labrinea added inline comments.
================
Comment at: lib/Basic/Targets.cpp:4785
@@ +4784,3 @@
+ if (Opts.C99 && !Opts.Freestanding) {
+ Builder.defineMacro("__ARM_FP_FENV_ROUNDING", "1");
+ Builder.defineMacro("__STDC_IEC_559__", "1");
----------------
rengolin wrote:
> I'm not convinced by the runtime rounding being depending on C99, or why it's
> not possible in freestanding environments.
I copied the same logic from AArch64.
================
Comment at: lib/Basic/Targets.cpp:4786
@@ +4785,3 @@
+ Builder.defineMacro("__ARM_FP_FENV_ROUNDING", "1");
+ Builder.defineMacro("__STDC_IEC_559__", "1");
+ }
----------------
Shall we define this unconditionally?
================
Comment at: lib/Basic/Targets.cpp:4790
@@ -4779,1 +4789,3 @@
+ if (!Opts.C11)
+ Builder.defineMacro("__SUPPORT_SNAN__", "1");
}
----------------
rengolin wrote:
> Er, well, the ACLE doc says it wasn't adopted in C11, but that doesn't mean
> that it was present in all others, just that there is a proposal for C11
> which wasn't accepted.
>
> I think this macro needs some other mechanism to be set, and unless
> Clang/LLVM supports signaling NANs unconditionally, we shouldn't add it.
Maybe we could skip this macro in this patch? @richard.barton.arm
================
Comment at: lib/Basic/Targets.cpp:5270
@@ +5269,3 @@
+ if (Opts.C99 && !Opts.Freestanding) {
+ Builder.defineMacro("__ARM_FP_FENV_ROUNDING", "1");
+ Builder.defineMacro("__STDC_IEC_559__", "1");
----------------
rengolin wrote:
> same comments as above.
This was present already.
http://reviews.llvm.org/D12633
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
