[llvm] r233464 - [CodeGen] "PromoteInteger" f32 to f64 doesn't make sense.

[Ahmed Bougacha]

Author: ab
Date: Fri Mar 27 20:22:37 2015
New Revision: 233464

URL: http://llvm.org/viewvc/llvm-project?rev=233464&view=rev
Log:
[CodeGen] "PromoteInteger" f32 to f64 doesn't make sense.

The original f32->f64 promotion logic was refactored into roughly the
currently shape in r37781. However, starting with r132263, the
legalizer has been split into different kinds, and the previous
"Promote" (which did the right thing) was search-and-replace'd into
"PromoteInteger". The divide gradually deepened, with type legalization
("PromoteInteger") being separated from ops legalization
("Promote", which still works for floating point ops).

Fast-forward to today: there's no in-tree target with legal f64 but
illegal f32 (rather: no tests were harmed in the making of this patch).

With such a target, i.e., if you trick the legalizer into going through
the PromoteInteger path for FP, you get the expected brokenness.
For instance, there's no PromoteIntRes_FADD (the name itself sounds
wrong), so we'll just hit some assert in the PromoteInteger path.

Don't pretend we can promote f32 to f64. Instead, always soften.

Modified:
    llvm/trunk/lib/CodeGen/TargetLoweringBase.cpp

Modified: llvm/trunk/lib/CodeGen/TargetLoweringBase.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/CodeGen/TargetLoweringBase.cpp?rev=233464&r1=233463&r2=233464&view=diff
==============================================================================
--- llvm/trunk/lib/CodeGen/TargetLoweringBase.cpp (original)
+++ llvm/trunk/lib/CodeGen/TargetLoweringBase.cpp Fri Mar 27 20:22:37 2015
@@ -1246,20 +1246,13 @@ void TargetLoweringBase::computeRegister
     ValueTypeActions.setTypeAction(MVT::f64, TypeSoftenFloat);
   }
 
-  // Decide how to handle f32. If the target does not have native support for
-  // f32, promote it to f64 if it is legal. Otherwise, expand it to i32.
+  // Decide how to handle f32. If the target does not have native f32 support,
+  // expand it to i32 and we will be generating soft float library calls.
   if (!isTypeLegal(MVT::f32)) {
-    if (isTypeLegal(MVT::f64)) {
-      NumRegistersForVT[MVT::f32] = NumRegistersForVT[MVT::f64];
-      RegisterTypeForVT[MVT::f32] = RegisterTypeForVT[MVT::f64];
-      TransformToType[MVT::f32] = MVT::f64;
-      ValueTypeActions.setTypeAction(MVT::f32, TypePromoteInteger);
-    } else {
-      NumRegistersForVT[MVT::f32] = NumRegistersForVT[MVT::i32];
-      RegisterTypeForVT[MVT::f32] = RegisterTypeForVT[MVT::i32];
-      TransformToType[MVT::f32] = MVT::i32;
-      ValueTypeActions.setTypeAction(MVT::f32, TypeSoftenFloat);
-    }
+    NumRegistersForVT[MVT::f32] = NumRegistersForVT[MVT::i32];
+    RegisterTypeForVT[MVT::f32] = RegisterTypeForVT[MVT::i32];
+    TransformToType[MVT::f32] = MVT::i32;
+    ValueTypeActions.setTypeAction(MVT::f32, TypeSoftenFloat);
   }
 
   if (!isTypeLegal(MVT::f16)) {