Fix BasicTTI::getCmpSelInstrCost to deal with illegal vector types

The default implementation of getCmpSelInstrCost, which provides the cost of
icmp/fcmp/select instructions, did not deal sensibly with illegal vector types
that were scalarized. We'd ask for the legalization cost of the vector type,
which would return something like (4, f64) given an input of <4 x double>, and
we'd then check the TLI status of the ISD opcode on that scalar type. This would
result in querying (ISD::VSELECT, f64), for example. Amusingly enough,
ISD::VSELECT on scalar types is marked as Legal by default (as with most other
operations), and most backends never change this because VSELECT is never
generated on scalars. However, seeing the resulting operation as Legal, we'd
neglect to add the scalarization cost before returning. The result is that we'd
grossly under-estimate the cost of cmps/selects on illegal vector types.

Now, if type legalization clearly results in scalarization, we skip the early
return and add the scalarization cost.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@217859 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Hal Finkel 2014-09-16 04:35:50 +00:00
parent 50c1eff629
commit 8def8d9263
2 changed files with 16 additions and 1 deletions

View File

@ -468,7 +468,8 @@ unsigned BasicTTI::getCmpSelInstrCost(unsigned Opcode, Type *ValTy,
std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(ValTy); std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(ValTy);
if (!TLI->isOperationExpand(ISD, LT.second)) { if (!(ValTy->isVectorTy() && !LT.second.isVector()) &&
!TLI->isOperationExpand(ISD, LT.second)) {
// The operation is legal. Assume it costs 1. Multiply // The operation is legal. Assume it costs 1. Multiply
// by the type-legalization overhead. // by the type-legalization overhead.
return LT.first * 1; return LT.first * 1;

View File

@ -0,0 +1,14 @@
; 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() {
; VSX is disabled, so this cost needs to include scalarization (because
; <4 x double> is legalized to scalars).
; CHECK: cost of 44 {{.*}} fcmp
%v1 = fcmp ugt <4 x double> undef, undef
ret void
}