Re: buildfarm - make check failures for leveret on 8.0 - Mailing list pgsql-hackers

From Andrew Dunstan
Subject Re: buildfarm - make check failures for leveret on 8.0
Date
Msg-id 44D789E1.3000100@dunslane.net
Whole thread Raw
In response to Re: buildfarm - make check failures for leveret on 8.0  (Jeremy Drake <pgsql@jdrake.com>)
List pgsql-hackers
Jeremy Drake wrote:
> On Mon, 7 Aug 2006, Tom Lane wrote:
>
>   
>> Stefan Kaltenbrunner <stefan@kaltenbrunner.cc> writes:
>>     
>>> *) why the large difference in the build-flags ?
>>>       
>> CVS HEAD configure.in knows about icc and the release branches don't.
>> I think the changes were only put into HEAD because of lack of testing,
>> but if we have buildfarm coverage I think it'd be OK to back-port the
>> configure logic to the prior branches.  Any objections?
>>     
>
> I sent the original patch.  I just sent it for HEAD because a) I could
> still deal with previous branches by editing Makefile.global by hand after
> configure, b) I reconfigured older branches seldom enough compared to
> HEAD that it didn't bother me nearly as much, and c) I figured it would be
> more readily accepted into HEAD than trying to get it back-ported.  Also I
> was not sure about the acceptance of such things into back branches, since
> it may be interpreted that supporting a new compiler is a "new feature"
> and most projects don't like to add new features to old releases.
>
>   

In general, yes, but I think if we support a new compiler it would be 
sensible to support at least the latest stable branch if possible, 
especially if that mainly means extra configuration rather than changing 
the code.

I'd be happy to see the config changes backported.

cheers

andrew


pgsql-hackers by date:

Previous
From: Jose Orlando Pereira
Date:
Subject: Re: standard interfaces for replication providers
Next
From: Alvaro Herrera
Date:
Subject: Re: CSStorm occurred again by postgreSQL8.2