> On Oct 14, 2016, at 11:07, Ryan Schmidt <ryandes...@macports.org> wrote:
> 
> On Oct 14, 2016, at 09:51, Richard L. Hamilton <rlha...@smart.net> wrote:
>> 
>> 
>>> On Oct 13, 2016, at 20:00, Ryan Schmidt <ryandes...@macports.org> wrote:
>>> 
>>> On Oct 13, 2016, at 16:58, Richard L. Hamilton <rlha...@smart.net> wrote:
>>>> 
>>>> This was a build prerequisite for py27-numpy, so I can't build that 
>>>> either.  It failed both with and without +universal (I think I have the 
>>>> latter set as default).
>>> 
>>> For some reason the 10.6 x86_64 builder didn't build it either so we didn't 
>>> have a gcc6-6.2.0_0.darwin_10.x86_64.tbz2 package. It's not easy to find 
>>> old buildbot logs so I don't know if it failed because of the same 
>>> bootstrap failure you saw or for some other reason.
>>> 
>>> Josh forced a build now, and it built successfully, so if you clean and try 
>>> again, without the universal variant, you'll get the package from us 
>>> without having to compile it yourself.
>> 
>> That worked, thanks!
>> 
>> I see that a similar situation applied to gcc49 and dpkg, that I couldn't 
>> build them for myself with +universal, but pre-built binaries were available 
>> without it.  That might well apply to the other things that aren't building 
>> for me too, except they're libraries, and I already have a bunch of 
>> dependencies on them in place, so that wouldn't be a good idea unless I 
>> redid everything without universal.
> 
> "Bootstrap comparison failure" is an error message specific to the GCC build 
> system. GCC is a very large and complex piece of software, and I don't 
> understand its build system or why this error sometimes happens for some 
> people.
> 
> If you are also having problems building other less complex software, like 
> dpkg, would you please show us the main.log file from those failures? They 
> may be easier to diagnose.
> 

dpkg +universal used to build with clang some time back according to my notes, 
but no more:

:info:build /usr/bin/llvm-g++-4.2 -DHAVE_CONFIG_H   
-DLOCALEDIR=\"/opt/local/share/locale\" -DADMINDIR=\"/opt/local/var/db/dpkg\" 
-DLIBDIR=\"/opt/local/lib/dpkg\" -DLOCALLIBDIR=\"/usr/local/lib/dpkg\" 
-idirafter ../lib/compat -iquote . -I.. -I../lib -I/opt/local/include -fno-rtti 
-Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers 
-Wmissing-declarations -Wmissing-format-attribute -Wformat -Wformat-security 
-Wpointer-arith -Winit-self -Wwrite-strings -Wcast-align -Wshadow -pipe -Os 
-arch x86_64 -arch i386 -MT baselist.o -MD -MP -MF .deps/baselist.Tpo -c -o 
baselist.o baselist.cc
:info:build llvm-g++-4.2: -E, -S, -save-temps and -M options are not allowed 
with multiple -arch flags
:info:build gnumake[3]: *** [basecmds.o] Error 1
:info:build gnumake[3]: *** Waiting for unfinished jobs....
:info:build llvm-g++-4.2: -E, -S, -save-temps and -M options are not allowed 
with multiple -arch flags
[...]

ditto with the default of gcc-4.2:

:info:build libtool: compile:  /usr/bin/gcc-4.2 -DHAVE_CONFIG_H -idirafter 
../../lib/compat -I../.. -DTEST_LIBCOMPAT=1 -I/opt/local/include -Wall -Wextra 
-Wno-unused-parameter -Wno-missing-field-initializers -Wmissing-declarations 
-Wmissing-format-attribute -Wformat -Wformat-security -Wpointer-arith 
-Winit-self -Wwrite-strings -Wcast-align -Wshadow -Wdeclaration-after-statement 
-Wnested-externs -Wbad-function-cast -Wstrict-prototypes -Wmissing-prototypes 
-Wold-style-definition -pipe -Os -arch x86_64 -arch i386 -MT 
libcompat_test_la-strnlen.lo -MD -MP -MF .deps/libcompat_test_la-strnlen.Tpo -c 
strnlen.c -o libcompat_test_la-strnlen.o
:info:build gcc-4.2: -E, -S, -save-temps and -M options are not allowed with 
multiple -arch flags
[...]

Hardware Version: Mac mini (Mid 2007)
Model Identifier: Macmini2,1
Mac OS X 10.6.8 10K549 (Snow Leopard)
Xcode 3.2.6 DevToolsSupport-1806.0
MacPorts Version: 2.3.4


_______________________________________________
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users

Reply via email to