[Python-Dev] cpython and parallel make

[Jonas Wagner]

Hi again,

the attached Makefile patch seems to fix the parallel build problems.

Is there a chance of getting this into trunk? Should I open an issue
or send the patch somewhere else?

Cheers,
Jonas

On Fri, Sep 5, 2014 at 12:15 PM, Jonas Wagner <jonas.wagner at epfl.ch> wrote:
>>> > Would people be interested in having a parallel version?
>>>
>>> See http://bugs.python.org/issue5309
>>
>> Cool! I'll look into this.
>
> The patch there works well for me. I've made one small update, and
> submitted the new version in the bug tracker.
>
> Regarding the other build problem, I might have found some hint:
> - Parser/pgen.o ends up in both the PARSER_OBJS and PGENOBJS variables
> in the Makefile
> - PARSER_OBJS is depended upon in a few places, hence it could be that
> make starts to build Parser/pgen.o
> - PGENOBJS is built when building PGEN, which happens *in a different
> make that is called recursively*
>
> I think the culprit is the rule for GRAMMAR_H which calls make
> recursively. Is there a reason that GRAMMAR_H has to generate PGEN
> like this? Couldn't it just depend on PGEN?
>
> Cheers,
> Jonas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: makefile_parallel.patch
Type: text/x-patch
Size: 1349 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-dev/attachments/20140905/c71a661f/attachment.bin>