From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mail4-relais-sop.national.inria.fr (mail4-relais-sop.national.inria.fr [192.134.164.105])
	by walapai.inria.fr (8.13.6/8.13.6) with ESMTP id q32LFNP7018595
	for <caml-list@sympa-roc.inria.fr>; Mon, 2 Apr 2012 23:15:25 +0200
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AsQDAHz4eU/Unw4Ujmdsb2JhbABFpimRUyIBAQEBCQsJCRIGI4IJAQEFCAIwNBcBAwIJDgMEAQEBDRoHGSwBCQgCBAESCwUCCIdzB7pYinqFdASNa4kHkgaBWw
X-IronPort-AV: E=Sophos;i="4.75,358,1330902000"; 
   d="scan'208";a="138770901"
Received: from avasout08.plus.net ([212.159.14.20])
  by mail4-smtp-sop.national.inria.fr with ESMTP; 02 Apr 2012 23:15:24 +0200
Received: from WinEight ([87.114.112.81])
	by avasout08 with smtp
	id t9FN1i0031lRaPK019FPDs; Mon, 02 Apr 2012 22:15:24 +0100
X-CM-Score: 0.00
X-CNFS-Analysis: v=2.0 cv=I8Dntacg c=1 sm=1 a=SNaS5AyKB+UI9Kq+unAWQg==:17
 a=-tnAv0XnCDgA:10 a=Xub9RBUEA-sA:10 a=kj9zAlcOel0A:10 a=SGJpDYsEAAAA:8
 a=ZOzjf2MOAAAA:8 a=CjxXgO3LAAAA:8 a=PrCVO86orOh0kYKJay0A:9
 a=SdW9BTy6fIdeNNktFHUA:7 a=CjuIK1q_8ugA:10 a=dTth3pfvbawA:10
 a=1CkXkM7HNqP2Dw45:21 a=Ki-PFQ4HIrNQNUIM:21 a=SNaS5AyKB+UI9Kq+unAWQg==:117
From: "Jon Harrop" <jon@ffconsultancy.com>
To: "'David Allsopp'" <dra-news@metastack.com>,
        "'Hongbo Zhang'" <bobzhang1988@gmail.com>,
        "'Caml List'" <caml-list@inria.fr>
References: <4F79B858.2090502@gmail.com> <E51C5B015DBD1348A1D85763337FB6D9C28E6C07@Remus.metastack.local>
In-Reply-To: <E51C5B015DBD1348A1D85763337FB6D9C28E6C07@Remus.metastack.local>
Date: Mon, 2 Apr 2012 22:15:20 +0100
Message-ID: <024101cd1115$b6962110$23c26330$@ffconsultancy.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQKchNmV+F/TI7QyZT9tQ+Lj0w4DWwFb478hlN5Bx5A=
Content-Language: en-gb
Subject: RE: [Caml-list] How could I implement an efficient ring buffer?


The "mod" and the write barrier will significantly degrade performance vs C.
Probably faster to replace "mod" with if-based wrap around but there's
nothing you can do about the write barrier.

Cheers,
Jon.

> -----Original Message-----
> From: David Allsopp [mailto:dra-news@metastack.com]
> Sent: 02 April 2012 16:05
> To: Hongbo Zhang; Caml List
> Subject: RE: [Caml-list] How could I implement an efficient ring buffer?
> 
> Hongbo Zhang wrote:
> > Hi List,
> >     I want to implement sliding window algorithm (in place, no memory
> > copy), I wonder whether I need to write c code.
> >
> >     To make it clear and simple,
> >        In c, you can record the head pointer of the array, and do the
> > modulo operations when get and set
> >        In ocaml, it seems you have an array a of type int array,  you
> > can not do things like this *(&a+5).
> 
> Perhaps I'm missing something, but if you're planning on using arrays,
what's
> wrong with retrieving the item using the array index modulo the length of
the
> array?
> 
> i.e.
> 
> let a = [| ... or whatever ... |] in
> a.(5 mod Array.length a)
> 
> If accessing an array by index instead of by pointer worries you, then
you're
> looking at the wrong language ;o)
> 
> 
> David
> 
> 
> --
> Caml-list mailing list.  Subscription management and archives:
> https://sympa-roc.inria.fr/wws/info/caml-list
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs