Version française
Home     About     Download     Resources     Contact us    
Browse thread
[Benchmark] NBody
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: -- (:)
From: Will M. Farr <farr@M...>
Subject: Re: [Caml-list] [Benchmark] NBody
When I ran it on my system using Shark (a Mac OS X profiling 
application that doesn't require hooks in the app itself to get the 
information), the breakdown of the significant values was

48.1% in advance
34% in __sqrt (system function)
+ random stuff

I'm not sure, in light of this, why the aggressive inlining makes any 
difference.

Will

On 7 Feb 2005, at 2:36 PM, Christophe TROESTLER wrote:

> On Mon, 7 Feb 2005, "Will M. Farr" <farr@MIT.EDU> wrote:
>>
>> You might try profiling (using gprof); maybe it will give you an
>> idea where your time is being spent.
>
> I did that but I could not see anything: the important spot reads:
>
> -----------------------------------------------
>                 8.05    0.00 1000000/1000000     camlNbody__entry [5]
> [6]     99.7    8.05    0.00 1000000         camlNbody__code_begin [6]
> -----------------------------------------------
>
> Did I miss something???
>
>> it looks like the java code is faster, but it clearly has a large
>> startup time.
>
> I thought that.  Still, I'd like to know whether there is a way to
> make Caml code that fast or if not why.
>
> Thanks for youe reply,
> ChriS