restrictions, and it's heuristic
was/is to only allow things not declared inline to be inlined if the
inlining is less than the call cos.
Since inline isn't used much in C code, ....

> * Start throttling recursive inlining later (e.g. 4 times the new limit, IOW
>   2 times the v2 one)
> * Maybe we can somehow also give leaf functions a bonus in the tree
>   inliner

You need a more general infrastructue to do this. I'll get around to
submitting/committing the callgraph and high level loop optimization
stuff i've been working on in my spare time, to the
ast-optimizer-branch, sometime in the next few months.

> * Maybe we can find a way to compute the minimum limit I fixed to 130
>   in v2 of the patch. It should be a function of argument list length etc.
>   Don't know, whether it's worth the trouble.


I tried this a few weeks ago, but it doesn't actually help.
The problem is that because we haven't generated the actual RTL yet,
the trees for the call arguments aren't representative at all of how
much code each argument will actually generate.



>
> Is the RTL inlining pass (integrate.c) still used, when we compile
> C++ with 3.0.1, or is the tree inliner (optimize.c) all that's being 
> applied, BTW?

C++ doesn't use RTL inlining anymore.

>
>> So, from my limited persective, I would like to recommend Kurt's patch for
>> GCC-3.0.2 unless somebody really fixes the inliner for good.
>
> Thx for testing and supporting!
> -- 
> Kurt Garloff                   <kurt@garloff.de>         [Eindhoven, NL]
> Physics: Plasma simulations  <K.Garloff@Phys.TUE.NL>  [TU Eindhoven, NL]
> Linux: SCSI, Security          <garloff@suse.de>    [SuSE Nuernberg, DE]
>  (See mail header or public key servers for PGP2 and GPG public keys.)

-- 
"A while ago, I went skiing in England.  It was a rare package:
two weeks in England, one night in Connecticut, two weeks in
England.  I said, "Yes, I'll take it."  I got on this chairlift
with this guy I didn't know.  We went halfway up the mountain
without saying a word.  Then he turned to me and said, "You
know, this is the first time I've gone skiing in ten years."  I
said, "Why did you take such a long time off?"  He said, "I was
in prison.  Want to know why?"  I said, "Not really.  Well, you
better tell me why."  He said, "I pushed a total stranger off a
Ferris wheel."  I said, "I remember you."
"-Steven Wright