svn commit: r384521 - head/Mk

[John Marino]

On 4/23/2015 13:06, John Marino wrote:
> On 4/22/2015 23:29, Gerald Pfeifer wrote:
>> Author: gerald
>> Date: Wed Apr 22 21:29:21 2015
>> New Revision: 384521
>> URL: https://svnweb.freebsd.org/changeset/ports/384521
>>
>> Log:
>>   Since there is not going to be any new version of GCC in the FreeBSD
>>   base system ever again, simplify the GCCVERSION table and logic to not
>>   worry about minimum system versions carrying a certain version of GCC.
>>   
>>   This also removes the _GCCVERSION_${v}_R variables and simplifies some
>>   logic and debug output.
>>
>> Modified:
>>   head/Mk/bsd.gcc.mk
>>
> 
> This look like "for fun" change, like several before it on bsd.gcc.mk.
> This is the most heavily patched Mk file for dports and most changes to
> it break dports.  DragonFly does indeed have GCC in base (two of them
> actually) and probably will for a long time to come.
> 
> I have not access the impacts yet, but this might be a sledgehammer to
> recover from.   In the future, could you at least give me a heads up and
> ideally put me on a phabric review for changes to anything compiler
> related?  In general, shouldn't bsd.gcc.mk go through a phabric review
> anyway?
> 
> Thanks.  I've been silently adjusting and not all the changes actually
> did anything except "simplify".
> 

Hi Gerald,
To follow up on this topic, I realized the situation was not
sustainable.  I wrote a replacement file called bsd.df.gcc.mk which gets
pulled in by bsd.port.mk instead of bsd.gcc.mk on DPorts.  As a bonus,
the makefile "code" is much clearer than bsd.gcc.mk and it's easy for me
to maintain (and basically static).

TLDR: DPorts doesn't use bsd.gcc.mk anymore, you can do what you want
with it without asking if it breaks DragonFly.

Thanks,
John