[Ecls-list] Huge patches being committed

[Gabriel Dos Reis]

On Tue, Jun 9, 2009 at 6:10 AM, Marko Kocić<marko.kocic at gmail.com> wrote:
> 2009/6/9 Juan Jose Garcia-Ripoll <juanjose.garciaripoll at googlemail.com>:
>> On Tue, Jun 9, 2009 at 12:24 PM, Marko Kocić<marko.kocic at gmail.com> wrote:
>>> Let me pretend now that I am build farm, and I'm reporting build
>>> failure on mingw32. It fails to build multitheraded build with
>>> included gc and system gmp. gcc-4.4 also.
>>> Note that mingw build doesn't work for some time.
>>
>> Yes, I know about the problems. There are many issues involved. Some
>> people reported problems with Boehm-Weiser GC and mingw and I myself
>> have problems running the mingw build in the virtual machine: it takes
>> 10x more resources than the MSVC build and my computer is not really
>> powerful. That makes it difficult for me to work on that port.
>
>
> True. That's why I used system gmp on mingw. It drastically reduces build time.
>
> I know you are a one man team, so I hesitated to bother you with port
> that noone else seems to use before I tried do fix it myself, but I
> didn't get far.

OpenAxiom is interested in ECL on mingw to compete or replace GCL on mongw.
I suspect that now that the Sage people have replaced CLisp with ECL they
would also be using ECL on mingw.

>
> As I can see this it the main problem with ecl. It clearly needs more
> maintainers, since one man, no matter how talented he is, just can't
> do all by himself.

Definitely.  And definitely true of many projects -- some of which mentioned
above.

-- Gaby