Re: failure to build current on amd64

2020-01-09 Thread Riccardo Mottola

Hi,

Robert Swindells wrote:

I use a mix of them on different architectures to speed up the build and
they seem to work. I'm not seeing clang built for amd64.

Is the test for the value case-sensitive ?


I supposed so, since the code checks for != "no"

So I changed it, but with limited success, see my other bug report, with 
the failure of building libgomp (CVS checkout of yesterday)


Riccardo


Re: failure to build current on amd64

2020-01-09 Thread Robert Swindells


Greg Troxel  wrote:
>Edgar Pettijohn  writes:
>
>> I have tried the MKXXX=NO for lots of stuff and they always fail, so
>> I'm not sure what the correct way to use them is.
>
>I suspect that they fail because they are non-deafult.   The right way
>is to debug them and send a patch :-)   But seriously, reporting
>problems is helpful.

I use a mix of them on different architectures to speed up the build and
they seem to work. I'm not seeing clang built for amd64.

Is the test for the value case-sensitive ? 


Re: failure to build current on amd64

2020-01-09 Thread Greg Troxel
Edgar Pettijohn  writes:

> I have tried the MKXXX=NO for lots of stuff and they always fail, so I'm not 
> sure what the correct way to use them is.

I suspect that they fail because they are non-deafult.   The right way
is to debug them and send a patch :-)   But seriously, reporting
problems is helpful.


Re: failure to build current on amd64

2020-01-08 Thread Riccardo Mottola

Hi Edgar,

Edgar Pettijohn wrote:

MKLLVM = NO
MKLLVMRT = NO


yet I think it is building it anyway

Riccardo

I built current yesterday no problem.

I have tried the MKXXX=NO for lots of stuff and they always fail, so I'm not 
sure what the correct way to use them is.


I think I was able to "convince" it with:

MKLLVM = no
HAVE_LLVM=no
MKLLVMRT = no

now I get a failure in a different place, although compilation still 
happens with clang!


Riccardo


Re: failure to build current on amd64

2020-01-08 Thread Edgar Pettijohn

On Jan 8, 2020 6:38 AM, Riccardo Mottola  wrote:
>
> Hi!
>
> current fails do build userland for me:
>
> #   compile  libclangARCMigrate/Transforms.o
> /usr/src/obj/tooldir.NetBSD-9.99.34-amd64/bin/x86_64--netbsd-c++ 
> -frandom-seed=dedfa0f2 -O2 -march=core2 -Werror -Wno-stringop-overflow 
> -fPIE -march=core2 -std=c++14 -fno-rtti -fno-exceptions 
> -fno-strict-aliasing  -ffunction-sections -fdata-sections 
> --sysroot=/usr/src/obj/destdir.amd64 -I. 
> -I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/include
>  
> -I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/llvm/include
>  
> -I/usr/src/external/apache2/llvm/include/obj 
> -I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../config 
> -I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../config  -c 
>
> /usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/lib/ARCMigrate/Transforms.cpp
>  
> -o Transforms.o
> In file included from 
> /usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/lib/ARCMigrate/Transforms.cpp:18:
> /usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/include/clang/Sema/Sema.h:2566:42:
>  
> error: 'clang::Sema::DefaultedFunctionKind::Comparison' is too small to 
> hold all values of 'enum class clang::Sema::DefaultedComparisonKind' 
> [-Werror]
>   DefaultedComparisonKind Comparison : 8;
>    ^
> cc1plus: all warnings being treated as errors
>
> *** Failed target:  Transforms.o
>
>
> but I have a further question. I put in my /etc/mk.conf
>
> MKLLVM = NO
> MKLLVMRT = NO
>
>
> yet I think it is building it anyway
>
> Riccardo

I built current yesterday no problem.

I have tried the MKXXX=NO for lots of stuff and they always fail, so I'm not 
sure what the correct way to use them is.

Edgar

failure to build current on amd64

2020-01-08 Thread Riccardo Mottola

Hi!

current fails do build userland for me:

#   compile  libclangARCMigrate/Transforms.o
/usr/src/obj/tooldir.NetBSD-9.99.34-amd64/bin/x86_64--netbsd-c++ 
-frandom-seed=dedfa0f2 -O2 -march=core2 -Werror -Wno-stringop-overflow 
-fPIE -march=core2 -std=c++14 -fno-rtti -fno-exceptions 
-fno-strict-aliasing  -ffunction-sections -fdata-sections 
--sysroot=/usr/src/obj/destdir.amd64 -I. 
-I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/include 
-I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/llvm/include 
-I/usr/src/external/apache2/llvm/include/obj 
-I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../config 
-I/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../config  -c 

/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/lib/ARCMigrate/Transforms.cpp 
-o Transforms.o
In file included from 
/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/lib/ARCMigrate/Transforms.cpp:18:
/usr/src/external/apache2/llvm/lib/libclangARCMigrate/../../dist/clang/include/clang/Sema/Sema.h:2566:42: 
error: 'clang::Sema::DefaultedFunctionKind::Comparison' is too small to 
hold all values of 'enum class clang::Sema::DefaultedComparisonKind' 
[-Werror]

 DefaultedComparisonKind Comparison : 8;
  ^
cc1plus: all warnings being treated as errors

*** Failed target:  Transforms.o


but I have a further question. I put in my /etc/mk.conf

MKLLVM = NO
MKLLVMRT = NO


yet I think it is building it anyway

Riccardo