From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Thread: a07f3367d7,d275ad890e7b3ecc
X-Google-Attributes: gida07f3367d7,public,usenet
X-Google-NewGroupId: yes
X-Google-Language: ENGLISH,ASCII-7-bit
Received: by 10.180.94.98 with SMTP id db2mr98617wib.2.1348190018645;
        Thu, 20 Sep 2012 18:13:38 -0700 (PDT)
Path: 
 q11ni3492521wiw.1!nntp.google.com!feeder1.cambriumusenet.nl!feed.tweaknews.nl!85.12.40.131.MISMATCH!xlned.com!feeder3.xlned.com!news.astraweb.com!border6.a.newsrouter.astraweb.com!border4.nntp.ams.giganews.com!border2.nntp.ams.giganews.com!border3.nntp.ams.giganews.com!border1.nntp.ams.giganews.com!nntp.giganews.com!newsreader4.netcologne.de!news.netcologne.de!goblin2!goblin.stu.neva.ru!aioe.org!eternal-september.org!feeder.eternal-september.org!mx04.eternal-september.org!.POSTED!not-for-mail
From: Simon Wright <simon@pushface.org>
Newsgroups: comp.lang.ada
Subject: Re: problem with Real_Matrix*Real_Matrix
Date: Tue, 18 Sep 2012 11:50:33 +0100
Organization: A noiseless patient Spider
Message-ID: <m24nmv61vq.fsf@pushface.org>
References: <cf82061e-ebae-46fc-b8d3-26c6dc9f84d9@googlegroups.com>
	<31325794-475b-430c-a419-a96da29251f0@googlegroups.com>
Mime-Version: 1.0
Injection-Info: mx04.eternal-september.org;
 posting-host="a9272c9c584810f74e3f95ce46cbeceb";
	logging-data="4662"; mail-complaints-to="abuse@eternal-september.org";
	posting-account="U2FsdGVkX1/ZbmPBDhcgMcwi69fI1y6vGiZuBTGFncU="
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.2 (darwin)
Cancel-Lock: sha1:Z/7HT2Oiuri7fPIgVCBFkEZX1Go=
	sha1:9rVzs6eFifvHIEcONicI3RVqOFY=
Content-Type: text/plain
Date: 2012-09-18T11:50:33+01:00
List-Id: <comp.lang.ada>

Anatoly Chernyshev <achernyshev@gmail.com> writes:

> Got the same mistake on Win7/GNAT 2012. As I'm doing matrix
> calculations too, I had to investigate, and here's the culprit from
> s-gearop.adb:
> ---------------------------------------------------------------
> function Vector_Matrix_Product (Left : Left_Vector; Right : Matrix)
> return Result_Vector is begin return R : Result_Vector (Right'Range
> (2)) do if Left'Length /= Right'Length (2) then raise Constraint_Error
> with "incompatible dimensions in vector-matrix multiplication"; end
> if;
>
>          for J in Right'Range (2) loop
>             declare
>                S : Result_Scalar := Zero;
>
>             begin
>                for K in Right'Range (1) loop
>                   S := S + Left (J - Right'First (1)
>                                    + Left'First) * Right (K, J);
>                end loop;
>
>                R (J) := S; end; end loop; end return; end
> Vector_Matrix_Product;
> --------------------------------------------------------------- 
> S := S + Left (J - Right'First (1) + Left'First) * Right (K, J); - in
> this fragment Left is kept constant within the K-cycle. Replacing
> Left'First with K would correct the problem.
>
> I believe, it was straightened for some reason for the Mac
> distribution; would be interesting to compare.

You're right about the location of the problem, but I think the fix is
instead to change

                  S := S + Left (J - Right'First (1)
to
                  S := S + Left (K - Right'First (1)

I've filed a bug in the GCC Bugzilla against GCC 4.7.0:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54614