[llvm] r313288 - [TargetTransformInfo] Detect 0 latency instructions

[Guozhi Wei via llvm-commits]

Author: carrot
Date: Thu Sep 14 12:20:02 2017
New Revision: 313288

URL: http://llvm.org/viewvc/llvm-project?rev=313288&view=rev
Log:
[TargetTransformInfo] Detect 0 latency instructions

For instructions that unlikely generate machine instructions, they should also have 0 latency.

Differential Revision: https://reviews.llvm.org/D37833


Modified:
    llvm/trunk/include/llvm/Analysis/TargetTransformInfoImpl.h
    llvm/trunk/test/Analysis/CostModel/X86/costmodel.ll

Modified: llvm/trunk/include/llvm/Analysis/TargetTransformInfoImpl.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/include/llvm/Analysis/TargetTransformInfoImpl.h?rev=313288&r1=313287&r2=313288&view=diff
==============================================================================
--- llvm/trunk/include/llvm/Analysis/TargetTransformInfoImpl.h (original)
+++ llvm/trunk/include/llvm/Analysis/TargetTransformInfoImpl.h Thu Sep 14 12:20:02 2017
@@ -775,7 +775,9 @@ public:
   }
 
   int getInstructionLatency(const Instruction *I) {
-    if (isa<PHINode>(I))
+    SmallVector<const Value *, 4> Operands(I->value_op_begin(),
+                                           I->value_op_end());
+    if (getUserCost(I, Operands) == TTI::TCC_Free)
       return 0;
 
     if (isa<CallInst>(I))

Modified: llvm/trunk/test/Analysis/CostModel/X86/costmodel.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Analysis/CostModel/X86/costmodel.ll?rev=313288&r1=313287&r2=313288&view=diff
==============================================================================
--- llvm/trunk/test/Analysis/CostModel/X86/costmodel.ll (original)
+++ llvm/trunk/test/Analysis/CostModel/X86/costmodel.ll Thu Sep 14 12:20:02 2017
@@ -3,6 +3,8 @@
 
 ; Tests if the interface TargetTransformInfo::getInstructionCost() works correctly.
 
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+
 define i64 @foo(i64 %arg) {
 
   ; LATENCY:  cost of 1 {{.*}} %I64 = add
@@ -13,6 +15,22 @@ define i64 @foo(i64 %arg) {
   ; CODESIZE: cost of 1 {{.*}} load
   load i64, i64* undef, align 4
 
+  ; LATENCY:  cost of 0 {{.*}} bitcast
+  ; CODESIZE: cost of 0 {{.*}} bitcast
+  %BC = bitcast i8* undef to i32*
+
+  ; LATENCY:  cost of 0 {{.*}} inttoptr
+  ; CODESIZE: cost of 0 {{.*}} inttoptr
+  %I2P = inttoptr i64 undef to i8*
+
+  ; LATENCY:  cost of 0 {{.*}} ptrtoint
+  ; CODESIZE: cost of 0 {{.*}} ptrtoint
+  %P2I = ptrtoint i8* undef to i64
+
+  ; LATENCY:  cost of 0 {{.*}} trunc
+  ; CODESIZE: cost of 0 {{.*}} trunc
+  %TC = trunc i64 undef to i32
+
   ; LATENCY:  cost of 1 {{.*}} ret
   ; CODESIZE: cost of 1 {{.*}} ret
   ret i64 undef