Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-29 Thread Bjarne D Mathiesen
I tried editing the libgcc8 Portfile "configure.cxxflags-append -Wl,-read_only_relocs,suppress" has no influence and just gives the original error. "configure.ldflags-append -read_only_relocs suppress" gives : #=> port -v install libgcc ---> Computing dependencies for libgcc.. The following

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-29 Thread Ken Cunningham
> /macports/bin/clang++-mp-5.0 -arch i386 -std=gnu++98 -g -DIN_GCC > -fno-strict-aliasing -fno-exceptions -fno-rtti > -fasynchronous-unwind-tables -W -Wall -Wno-narrowing -Wwrite-strings > -Wcast-qual -Wno-format -Wmissing-format-attribute -Woverloaded-virtual > -pedantic -Wno-long-long

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-29 Thread Bjarne D Mathiesen
OK - I've re-installed the toolchain so presently I've got : #=> port installed name:^clang name:^llvm name:*gcc* The following ports are currently installed: clang-3.4 @3.4.2_12 clang-3.7 @3.7.1_5+analyzer (active) clang-5.0 @5.0.2_3+analyzer+defaultlibcxx+emulated_tls+libstdcxx (active)

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-28 Thread Bjarne D Mathiesen
Ken Cunningham wrote: >> Bootstrapping 3: >> (5) & (7) : >> the instructions state +llvm39 ; >> but shouldn't it be +llwm37 ?!? > > The instructions state +llvm50 (I changed them from llvm39 about a week ago, > in anticipation of llvm-3.9 being retired) > My bad ;-) I've just had

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-28 Thread Ken Cunningham
> Bootstrapping 3: > (5) & (7) : > the instructions state +llvm39 ; > but shouldn't it be +llwm37 ?!? The instructions state +llvm50 (I changed them from llvm39 about a week ago, in anticipation of llvm-3.9 being retired) > (10) : > according to your procedure at described above, >

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-28 Thread Bjarne D Mathiesen
OK ! There have been some major changes in a lot of PortFile recently ! Ive just had to do this : #=> port -vN install libcxx ---> Computing dependencies for libcxx ---> Dependencies to be installed: clang-3.4 cctools

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-28 Thread Bjarne D Mathiesen
Further notes regarding LibcxxOnOlderSystems : (12) : mentions ripping clang-3.4 out ; but isn't it enough to remove it from default_compilers ?!? so that one still has it in case of an emergency ?!? alternatively, can one use eg "port (pkg|dmg) (clang|llvm)-x.y" to store the

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-28 Thread Bjarne D Mathiesen
Ken Cunningham wrote: >> On 27 Aug 2019, at 15:28, Ken Cunningham wrote: >> >> >/Luckily, we'll always have clang-3.7, / >> That seems like an oxymoron... >> >> Is there really no hope of ever getting away from needing such an old >> version? > > Practically, no, not on MacPorts for

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-27 Thread Ken Cunningham
> On 27 Aug 2019, at 15:28, Ken Cunningham wrote: > > > Luckily, we'll always have clang-3.7, > > That seems like an oxymoron... > > Is there really no hope of ever getting away from needing such an old > version? Practically, no, not on MacPorts for bootstrapping anyway. clang-3.4 cannot

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-27 Thread Bill Cole
On 27 Aug 2019, at 15:28, Ken Cunningham wrote: Luckily, we'll always have clang-3.7, That seems like an oxymoron... Is there really no hope of ever getting away from needing such an old version? I've just been through the 3rd or 4th round of re-bootstrapping my libc++ on 10.6.8 world,

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-27 Thread Ken Cunningham
> On 27/08/2019 1:31 pm, Ken Cunningham wrote: > > that is a 32 bit error that is usually suppressed by passing > > "-read_only_relocs suppress"' to the linker, IIRC. > > > > Why are you seeing it, though, I wonder? > > > > I don't think anyone has likely tried to build libgcc8 32bit on 10.6.8

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-27 Thread Bjarne D Mathiesen
Chris Jones wrote: > > > On 27/08/2019 1:31 pm, Ken Cunningham wrote: >> that is a 32 bit error that is usually suppressed by passing >> "-read_only_relocs suppress"' to the linker, IIRC. >> >> Why are you seeing it, though, I wonder? >> >> I don't think anyone has likely tried to build

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-27 Thread Ken Cunningham
defaulting systems to clang-8.0 for all builds will be pretty interesting, and likely will trigger quite a few build errors as it's so strict and defaults to c++17...but certainly will be useful in the end to help pave the way for newer xcode clangs to follow...

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-27 Thread Chris Jones
On 27/08/2019 1:31 pm, Ken Cunningham wrote: that is a 32 bit error that is usually suppressed by passing "-read_only_relocs suppress"' to the linker, IIRC. Why are you seeing it, though, I wonder? I don't think anyone has likely tried to build libgcc8 32bit on 10.6.8 with clang-8.0...

Re: [libgcc8] compile failure on i386 on 10.6 Snow

2019-08-27 Thread Ken Cunningham
that is a 32 bit error that is usually suppressed by passing "-read_only_relocs suppress"' to the linker, IIRC. Why are you seeing it, though, I wonder? I don't think anyone has likely tried to build libgcc8 32bit on 10.6.8 with clang-8.0... Why don't you try it with clang-3.9 K