[llvm] r217056 - unique_ptrify IRObjectFile::createIRObjectFile

[David Blaikie]

On Thu, Sep 4, 2014 at 6:23 PM, Robinson, Paul <
Paul_Robinson at playstation.sony.com> wrote:

> > -----Original Message-----
> > From: llvm-commits-bounces at cs.uiuc.edu [mailto:llvm-commits-
> > bounces at cs.uiuc.edu] On Behalf Of Rafael Espíndola
> > Sent: Thursday, September 04, 2014 11:24 AM
> > To: David Blaikie
> > Cc: llvm-commits
> > Subject: Re: [llvm] r217056 - unique_ptrify
> > IRObjectFile::createIRObjectFile
> >
> > On 4 September 2014 14:10, David Blaikie <dblaikie at gmail.com> wrote:
> > > Hmm, and now that I do actually run all the tests (maybe after
> > explicitly
> > > building LLVMgold.so the build system decided to also run its tests on
> > > "ninja check-all") I get this failure:
> > >
> > > ld: error: unrecognized emulation elf32ppc
> > > ld: error:
> > >
> >
> /mnt/fast/dev/llvm/build/clang/debug/split/notypes/nostandalone/test/tools
> > /gold/Output/mtriple.ll.tmp3.o:
> > > not configured to support 32-bit big-endian object
> > >
> > > Any idea if I can actually address this? Do I have to build my own
> 'ld'?
> > ...
> > > otherwise I'll probably just disable it again.
> >
> > I guess the version of gold you are using has been explicitly
> > configured to not support 32 bit big-endian :-(
> > The only solution I can think of is trying to update
> > have_ld_plugin_support to check for that.
>
> I've just run into this and blamed it on having a plugin-capable gold
> but not building the ppc backend.  In which case the right fix is in
> test/tools/gold/lit.local.cfg to require ppc target as well as x86.
>

I don't /think/ that's the issue. I build with all targets enabled and I
was still seeing these problems. Pretty sure as Rafael mentioned, it's
something in the way I build my 'gold'. But I'm not too fussed about fixing
that so I just went back to not building the gold plugin, as before. I'll
fix any minor build breaks I introduce in the future after the buildbots
complain. (it's infrequent enough that I'm not too fussed about that)

- David


> --paulr
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140905/546a192b/attachment.html>