gm2
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Catching up, some old some new problems. test errors?


From: Gaius Mulley
Subject: Re: Catching up, some old some new problems. test errors?
Date: Sun, 03 May 2020 12:43:35 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux)

Dragiša Durić <address@hidden> writes:

> Hi,
>
> I am starting to compile my old codebase, a work I tried first in
> early phases of gm2 dev... after first pass I am getting a lot of
>
> implementation restriction, array is too large
>
> and one example, maybe telling, is this:
>
>  gm2 -fiso -c Sort.mod                                                       
>                                             
> MakeArgs.def:24:1: error implementation restriction, array is too large 
> MakeArgs.def:24:1: error implementation restriction, array is too large 
>
>  ~/bluewd-m2/gm2.lib0  cat MakeArgs.def                                     
>                                                            
> DEFINITION MODULE FOR "C" MakeArgs;
>
> FROM
>   SYSTEM
> IMPORT
>   ADDRESS;
>
> EXPORT QUALIFIED Args, Do, ByNumber, Free;
>
> TYPE
>   Args = POINTER TO RECORD
>     format: POINTER TO ARRAY [0..511] OF CHAR;
>     count: CARDINAL; (* SIZE() == SIZE(ADDRESS) *)
>     params: ARRAY [0..31] OF ADDRESS;
>   END;
>
> PROCEDURE Do (format: ARRAY OF CHAR; ...): Args;
>
> PROCEDURE ByNumber(argCount: CARDINAL; ...): Args;
>
> PROCEDURE Free (args: Args);
>
> END MakeArgs.
>
> Sort imports some modules, one is MakeArgs. I do not have other errors
> in few tens more source files I have in this lib0 project. Is this
> something known/fixable from my side?
>
> Also, I built gm2 on RPi 4 (w/ 4GB RAM). It went fairly well, had to
> add --build. When I ran make check-gm2, I had, IIRC, some 40 errors on
> x864_64 box and some 68 on RPi. Is this expected or I did somethin
> wrong ith building?
>
> TIA,
> dd

Hi Dragiša,

this is an interesting error - I've not seen an array too large error.
Out of interest which version of gm2 are you using?
I get 1 error from building gm2 master branch on a x86_64 debian stretch
system and no errors using gm2 master on aarch64 debian stretch.

Is there any chance the code can be reduced down to an email-able test
case? - I'd be interested to fix the bug,

regards,
Gaius



reply via email to

[Prev in Thread] Current Thread [Next in Thread]