Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.c++ > #9833

Re: does virtualizing all methods slow C++ down ?

Show all headers | View raw

On 9/6/2011 3:40 PM, Alain Ketterlin wrote:
> jacob navia<jacob@spamsink.net>  writes:
>
>> Le 06/09/11 21:13, Lynn McGuire a écrit :
>>> Thanks, I had not thought about the direct calls just
>>> directly binding to the proper method. I'll bet that
>>> there are few direct calls in our code though as we
>>> have many children classes.
>>>
>>> ObjPtr ->  DataGroup ->  GenGroup
>>>
>>> ObjPtr ->  DataGroup ->  NodeGroup ->  StreamGroup
>>>
>>> are a couple of example lineages.
>>>
>>> Lynn
>>
>> This is completely ridiculous. An indirect call vs a direct call
>> will cost you some pipeline turbulence and the cache may be flushed,
>> really not a big deal.
>
> Depends on the code. With lots of small methods (a la Java, with
> getters/setters all over the place), it may be significant, mainly
> because of the impossibility to inline. Take any program, compile it
> with inlining disabled: the difference may be significant.
>
>> Have you MEASURED the slowdown? Is it significant in relationship
>> to OTHER improvements in your code?
>
> Hard to measure. You need to change tens/hundreds of classes...
>
>> Have you profiled your code to KNOW where it is spending 90% of the
>> time?
>
> In the case I mention above (many small functions), the overhead would
> be spread over a large part of the code. Profiling becomes less useful.
>
>> Before you answer THOSE questions, worrying about direc/indirect
>> calls is WASTED time and effort.
>
> I've just done this on a piece code that was evaluating an expression
> tree on each vertex of a 4D grid, with an insane number of
> floating-point ops. Incredible speedup (~120). Granted, the original was
> over-engineered, and this may be a corner case.
>
> Regarding Lynn's case, I would say you're right: he may be better off
> keeping his current design safe rather than trying to save a few cycles.
>
> -- Alain.

You are more than correct, I was just wondering.  In fact, we
saved millions of cycles by porting our code from Smalltalk
to C++.  We had to do it anyway, our Smalltalk variant was
Win16 (don't even go there !) without a Win32 port.  I would
not be surprised if we got a 100X speedup by converting to
C++ and Win32.  Our current design is solid and will not be
changed - we have many other dragons to slay.

Thanks,
Lynn

Back to comp.lang.c++ | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

does virtualizing all methods slow C++ down ? Lynn McGuire <lmc@winsim.com> - 2011-09-06 11:31 -0500
  Re: does virtualizing all methods slow C++ down ? Christopher <cpisz@austin.rr.com> - 2011-09-06 10:00 -0700
    Re: does virtualizing all methods slow C++ down ? Lynn McGuire <lmc@winsim.com> - 2011-09-06 14:09 -0500
      Re: does virtualizing all methods slow C++ down ? Christopher <cpisz@austin.rr.com> - 2011-09-06 12:35 -0700
        Re: does virtualizing all methods slow C++ down ? "Kevin P. Fleming" <news@kpfleming.us> - 2011-09-06 16:41 -0500
      Re: does virtualizing all methods slow C++ down ? "Balog Pal" <pasa@lib.hu> - 2011-09-11 12:02 +0200
    Re: does virtualizing all methods slow C++ down ? "Thomas J. Gritzan" <phygon_antispam@gmx.de> - 2011-09-07 02:42 +0200
      Re: does virtualizing all methods slow C++ down ? Noah Roberts <roberts.noah@gmail.com> - 2011-09-07 07:39 -0700
        Re: does virtualizing all methods slow C++ down ? Richard Damon <news.x.richarddamon@xoxy.net> - 2011-09-07 11:46 -0400
    Re: does virtualizing all methods slow C++ down ? Nobody <nobody@nowhere.com> - 2011-09-07 01:45 +0100
  Re: does virtualizing all methods slow C++ down ? Alain Ketterlin <alain@dpt-info.u-strasbg.fr> - 2011-09-06 19:12 +0200
    Re: does virtualizing all methods slow C++ down ? Lynn McGuire <lmc@winsim.com> - 2011-09-06 14:13 -0500
      Re: does virtualizing all methods slow C++ down ? jacob navia <jacob@spamsink.net> - 2011-09-06 21:57 +0200
        Re: does virtualizing all methods slow C++ down ? Alain Ketterlin <alain@dpt-info.u-strasbg.fr> - 2011-09-06 22:40 +0200
          Re: does virtualizing all methods slow C++ down ? Lynn McGuire <lmc@winsim.com> - 2011-09-06 16:13 -0500
            Re: does virtualizing all methods slow C++ down ? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2011-09-07 08:01 +0000
              Re: does virtualizing all methods slow C++ down ? Lynn McGuire <lmc@winsim.com> - 2011-09-14 12:08 -0500
                Re: does virtualizing all methods slow C++ down ? Lynn McGuire <lmc@winsim.com> - 2011-09-14 14:04 -0500
        Re: does virtualizing all methods slow C++ down ? Lynn McGuire <lmc@winsim.com> - 2011-09-06 16:11 -0500
          Re: does virtualizing all methods slow C++ down ? Paavo Helde <myfirstname@osa.pri.ee> - 2011-09-07 11:44 -0500
            Re: does virtualizing all methods slow C++ down ? Joshua Maurice <joshuamaurice@gmail.com> - 2011-09-07 14:11 -0700
              Re: does virtualizing all methods slow C++ down ? Paavo Helde <myfirstname@osa.pri.ee> - 2011-09-08 01:26 -0500
              Re: does virtualizing all methods slow C++ down ? Nobody <nobody@nowhere.com> - 2011-09-08 07:37 +0100
              Re: does virtualizing all methods slow C++ down ? zindorsky <nephi.allred@gmail.com> - 2011-09-09 10:20 -0700
                Re: does virtualizing all methods slow C++ down ? "Kevin P. Fleming" <news@kpfleming.us> - 2011-09-10 11:48 -0400
                Re: does virtualizing all methods slow C++ down ? Nobody <nobody@nowhere.com> - 2011-09-10 17:58 +0100
  Re: does virtualizing all methods slow C++ down ? Juha Nieminen <nospam@thanks.invalid> - 2011-09-07 06:19 +0000
  Re: does virtualizing all methods slow C++ down ? Goran <goran.pusic@gmail.com> - 2011-09-06 23:52 -0700
    Re: does virtualizing all methods slow C++ down ? Juha Nieminen <nospam@thanks.invalid> - 2011-09-07 07:57 +0000
      Re: does virtualizing all methods slow C++ down ? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2011-09-07 10:39 +0000
      Re: does virtualizing all methods slow C++ down ? Goran <goran.pusic@gmail.com> - 2011-09-07 04:30 -0700
        Re: does virtualizing all methods slow C++ down ? Juha Nieminen <nospam@thanks.invalid> - 2011-09-07 13:04 +0000
          Re: does virtualizing all methods slow C++ down ? Goran <goran.pusic@gmail.com> - 2011-09-07 06:18 -0700
  Re: does virtualizing all methods slow C++ down ? ld <laurent.deniau@gmail.com> - 2011-09-08 05:34 -0700

csiph-web