Fix multi-register costs in BasicTTI::getCastInstrCost

For an cast (extension, etc.), the currently logic predicts a low cost if the
associated operation (keyed on the destination type) is legal (or promoted).
This is not true when the number of values required to legalize the type is
changing. For example, <8 x i16> being sign extended by <8 x i32> is not
generically cheap on PPC with VSX, even though sign extension to v4i32 is
legal, because two output v4i32 values are required compared to the single
v8i16 input value, and without custom logic in the target, this conversion will
scalarize.

llvm-svn: 205487

2 files changed, 23 insertions, 1 deletions

diff --git a/llvm/lib/CodeGen/BasicTargetTransformInfo.cpp b/llvm/lib/CodeGen/BasicTargetTransformInfo.cpp
index 03c5eba3692..25486cf3919 100644
--- a/llvm/lib/CodeGen/BasicTargetTransformInfo.cpp
+++ b/llvm/lib/CodeGen/BasicTargetTransformInfo.cpp
@@ -297,7 +297,8 @@ unsigned BasicTTI::getCastInstrCost(unsigned Opcode, Type *Dst,
     return 0;
 
   // If the cast is marked as legal (or promote) then assume low cost.
-  if (TLI->isOperationLegalOrPromote(ISD, DstLT.second))
+  if (SrcLT.first == DstLT.first &&
+      TLI->isOperationLegalOrPromote(ISD, DstLT.second))
     return 1;
 
   // Handle scalar conversions.
diff --git a/llvm/test/Analysis/CostModel/PowerPC/ext.ll b/llvm/test/Analysis/CostModel/PowerPC/ext.ll
new file mode 100644
index 00000000000..daaa8f5bac3
--- /dev/null
+++ b/llvm/test/Analysis/CostModel/PowerPC/ext.ll
@@ -0,0 +1,21 @@
+; RUN: opt < %s -cost-model -analyze -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=+vsx | FileCheck %s
+target datalayout = "E-m:e-i64:64-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+define void @exts() {
+
+  ; CHECK: cost of 1 {{.*}} sext
+  %v1 = sext i16 undef to i32
+
+  ; CHECK: cost of 1 {{.*}} sext
+  %v2 = sext <2 x i16> undef to <2 x i32>
+
+  ; CHECK: cost of 1 {{.*}} sext
+  %v3 = sext <4 x i16> undef to <4 x i32>
+
+  ; CHECK: cost of 216 {{.*}} sext
+  %v4 = sext <8 x i16> undef to <8 x i32>
+
+  ret void
+}
+