llvm dev - Feb 2016 - Can lit be upgraded to assume Python 2.7?

This sounds like a good idea to me!  I can’t think of any common platform where
you can’t get 2.7.  Lets get rid of that legacy cruft!

> On Feb 23, 2016, at 1:32 PM, Eric Christopher <echristo at gmail.com>
wrote:
> 
> Seems reasonable to me. Chris?
> 
> On Mon, Feb 22, 2016, 8:40 PM Sean Silva via llvm-dev <llvm-dev at
lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
> In the root llvm/CMakeLists.txt we have:
> 
> if( ${PYTHON_VERSION_STRING} VERSION_LESS 2.7 )
>   message(FATAL_ERROR "Python 2.7 or newer is required")
> endif()
> 
> But lit seems to still be stuck in a Python 2.5 world. For example,
detectCPUs is redundant now that we have multiprocessing.cpu_count()
(multiprocessing requires >=2.6). And there are a bunch of other Python 2.5
workarounds floating around inside lit. I'm actually not sure if there are
2.6 workarounds.
> 
> Anyway, does anybody know if somehow, despite the CMake check, lit is still
being run with Python <2.7 on any bots or anywhere? If not, I'd like to
make a couple cleanup patches.
> 
> -- Sean Silva
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
<http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20160223/591bc743/attachment.html>

Great! I'll circle around to this at some point. Despite the
"obvious"
nature of it I still am wary of underestimating the cruftiness of the
buildbots, so I'll probably do it some time at night when the bots are
mostly green so that I can easily see if any bots *are* broken by this.

-- Sean Silva

On Tue, Feb 23, 2016 at 6:55 PM, Chris Matthews <chris.matthews at
apple.com>
wrote:
> This sounds like a good idea to me!  I can’t think of any common platform
> where you can’t get 2.7.  Lets get rid of that legacy cruft!
>
>
> On Feb 23, 2016, at 1:32 PM, Eric Christopher <echristo at gmail.com>
wrote:
>
> Seems reasonable to me. Chris?
>
> On Mon, Feb 22, 2016, 8:40 PM Sean Silva via llvm-dev <
> llvm-dev at lists.llvm.org> wrote:
>
>> In the root llvm/CMakeLists.txt we have:
>>
>> if( ${PYTHON_VERSION_STRING} VERSION_LESS 2.7 )
>>   message(FATAL_ERROR "Python 2.7 or newer is required")
>> endif()
>>
>> But lit seems to still be stuck in a Python 2.5 world. For example,
>> detectCPUs is redundant now that we have multiprocessing.cpu_count()
>> (multiprocessing requires >=2.6). And there are a bunch of other
Python 2.5
>> workarounds floating around inside lit. I'm actually not sure if
there are
>> 2.6 workarounds.
>>
>> Anyway, does anybody know if somehow, despite the CMake check, lit is
>> still being run with Python <2.7 on any bots or anywhere? If not,
I'd like
>> to make a couple cleanup patches.
>>
>> -- Sean Silva
>> _______________________________________________
>> LLVM Developers mailing list
>> llvm-dev at lists.llvm.org
>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20160224/a83dc101/attachment.html>

On 24 February 2016 at 19:49, Sean Silva via llvm-dev
<llvm-dev at lists.llvm.org> wrote:
> Great! I'll circle around to this at some point. Despite the
"obvious"
> nature of it I still am wary of underestimating the cruftiness of the
> buildbots, so I'll probably do it some time at night when the bots are
> mostly green so that I can easily see if any bots *are* broken by this.
While at it, can you call "env python2" in the shebang line? It makes
it a lot easier for Python3 systems, like Arch.

cheers,
--renato

Still haven't gotten around to this, but just now I've randomly run into
`set(Python_ADDITIONAL_VERSIONS 2.7 2.6 2.5)` in compiler-rt/CMakeLists.txt
I assume that can be nuked too?

-- Sean Silva

On Wed, Feb 24, 2016 at 11:49 AM, Sean Silva <chisophugis at gmail.com>
wrote:
> Great! I'll circle around to this at some point. Despite the
"obvious"
> nature of it I still am wary of underestimating the cruftiness of the
> buildbots, so I'll probably do it some time at night when the bots are
> mostly green so that I can easily see if any bots *are* broken by this.
>
> -- Sean Silva
>
> On Tue, Feb 23, 2016 at 6:55 PM, Chris Matthews <chris.matthews at
apple.com>
> wrote:
>
>> This sounds like a good idea to me!  I can’t think of any common
platform
>> where you can’t get 2.7.  Lets get rid of that legacy cruft!
>>
>>
>> On Feb 23, 2016, at 1:32 PM, Eric Christopher <echristo at
gmail.com> wrote:
>>
>> Seems reasonable to me. Chris?
>>
>> On Mon, Feb 22, 2016, 8:40 PM Sean Silva via llvm-dev <
>> llvm-dev at lists.llvm.org> wrote:
>>
>>> In the root llvm/CMakeLists.txt we have:
>>>
>>> if( ${PYTHON_VERSION_STRING} VERSION_LESS 2.7 )
>>>   message(FATAL_ERROR "Python 2.7 or newer is required")
>>> endif()
>>>
>>> But lit seems to still be stuck in a Python 2.5 world. For example,
>>> detectCPUs is redundant now that we have
multiprocessing.cpu_count()
>>> (multiprocessing requires >=2.6). And there are a bunch of other
Python 2.5
>>> workarounds floating around inside lit. I'm actually not sure
if there are
>>> 2.6 workarounds.
>>>
>>> Anyway, does anybody know if somehow, despite the CMake check, lit
is
>>> still being run with Python <2.7 on any bots or anywhere? If
not, I'd like
>>> to make a couple cleanup patches.
>>>
>>> -- Sean Silva
>>> _______________________________________________
>>> LLVM Developers mailing list
>>> llvm-dev at lists.llvm.org
>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20160226/eb6c8582/attachment.html>