Prev: C preprocessor
Next: weird g77 and gfortran
From: Thomas Koenig on 5 May 2008 18:28
Steven G. Kargl <kargl(a)troutmask.apl.washington.edu> wrote:

> Thomas Koenig <tkoenig(a)netcologne.de> writes:
>> Steven G. Kargl <kargl(a)troutmask.apl.washington.edu> wrote:
>>> http://pcc.ludd.ltu.se/news/
>>
>> Any timeline for F 95 and F 2003? :-)
>
> Not that I know. :-O

[...]

> and that there is no official f77 maintainer.

Sounds like a fun project... if anybody wanted to do this, he
or she could write the compiler in Fortran, use C binding for
interface to the pcc backend, and use gfortran or g95 for
bootstrap.

And no, I am most definitely NOT volunteering :-)
From: Steven G. Kargl on 5 May 2008 19:10
In article <slrng1v2gg.78k.tkoenig(a)meiner.onlinehome.de>,
Thomas Koenig <tkoenig(a)netcologne.de> writes:
> Steven G. Kargl <kargl(a)troutmask.apl.washington.edu> wrote:
>
>> Thomas Koenig <tkoenig(a)netcologne.de> writes:
>>> Steven G. Kargl <kargl(a)troutmask.apl.washington.edu> wrote:
>>>> http://pcc.ludd.ltu.se/news/
>>>
>>> Any timeline for F 95 and F 2003? :-)
>>
>> Not that I know. :-O
>
> [...]
>
>> and that there is no official f77 maintainer.
>
> Sounds like a fun project... if anybody wanted to do this, he
> or she could write the compiler in Fortran, use C binding for
> interface to the pcc backend, and use gfortran or g95 for
> bootstrap.
>
> And no, I am most definitely NOT volunteering :-)

My exchanges with Anders has given him an extra boost of energy.
In 2 days, he's gotten f77 to go from compiling 21 of 192 NIST
programs to 109 programs. In ab execution test, 5 programs
segfault due to the ancient runtime library. I just sent Anders
a patch that fixes the seqfaults. Unfortunately, the tests are
only partially passed for these 5 files. Four others die with an
abort().


--
Steve
http://troutmask.apl.washington.edu/~kargl/
 | 
Pages: 1
Prev: C preprocessor
Next: weird g77 and gfortran