[m-rev.] diff: disable mfilterjavac by default

Julien Fischer jfischer at opturion.com
Mon Apr 29 13:46:13 AEST 2013



On Mon, 29 Apr 2013, Paul Bone wrote:

> On Mon, Apr 29, 2013 at 02:11:05AM +1000, Julien Fischer wrote:
>>
>> On Sun, 28 Apr 2013, Paul Bone wrote:
>>
>>> On Tue, Apr 23, 2013 at 05:41:43PM +1000, Julien Fischer wrote:
>>>> Disable mfilterjavac for now.
>>>>
>>>> mfilterjavac is currently causing problems when installing the java grade
>>>> from
>>>> the stage 2 dir.  It can be re-enabled when the following sequence of
>>>> actions
>>>> works correctly:
>>>>
>>>> (1) Build stage 1 using rotd-2013-04-18
>>>> (2) Bootcheck (or otherwise build the stage 2 compiler).
>>>> (3) cd stage2 && mmake install (with the java grade as one of the grades
>>>>     to install).
>>>>
>>>
>>> I've tested these steps. they work for me, the mfilterjavac tool is built
>>> and installed bin/.  I've built this ROTD from the current version of
>>> master.
>>>
>>> http://dl.mercurylang.org/rotd/mercury-srcdist-rotd-2013-04-27.tar.gz
>>
>> The above tarball has a few problems, the version string encoded in the
>> pre-generated source files is "version 12.08beta-mc1", not
>> "rotd-2013-04-27" -- you presumably left an old copy of VERSION lying
>> around.  Also we now build the source distribution in the hlc.gc.pregen
>> grade (i.e. HIGHLEVEL_CODE shoud equal "yes" in the comments at the
>> head of the pre-generated .c files).  Since you didn't use the pregen
>> grades here it's likely that this particular tarball is affected
>> by the problems that grade was intended to fix (i.e. it's broken).
>>
>> Can you please remove it from the download page.
>
> Done, google will take some time to refresh the distributed copies though.
> The cloud storage service is a CDN.

I've uploaded a correctly built ROTD (rotd-2013-04-29) to the Opturion
server.

> Can you document the correct process and put it in scripts/?  I followed the
> process in scripts/test_mercury, a comment there saying that that process is
> wrong would be helpful.

The procedure in test_mecury is correct, except that it doens't set the
grade.  I've updated it to do that.  (I'll send you the script I
currently use to build the ROTDs as well -- or, if I get time to clean
it up, add it to the repository.)

Cheers,
Julien.



More information about the reviews mailing list