On 03/10/2016 01:22 AM, jhu wrote:
> "clipka" <nomail@nomail> wrote:
>> We've been having problems with the Unix build process in the past, often
>> related to the boost libraries. I'm wondering where we're at with these issues
>> right now with the current master branch.
>>
>> If the 3.7 build process has ever failed to work "out of the box" for you (i.e.
>> strictly following the baseline build instructions described in `unix/README`),
>> I'd love to hear from you:
>>
>> - What variant of Unix are you using?
>> - What version of the boost library are you using?
>>
>> - Does the official stable 3.7.0.0 build "out of the box" on your system?
>> Otherwise, what problems do you encounter, and how (if at all) do you manage to
>> get it to build?
>>
>> - Does the current master 3.7.1-alpha.8344773 build "out of the box" on your
>> system? Otherwise, what problems do you encounter, and how (if at all) do you
>> manage to get it to build?
>>
>> - Any other observations?
>>
>>
>> (Please do NOT reply if the 3.7 build process has always worked perfectly fine
>> for you.)
>
> Fails at the ./prebuild.sh in the unix/ directory. This is both on FreeBSD 10.2
> and Debian testing with the official stable 3.7.0.0 build. On both systems I
> get:
>
> configure.ac:302: the top level
> + cat ./configure
> + sed -e 's,configure.gnu  --help=recursive,& --srcdir=$ac_srcdir,g' -e 's,\(cd
> \)\($ac_\)\(pop\)*\(dir\),\1"\2\3\4",g' -e
> 's,$am_aux_dir/missing,\\"$am_aux_dir\\"/missing,g'
> + mv -f ./configure.tmp ./configure
> + chmod +x ./configure
> + rm -f -r ./autom4te.cache
> sed: ./Makefile.in: No such file or directory
>
> So, it's not generating Makefile.in for some reason.
>
>
>
Does the suggested fix at the bottom of git issue 54 fix things for your 
3.7.0-stable build ?

See:  https://github.com/POV-Ray/povray/issues/54

Aside:
-----
Christoph! I believe you can close git issue 54 as that user is 
compiling 3.7.0-stable OK.

Also a reminder one of the main reasons for staying with 3.7.0-stable is 
the 17-19% performance hit in 3.7.1. My working machines is currently 
just a 2 core i3... I confess to not having lately tested performance 
for 3.7.1. Have the developers at all addressed the performance issue?

Bill P.

Post a reply to this message