Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
This is now handled by clang
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
These should be handled by clang
Reverts b72f4056 Introduce per device defines.
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
LGTM for all three patches.
-Tom
Any reason why these three and previous two patches from Jan were not applied?
Regards,
Vedran
> > Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
>
> LGTM for all three patches.
>
> -Tom
>Any reason why these three and previous two patches from Jan were
not
applied?
Not for 1/3, but I did not want to push 2/3 and 3/3 before
http://reviews.llvm.org/D20389 lands.
The current state (build failure for nvptx) is imo preferable to the
alternative of miscompiling nvptx libclc (compiling without
cl_khr_fp64).
the get_global_offset patches depend on:
http://reviews.llvm.org/D20297
http://reviews.llvm.org/D20298
the other dependencies are already accepted:
http://reviews.llvm.org/D20691
http://reviews.llvm.org/D20299
regards,
Jan