[PATCH] D19265: [Sparc] Add Soft Float support

[James Y Knight via llvm-commits]

jyknight added a subscriber: echristo.
jyknight added a comment.

What's the ABI impact of this? Will this use the same calling convention as GCC does?


================
Comment at: lib/Target/Sparc/SparcTargetMachine.cpp:87
@@ +86,3 @@
+  // function, so we can enable it as a subtarget feature.                      
+  bool softFloat =                                                              
+      F.hasFnAttribute("use-soft-float") &&                                     
----------------
I wonder what the point of having this redundant way of specifying soft-float via "use-soft-float" is. Clang seems to emit *both* use-soft-float=true and +soft-float. Hopefully this isn't needed these days. @echristo?

================
Comment at: lib/Target/Sparc/SparcTargetMachine.h:25
@@ -24,2 +24,3 @@
   std::unique_ptr<TargetLoweringObjectFile> TLOF;
   SparcSubtarget Subtarget;
+  mutable StringMap<std::unique_ptr<SparcSubtarget>> SubtargetMap;
----------------
Pretty sure this ought to be deleted while adding the SubtargetMap.


http://reviews.llvm.org/D19265