[llvm] r286592 - [AArch64] Enable merging of adjacent zero stores for all subtargets.

[Chad Rosier via llvm-commits]

Author: mcrosier
Date: Fri Nov 11 08:10:12 2016
New Revision: 286592

URL: http://llvm.org/viewvc/llvm-project?rev=286592&view=rev
Log:
[AArch64] Enable merging of adjacent zero stores for all subtargets.

This optimization merges adjacent zero stores into a wider store.

e.g.,

strh wzr, [x0]
strh wzr, [x0, #2]
; becomes
str wzr, [x0]

e.g.,

str wzr, [x0]
str wzr, [x0, #4]
; becomes
str xzr, [x0]

Previously, this was only enabled for Kryo and Cortex-A57.

Differential Revision: https://reviews.llvm.org/D26396

Modified:
    llvm/trunk/lib/Target/AArch64/AArch64.td
    llvm/trunk/lib/Target/AArch64/AArch64LoadStoreOptimizer.cpp
    llvm/trunk/lib/Target/AArch64/AArch64Subtarget.h
    llvm/trunk/test/CodeGen/AArch64/arm64-narrow-st-merge.ll

Modified: llvm/trunk/lib/Target/AArch64/AArch64.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AArch64/AArch64.td?rev=286592&r1=286591&r2=286592&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AArch64/AArch64.td (original)
+++ llvm/trunk/lib/Target/AArch64/AArch64.td Fri Nov 11 08:10:12 2016
@@ -61,11 +61,6 @@ def FeatureReserveX18 : SubtargetFeature
                                          "Reserve X18, making it unavailable "
                                          "as a GPR">;
 
-def FeatureMergeNarrowZeroSt : SubtargetFeature<"merge-narrow-zero-st",
-                                                "MergeNarrowZeroStores", "true",
-                                                "Merge narrow zero store "
-                                                "instructions">;
-
 def FeatureUseAA : SubtargetFeature<"use-aa", "UseAA", "true",
                                     "Use alias analysis during codegen">;
 
@@ -182,7 +177,6 @@ def ProcA57     : SubtargetFeature<"a57"
                                    FeatureCrypto,
                                    FeatureCustomCheapAsMoveHandling,
                                    FeatureFPARMv8,
-                                   FeatureMergeNarrowZeroSt,
                                    FeatureNEON,
                                    FeaturePerfMon,
                                    FeaturePostRAScheduler,
@@ -253,7 +247,6 @@ def ProcKryo    : SubtargetFeature<"kryo
                                    FeatureCrypto,
                                    FeatureCustomCheapAsMoveHandling,
                                    FeatureFPARMv8,
-                                   FeatureMergeNarrowZeroSt,
                                    FeatureNEON,
                                    FeaturePerfMon,
                                    FeaturePostRAScheduler,

Modified: llvm/trunk/lib/Target/AArch64/AArch64LoadStoreOptimizer.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AArch64/AArch64LoadStoreOptimizer.cpp?rev=286592&r1=286591&r2=286592&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AArch64/AArch64LoadStoreOptimizer.cpp (original)
+++ llvm/trunk/lib/Target/AArch64/AArch64LoadStoreOptimizer.cpp Fri Nov 11 08:10:12 2016
@@ -1699,8 +1699,7 @@ bool AArch64LoadStoreOpt::runOnMachineFu
   UsedRegs.resize(TRI->getNumRegs());
 
   bool Modified = false;
-  bool enableNarrowZeroStOpt =
-      Subtarget->mergeNarrowStores() && !Subtarget->requiresStrictAlign();
+  bool enableNarrowZeroStOpt = !Subtarget->requiresStrictAlign();
   for (auto &MBB : Fn)
     Modified |= optimizeBlock(MBB, enableNarrowZeroStOpt);
 

Modified: llvm/trunk/lib/Target/AArch64/AArch64Subtarget.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AArch64/AArch64Subtarget.h?rev=286592&r1=286591&r2=286592&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AArch64/AArch64Subtarget.h (original)
+++ llvm/trunk/lib/Target/AArch64/AArch64Subtarget.h Fri Nov 11 08:10:12 2016
@@ -71,7 +71,6 @@ protected:
 
   // StrictAlign - Disallow unaligned memory accesses.
   bool StrictAlign = false;
-  bool MergeNarrowZeroStores = false;
   bool UseAA = false;
   bool PredictableSelectIsExpensive = false;
   bool BalanceFPOps = false;
@@ -179,7 +178,6 @@ public:
   bool hasCrypto() const { return HasCrypto; }
   bool hasCRC() const { return HasCRC; }
   bool hasRAS() const { return HasRAS; }
-  bool mergeNarrowStores() const { return MergeNarrowZeroStores; }
   bool balanceFPOps() const { return BalanceFPOps; }
   bool predictableSelectIsExpensive() const {
     return PredictableSelectIsExpensive;

Modified: llvm/trunk/test/CodeGen/AArch64/arm64-narrow-st-merge.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AArch64/arm64-narrow-st-merge.ll?rev=286592&r1=286591&r2=286592&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/AArch64/arm64-narrow-st-merge.ll (original)
+++ llvm/trunk/test/CodeGen/AArch64/arm64-narrow-st-merge.ll Fri Nov 11 08:10:12 2016
@@ -1,6 +1,4 @@
-; RUN: llc < %s -mtriple aarch64--none-eabi -mcpu=cortex-a57 -verify-machineinstrs | FileCheck %s
-; RUN: llc < %s -mtriple aarch64_be--none-eabi -mcpu=cortex-a57 -verify-machineinstrs | FileCheck %s
-; RUN: llc < %s -mtriple aarch64--none-eabi -mcpu=kryo -verify-machineinstrs | FileCheck %s
+; RUN: llc < %s -mtriple aarch64--none-eabi -verify-machineinstrs | FileCheck %s
 
 ; CHECK-LABEL: Strh_zero
 ; CHECK: str wzr