https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61315
Jack Howarth <howarth at nitro dot med.uc.edu> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |howarth at nitro dot med.uc.edu
--- Comment #5 from Jack Howarth <howarth at nitro dot med.uc.edu> ---
(In reply to Andrew Pinski from comment #3)
> Still does not matter, the compiler is broken and should be reported to
> Apple. The time for work around in broken compilers is so 90s and really
> should not be done. This is clang being broken and really should be fixed
> there instead.
Is the object here to burn all bridges with the darwin target and leave those
users only the option of using llvm based compilers as of gcc 4.10? While I
realize you are all frustrated with the abandonment of post-GPLv2 tools in
darwin., this action simply makes it extremely easy for Apple to depreciate
their ancient gas assembler in the next Xcode for darwin14 since Apple can
rightly say that FSF gcc no longer bootstraps on darwin anyway. I am certain
that gas will be depreciated from some future darwin but hopefully will be
replaced by a stand-alone assembler based on the clang built-in assembler.
However, if FSF gcc starts to get overly self-self-rightous about the compiler
details of clang, we might as well fold up our tents on the darwin target.
ps Be careful what you wish for. I just updated the llvm34-3.4.1-0e package in
fink to use the openmp svn trunk and a back port of the current clang-omp
github based on clang 3.4. The resulting clang/clang++ compilers have excellent
OpenMP performance defaulted to -fopenmp as -liomp5 rather than -lgomp.
http://fink.9193.n7.nabble.com/llvm34-3-4-1-0e-and-OpenMP3-1-Validation-tt46118.html
While you are still lucky that there has been little effort on a fortran
llvm-based compiler, the rest is a work in rapid progress. So go ahead and
dismiss that user base if that is what you desire.
