Sent: Friday, October 07, 2016 at 6:26 PM
From: "Bruce Dubbs" <bruce.du...@gmail.com>
To: "BLFS Support List" <blfs-support@lists.linuxfromscratch.org>
Subject: Re: [blfs-support] Solid failure in Frameworks-5.25.0
Cliff McDiarmid wrote:
> Sent: Friday, October 07, 2016 at 11:01 AM
> From: "Cliff McDiarmid" <cliffhan...@gardener.com>
> To: blfs-support@lists.linuxfromscratch.org
> Subject: Re: [blfs-support] Solid failure in Frameworks-5.25.0
> Sent: Thursday, October 06, 2016 at 1:05 PM
> From: akhiezer <lf...@cruziero.com>
> To: "BLFS Support List" <blfs-support@lists.linuxfromscratch.org>
> Subject: Re: [blfs-support] Solid failure in Frameworks-5.25.0
>> From: "Cliff McDiarmid" <cliffhan...@gardener.com>
>> Date: Thu, 6 Oct 2016 12:17:19 +0200
>> Subject: Re: [blfs-support] Solid failure in Frameworks-5.25.0
>>
>> Sent: Wednesday, October 05, 2016 at 11:29 PM
>> From: "Ken Moffat" <zarniwh...@ntlworld.com>
>> To: "BLFS Support List" <blfs-support@lists.linuxfromscratch.org>
>> Subject: Re: [blfs-support] Solid failure in Frameworks-5.25.0
>> On Wed, Oct 05, 2016 at 09:50:30PM +0100, akhiezer wrote:
>>>> From: "Cliff McDiarmid" <cliffhan...@gardener.com>
>>>> Date: Wed, 5 Oct 2016 20:35:36 +0200
>>>> Subject: [blfs-support] Solid failure in Frameworks-5.25.0
>>>>
>>>> Hi
>>>>
>>>> Does anyone know what's causing this when building Solid as part of 
>>>> Frameworks? Or has anyone seen this before?
>>>>
>>>>
>>>> [ 40%] Building C object 
>>>> src/solid/CMakeFiles/KF5Solid_static.dir/predicate_lexer.c.o
>>>> /mnt/home/cliffys/frameworks-5.26.0/solid-5.25.0/build/src/solid/predicate_lexer.c:
>>>>  In function âSolidensure_buffer_stackâ:
>>>> /mnt/home/cliffys/frameworks-5.26.0/solid-5.25.0/build/src/solid/predicate_lexer.c:1587:21:
>>>>  error: C++ style comments are not allowed in ISO C90
>>>> num_to_alloc = 1; // After all that talk, this was set to 1 anyways...
>>>> ^
>>>
>>> Dealing with the info 'in isolation' - haven't gone back and looked at book 
>>> pages:
>>>
>>> either:
>>> ====
>>> * change that line, in:
>>> /mnt/home/cliffys/frameworks-5.26.0/solid-5.25.0/build/src/solid/predicate_lexer.c
>>> , to:
>>> num_to_alloc = 1; /* After all that talk, this was set to 1 anyways... */
>>> But you might hit similar issue in other files: so ref following option, 
>>> below.
>>>
>>> * or:
>>> tell the compiler to use an appropriate C version later than 'ISO C90'; or
>>> at least to be relaxed about such '// ...' comment styles.
>>> ====
>>>
>>
>>> Yes, it seems strange that it is setting things up as if for an old
>>> compiler.
>>
>> Not exactly old - Solid-5.25.0 on mine reported
>>
>> -- The C compiler identification is GNU 6.1.0
>> -- The CXX compiler identification is GNU 6.1.0
>> -- Check for working C compiler: /usr/bin/cc
>> -- Check for working C compiler: /usr/bin/cc -- works
>> -- Detecting C compiler ABI info
>> -- Detecting C compiler ABI info - done
>> -- Detecting C compile features
>> -- Detecting C compile features - done
>> -- Check for working CXX compiler: /usr/bin/c++
>> -- Check for working CXX compiler: /usr/bin/c++ -- works
>> -- Detecting CXX compiler ABI info
>> -- Detecting CXX compiler ABI info - done
>> -- Detecting CXX compile features
>> -- Detecting CXX compile features - done
>>
>>
>>> In 7.9-rc1 I had similar output with 5.18, except the compiler was
>>> 5.3.0. I can't go back and compare that source to 5.25.0 because
>>> http://download.kde.org/stable/frameworks/ has nothing before 5.20,
>>> but I will be very surprised if '//' comments are a recent addition.
>>
>>> So, which compiler are you using, and what does cmake think it is
>>> when trying to build solid ? Or did you accidentally use some
>>> CFLAGS or CXXFLAGS to force the version ?
>>
>> I would have thought GCC 6.1 would have been fine. Not using CFLAGS or 
>> CXXFLAGS, I use to use them don't anymore.
>>
>
>
>> It's not so much the gcc version per se: it's which 'C' standard it's
>> trying to conform to - ref e.g.:
>> * 
>> https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html[https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html][https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html[https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html]][https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html[https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html][https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html[https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html]]]
>> (& sub-ref: 
>> https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards[https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards][https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards[https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards]][https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards[https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards][https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards[https://gcc.gnu.org/onlinedocs/gcc/Standards.html#Standards]]])
>> - see the top two items, '-ansi' & '-std='; & incl the examples they
>> give re '//'-style comments.
> ---
>> Basically, you may have some command-line switch &/or environment-var
>> that is reaching as far as the gcc cmdline, and causing the problem.

> Just as an aside. Are these sort of messages typical when building KF5? They 
> appear in various forms in many packages.

>Warnings are not a problem. They often occur because the compiler used
>upstream is a little older than the one we use. You really only need to
>be concerned about error messages.

> I can find nothing in envs or in profile etc that would account for this.  
> Have made the alteration in the file in question and am going forward on that 
> basis at the moment.
> Incidently, Solid 5.25.0 is also failing at that stage and yet I know I 
> compiled this several months ago without errors.

>Which version of flex are you using,

>predicate_lexer.c is generated by flex from a template

>if you are using 2.6.0 , it needs a patch attached(courtesy of Arch), 

>2.6.1 is ok

>Martin

That's it.  Upgraded to Flex 2.6.1 and gcc behaves.  Many thanks Martin and to 
everyone else.

Cliff


-- 
http://lists.linuxfromscratch.org/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to