[vtk-developers] BandedPolyDataContourFilter failes to compile on Ubuntu 14

[Robert Maynard]

VTK has documented that it supports gcc 4.8.3+ (In the readme, in the
buildsystem, and in the code). and therefore we better do so.

In the past when updating minimum compiler versions have done so
through first verifying that the compilers can be updated without
major inconvenience to users. I haven't seen anyone do a survey for
dropping GCC 4.8. Are we happy dropping the default compiler for the
latest Red Hat Enterprise release?

On Tue, Feb 5, 2019 at 4:40 PM Bill Lorensen <bill.lorensen at gmail.com> wrote:
>
> I saw the MR. My question is why are we supporting an old compiler. I think that's what the argument against my first post in this thread.
>
> On Tue, Feb 5, 2019, 1:34 PM Sankhesh Jhaveri <sankhesh.jhaveri at kitware.com wrote:
>>
>> FWIW, I have a MR here: https://gitlab.kitware.com/vtk/vtk/merge_requests/5157
>>
>>
>> On Tue, Feb 5, 2019 at 4:32 PM Robert Maynard <robert.maynard at kitware.com> wrote:
>>>
>>> I meant GCC 4.8.3/5 is a supported compiler by VTK not GCC itself.
>>>
>>> On Tue, Feb 5, 2019 at 4:29 PM Bill Lorensen <bill.lorensen at gmail.com> wrote:
>>> >
>>> > Gcc 4.8 is no longer maintained. See
>>> > https://www.gnu.org/software/gcc/gcc-4.8/
>>> >
>>> >
>>> >
>>> > On Tue, Feb 5, 2019, 1:27 PM Robert Maynard <robert.maynard at kitware.com wrote:
>>> >>
>>> >> I wasn't paying attention to the thread. GCC 4.8.3 is a supported
>>> >> compiler and until we remove it we have to accept that we need to
>>> >> support it.
>>> >>
>>> >> On Tue, Feb 5, 2019 at 4:21 PM Bill Lorensen <bill.lorensen at gmail.com> wrote:
>>> >> >
>>> >> > What happened to the arguments against my complaints about this error?
>>> >> >
>>> >> > Bill
>>> >> >
>>> >> > On Tue, Feb 5, 2019, 9:23 AM Sankhesh Jhaveri <sankhesh.jhaveri at kitware.com wrote:
>>> >> >>
>>> >> >> Sweet, thanks Rob,
>>> >> >>
>>> >> >>
>>> >> >> On Tue, Feb 5, 2019 at 11:31 AM Robert Maynard <robert.maynard at kitware.com> wrote:
>>> >> >>>
>>> >> >>> Here is a drop-in wrapper for rotate that has the C++11 behavior but
>>> >> >>> also works with GCC 4.8 ( https://godbolt.org/z/1aVjiZ ).
>>> >> >>>
>>> >> >>> On Tue, Feb 5, 2019 at 10:29 AM Bill Lorensen <bill.lorensen at gmail.com> wrote:
>>> >> >>> >
>>> >> >>> > I don't have a fix. Best if you can look.
>>> >> >>> >
>>> >> >>> > I've built gcc 8.2 do I don't have a problem.
>>> >> >>> >
>>> >> >>> > Bill
>>> >> >>> >
>>> >> >>> > On Tue, Feb 5, 2019, 7:20 AM Sankhesh Jhaveri <sankhesh.jhaveri at kitware.com wrote:
>>> >> >>> >>
>>> >> >>> >> Bill,
>>> >> >>> >>
>>> >> >>> >> It would be great if you have a fix, neverthless. A customer noticed the same issue on Red Hat GCC v4.8.5 20150623. The README says VTK requires gcc  v4.8.3 or newer.
>>> >> >>> >>
>>> >> >>> >> Let me know if you would rather I look into a fix for it.
>>> >> >>> >>
>>> >> >>> >> Thanks,
>>> >> >>> >> Sankhesh
>>> >> >>> >>
>>> >> >>> >>
>>> >> >>> >> On Tue, Feb 5, 2019 at 1:27 AM Bill Lorensen <bill.lorensen at gmail.com> wrote:
>>> >> >>> >>>
>>> >> >>> >>> Sean,
>>> >> >>> >>>
>>> >> >>> >>> Thanks for the suggestion. I was able to build gcc 8.2 on my Ubuntu
>>> >> >>> >>> 14.04 system. Only a little tricky. Took several hours to compile.
>>> >> >>> >>>
>>> >> >>> >>> I have successfully built VTK HEAD and the VTKExamples. All VTK tests passed.
>>> >> >>> >>>
>>> >> >>> >>> No more bitching from me (about this topic anyway0>
>>> >> >>> >>>
>>> >> >>> >>> The compiler did expose some new warnings which I'll fix.
>>> >> >>> >>>
>>> >> >>> >>> Bill
>>> >> >>> >>>
>>> >> >>> >>> On Mon, Feb 4, 2019 at 8:38 AM Sean McBride <sean at rogue-research.com> wrote:
>>> >> >>> >>> >
>>> >> >>> >>> > On Thu, 31 Jan 2019 20:37:32 -0500, David Thompson via vtk-developers said:
>>> >> >>> >>> >
>>> >> >>> >>> > >I'm not sure it is a good idea to support old compilers and distributions:
>>> >> >>> >>> > >
>>> >> >>> >>> > >1. Ubuntu 14.04 LTS will lose support from Canonical in 3 months. I
>>> >> >>> >>> > >would be surprised if what is in git master is released by then. Why
>>> >> >>> >>> > >should we support platforms no longer supported by their distributor?
>>> >> >>> >>> >
>>> >> >>> >>> > Is the compiler so tied to the OS?  Bill, can't you upgrade your compiler without updating your whole OS?
>>> >> >>> >>> >
>>> >> >>> >>> > >2. If we keep forcing workarounds, the next time something breaks, then
>>> >> >>> >>> > >*it* will be the only compiler error. That pattern just keeps accruing
>>> >> >>> >>> > >technical debt. Ultimately, I believe that technical debt will push
>>> >> >>> >>> > >developers away from VTK so I want to avoid it.
>>> >> >>> >>> > >
>>> >> >>> >>> > >3. Not too long ago, it was common practice for systems in production
>>> >> >>> >>> > >use to never get patched (lest the patch break what was working).
>>> >> >>> >>> >
>>> >> >>> >>> > Not that we (programmers collectively) have solved the problem of patches always breaking things that work.  It's still the norm.  :(
>>> >> >>> >>> >
>>> >> >>> >>> > >However, security issues and increased connectivity prevalent today has
>>> >> >>> >>> > >pretty much put a stop to this practice and I think that translates into
>>> >> >>> >>> > >a much weaker argument for supporting older systems.
>>> >> >>> >>> >
>>> >> >>> >>> > I largely agree, but there are exceptions.
>>> >> >>> >>> >
>>> >> >>> >>> > Perhaps less on linux, but newer versions of macOS often drop support for older hardware.  So it's not always easy or cheap to just keep all software up-to-date.
>>> >> >>> >>> >
>>> >> >>> >>> > Cheers,
>>> >> >>> >>> >
>>> >> >>> >>> > --
>>> >> >>> >>> > Sean
>>> >> >>> >>> >
>>> >> >>> >>> >
>>> >> >>> >>>
>>> >> >>> >>>
>>> >> >>> >>> --
>>> >> >>> >>> Unpaid intern in BillsParadise at noware dot com
>>> >> >>> >>> _______________________________________________
>>> >> >>> >>> Powered by www.kitware.com
>>> >> >>> >>>
>>> >> >>> >>> Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
>>> >> >>> >>>
>>> >> >>> >>> Search the list archives at: http://markmail.org/search/?q=vtk-developers
>>> >> >>> >>>
>>> >> >>> >>> Follow this link to subscribe/unsubscribe:
>>> >> >>> >>> https://vtk.org/mailman/listinfo/vtk-developers
>>> >> >>> >>>
>>> >> >>> >> --
>>> >> >>> >>
>>> >> >>> >> Sankhesh Jhaveri
>>> >> >>> >>
>>> >> >>> >> Sr. Research & Development Engineer | Kitware | (518) 881-4417
>>> >> >>> >
>>> >> >>> > _______________________________________________
>>> >> >>> > Powered by www.kitware.com
>>> >> >>> >
>>> >> >>> > Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
>>> >> >>> >
>>> >> >>> > Search the list archives at: http://markmail.org/search/?q=vtk-developers
>>> >> >>> >
>>> >> >>> > Follow this link to subscribe/unsubscribe:
>>> >> >>> > https://vtk.org/mailman/listinfo/vtk-developers
>>> >> >>> >
>>> >> >>
>>> >> >> --
>>> >> >>
>>> >> >> Sankhesh Jhaveri
>>> >> >>
>>> >> >> Sr. Research & Development Engineer | Kitware | (518) 881-4417
>>
>> --
>>
>> Sankhesh Jhaveri
>>
>> Sr. Research & Development Engineer | Kitware | (518) 881-4417