Re: G95 is OK to *learn* Fortran?
From: Dr Chaos (mbkennelSPAMBEGONE_at_NOSPAMyahoo.com)
Date: 04/22/04
- Next message: Dr Chaos: "Re: G95 is OK to *learn* Fortran?"
- Previous message: Rich Townsend: "Re: Modules and shared libraries"
- In reply to: Joost VandeVondele: "Re: G95 is OK to *learn* Fortran?"
- Next in thread: Jos R Bergervoet: "Re: G95 is OK to *learn* Fortran?"
- Reply: Jos R Bergervoet: "Re: G95 is OK to *learn* Fortran?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 21 Apr 2004 23:44:44 +0000 (UTC)
On 16 Apr 2004 10:32:10 -0700, Joost VandeVondele <jv244@cam.ac.uk> wrote:
>> Neither one is making very quick progress, either.
>
> Ok 'useable' was defined rather strictly, I guess that might be the
> same for 'quick'. Some data as far as I know
> jan 2003: 'hello world' working
> dec 2003: LAPACK passing its testsuite
> apr 2004: several scientific codes working correctly. F90-ish programs
> like FMLIB or LAPACK95 Ok.
>
> My feeling is that progress is steady, and that's even better than
> quick.
>
> Clearly g95 doesn't yet compile my application correctly, but that's
> about the same status as the average commercial compiler reaches.
I tried the most recent g95 on my program, which uses a fair number of
g95 features. I've found three showstopper bugs so far in two hours. :(
(as in internal compiler error, failure to compiler legal program,
and erroneous value being returned with optimization)
Yup, sent in bug reports.
- Next message: Dr Chaos: "Re: G95 is OK to *learn* Fortran?"
- Previous message: Rich Townsend: "Re: Modules and shared libraries"
- In reply to: Joost VandeVondele: "Re: G95 is OK to *learn* Fortran?"
- Next in thread: Jos R Bergervoet: "Re: G95 is OK to *learn* Fortran?"
- Reply: Jos R Bergervoet: "Re: G95 is OK to *learn* Fortran?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|