[llvm] 4167645 - GlobalISel: Move Localizer::shouldLocalize(..) to TargetLowering
Volkan Keles via llvm-commits
llvm-commits at lists.llvm.org
Mon Mar 2 09:16:51 PST 2020
Author: Volkan Keles
Date: 2020-03-02T09:15:40-08:00
New Revision: 4167645d1e6a5ecc8790f0aba450799c4441882d
URL: https://github.com/llvm/llvm-project/commit/4167645d1e6a5ecc8790f0aba450799c4441882d
DIFF: https://github.com/llvm/llvm-project/commit/4167645d1e6a5ecc8790f0aba450799c4441882d.diff
LOG: GlobalISel: Move Localizer::shouldLocalize(..) to TargetLowering
Add a new target hook for shouldLocalize so that
targets can customize the logic.
https://reviews.llvm.org/D75207
Added:
Modified:
llvm/include/llvm/CodeGen/TargetLowering.h
llvm/lib/CodeGen/GlobalISel/Localizer.cpp
llvm/lib/CodeGen/TargetLoweringBase.cpp
Removed:
################################################################################
diff --git a/llvm/include/llvm/CodeGen/TargetLowering.h b/llvm/include/llvm/CodeGen/TargetLowering.h
index 9583e2b718e5..0552420c3c33 100644
--- a/llvm/include/llvm/CodeGen/TargetLowering.h
+++ b/llvm/include/llvm/CodeGen/TargetLowering.h
@@ -2754,6 +2754,13 @@ class TargetLoweringBase {
/// The default implementation just freezes the set of reserved registers.
virtual void finalizeLowering(MachineFunction &MF) const;
+ //===----------------------------------------------------------------------===//
+ // GlobalISel Hooks
+ //===----------------------------------------------------------------------===//
+ /// Check whether or not \p MI needs to be moved close to its uses.
+ virtual bool shouldLocalize(const MachineInstr &MI, const TargetTransformInfo *TTI) const;
+
+
private:
const TargetMachine &TM;
diff --git a/llvm/lib/CodeGen/GlobalISel/Localizer.cpp b/llvm/lib/CodeGen/GlobalISel/Localizer.cpp
index 1c4a668e5f31..9aac47ecb350 100644
--- a/llvm/lib/CodeGen/GlobalISel/Localizer.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/Localizer.cpp
@@ -13,6 +13,7 @@
#include "llvm/ADT/DenseMap.h"
#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
+#include "llvm/CodeGen/TargetLowering.h"
#include "llvm/InitializePasses.h"
#include "llvm/Support/Debug.h"
@@ -40,60 +41,6 @@ void Localizer::init(MachineFunction &MF) {
TTI = &getAnalysis<TargetTransformInfoWrapperPass>().getTTI(MF.getFunction());
}
-bool Localizer::shouldLocalize(const MachineInstr &MI) {
- // Assuming a spill and reload of a value has a cost of 1 instruction each,
- // this helper function computes the maximum number of uses we should consider
- // for remat. E.g. on arm64 global addresses take 2 insts to materialize. We
- // break even in terms of code size when the original MI has 2 users vs
- // choosing to potentially spill. Any more than 2 users we we have a net code
- // size increase. This doesn't take into account register pressure though.
- auto maxUses = [](unsigned RematCost) {
- // A cost of 1 means remats are basically free.
- if (RematCost == 1)
- return UINT_MAX;
- if (RematCost == 2)
- return 2U;
-
- // Remat is too expensive, only sink if there's one user.
- if (RematCost > 2)
- return 1U;
- llvm_unreachable("Unexpected remat cost");
- };
-
- // Helper to walk through uses and terminate if we've reached a limit. Saves
- // us spending time traversing uses if all we want to know is if it's >= min.
- auto isUsesAtMost = [&](unsigned Reg, unsigned MaxUses) {
- unsigned NumUses = 0;
- auto UI = MRI->use_instr_nodbg_begin(Reg), UE = MRI->use_instr_nodbg_end();
- for (; UI != UE && NumUses < MaxUses; ++UI) {
- NumUses++;
- }
- // If we haven't reached the end yet then there are more than MaxUses users.
- return UI == UE;
- };
-
- switch (MI.getOpcode()) {
- default:
- return false;
- // Constants-like instructions should be close to their users.
- // We don't want long live-ranges for them.
- case TargetOpcode::G_CONSTANT:
- case TargetOpcode::G_FCONSTANT:
- case TargetOpcode::G_FRAME_INDEX:
- case TargetOpcode::G_INTTOPTR:
- return true;
- case TargetOpcode::G_GLOBAL_VALUE: {
- unsigned RematCost = TTI->getGISelRematGlobalCost();
- Register Reg = MI.getOperand(0).getReg();
- unsigned MaxUses = maxUses(RematCost);
- if (MaxUses == UINT_MAX)
- return true; // Remats are "free" so always localize.
- bool B = isUsesAtMost(Reg, MaxUses);
- return B;
- }
- }
-}
-
void Localizer::getAnalysisUsage(AnalysisUsage &AU) const {
AU.addRequired<TargetTransformInfoWrapperPass>();
getSelectionDAGFallbackAnalysisUsage(AU);
@@ -119,9 +66,10 @@ bool Localizer::localizeInterBlock(MachineFunction &MF,
// we only localize instructions in the entry block here. This might change if
// we start doing CSE across blocks.
auto &MBB = MF.front();
+ auto &TL = *MF.getSubtarget().getTargetLowering();
for (auto RI = MBB.rbegin(), RE = MBB.rend(); RI != RE; ++RI) {
MachineInstr &MI = *RI;
- if (!shouldLocalize(MI))
+ if (!TL.shouldLocalize(MI, TTI))
continue;
LLVM_DEBUG(dbgs() << "Should localize: " << MI);
assert(MI.getDesc().getNumDefs() == 1 &&
diff --git a/llvm/lib/CodeGen/TargetLoweringBase.cpp b/llvm/lib/CodeGen/TargetLoweringBase.cpp
index 95c63d09718c..436857d6b215 100644
--- a/llvm/lib/CodeGen/TargetLoweringBase.cpp
+++ b/llvm/lib/CodeGen/TargetLoweringBase.cpp
@@ -18,6 +18,7 @@
#include "llvm/ADT/Triple.h"
#include "llvm/ADT/Twine.h"
#include "llvm/Analysis/Loads.h"
+#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/CodeGen/Analysis.h"
#include "llvm/CodeGen/ISDOpcodes.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
@@ -2072,3 +2073,64 @@ TargetLoweringBase::getAtomicMemOperandFlags(const Instruction &AI,
Flags |= getTargetMMOFlags(AI);
return Flags;
}
+
+//===----------------------------------------------------------------------===//
+// GlobalISel Hooks
+//===----------------------------------------------------------------------===//
+
+bool TargetLoweringBase::shouldLocalize(const MachineInstr &MI,
+ const TargetTransformInfo *TTI) const {
+ auto &MF = *MI.getMF();
+ auto &MRI = MF.getRegInfo();
+ // Assuming a spill and reload of a value has a cost of 1 instruction each,
+ // this helper function computes the maximum number of uses we should consider
+ // for remat. E.g. on arm64 global addresses take 2 insts to materialize. We
+ // break even in terms of code size when the original MI has 2 users vs
+ // choosing to potentially spill. Any more than 2 users we we have a net code
+ // size increase. This doesn't take into account register pressure though.
+ auto maxUses = [](unsigned RematCost) {
+ // A cost of 1 means remats are basically free.
+ if (RematCost == 1)
+ return UINT_MAX;
+ if (RematCost == 2)
+ return 2U;
+
+ // Remat is too expensive, only sink if there's one user.
+ if (RematCost > 2)
+ return 1U;
+ llvm_unreachable("Unexpected remat cost");
+ };
+
+ // Helper to walk through uses and terminate if we've reached a limit. Saves
+ // us spending time traversing uses if all we want to know is if it's >= min.
+ auto isUsesAtMost = [&](unsigned Reg, unsigned MaxUses) {
+ unsigned NumUses = 0;
+ auto UI = MRI.use_instr_nodbg_begin(Reg), UE = MRI.use_instr_nodbg_end();
+ for (; UI != UE && NumUses < MaxUses; ++UI) {
+ NumUses++;
+ }
+ // If we haven't reached the end yet then there are more than MaxUses users.
+ return UI == UE;
+ };
+
+ switch (MI.getOpcode()) {
+ default:
+ return false;
+ // Constants-like instructions should be close to their users.
+ // We don't want long live-ranges for them.
+ case TargetOpcode::G_CONSTANT:
+ case TargetOpcode::G_FCONSTANT:
+ case TargetOpcode::G_FRAME_INDEX:
+ case TargetOpcode::G_INTTOPTR:
+ return true;
+ case TargetOpcode::G_GLOBAL_VALUE: {
+ unsigned RematCost = TTI->getGISelRematGlobalCost();
+ Register Reg = MI.getOperand(0).getReg();
+ unsigned MaxUses = maxUses(RematCost);
+ if (MaxUses == UINT_MAX)
+ return true; // Remats are "free" so always localize.
+ bool B = isUsesAtMost(Reg, MaxUses);
+ return B;
+ }
+ }
+}
More information about the llvm-commits
mailing list