svn commit: r395079 - in head/graphics: . mitsuba mitsuba/files

[John Marino]

On 8/24/2015 10:48 AM, Baptiste Daroussin wrote:
> On Mon, Aug 24, 2015 at 07:09:15AM +0000, Alexey Dokuchaev wrote:
>> On Sun, Aug 23, 2015 at 11:43:21PM +0100, Mathieu Arnold wrote:
>>> +--On 23 ao??t 2015 08:56:58 +0000 Alexey Dokuchaev <danfe at FreeBSD.org>
>>> wrote:
>>> | +# Unfortunately, the code is not GCC 4.2.1 compatible (due to bugs in GCC
>>> | +# itself).  Try to use system compiler if it is Clang, otherwise fallback
>>> | +# to modern GCC (on older versions of FreeBSD).
>>> | +.if ${OSVERSION} < 900014
>>> | +USE_GCC=	yes
>>>
>>> OSVERSION < 900014 will always be false.  Please, don't add more checks on
>>> unsupported versions.
>>
>> The fact that 8.X had reached its EOL merely means we cannot promise that
>> things won't break.  It does not mean we should immediately pull the plug
>> and remove two lines from the Makefile, leaving 8.4 users with a broken
>> port.
>>
>> It's up to maintainers to decide whether it's worth supporting some old
>> FreeBSD version.  Keeping around intricate patches and having to rebase
>> and test on every update is a nuisance, so those patches probably deserve
>> to rest in attic now.  Two lines in the Makefile are hardly a burden.
>>
>> I typically would try to support as many FreeBSD versions as long the
>> port stays manageable.  Often it's not a problem, and is very polite to
>> our users that do not want to upgrade for some reason (e.g. me and my
>> laptop are happy with stable/8 and see no reason to update, as it will
>> likely pessimize work experience without bringing any benefits over 8.4).
>>
>> That said, OSVERSION < 900014 will not always be false.  And why do you
>> want to be rude to our users and deliberately break things just because
>> we no longer promise that they will work?
>>
> 
> Even in that case, this is wrong anyway as you enforce GCC while should should
> require a modern compiler via compiler.mk.

Not to mention that it's pointless to support earlier than officially
supported platforms because everyone else is ripping out support every
time they touch a port and see it, actively.  In some cases, the only
change in the commit is removing support.

so that policy is:
1) contrary to what the rest of the devs are doing
2) ineffective (due to 1)

EOL means EOL.

John