[llvm] r367898 - [AArch64] Expand bcmp() for small block lengths
Evandro Menezes via llvm-commits
llvm-commits at lists.llvm.org
Mon Aug 5 11:09:14 PDT 2019
Author: evandro
Date: Mon Aug 5 11:09:14 2019
New Revision: 367898
URL: http://llvm.org/viewvc/llvm-project?rev=367898&view=rev
Log:
[AArch64] Expand bcmp() for small block lengths
Patch D56593 by @courbet results in calls to `bcmp()` in some cases, should
the target support the it. Unless `TTI::MemCmpExpansionOptions()`
is overridden by the target.
In a proprietary benchmark we see a performance drop of about 12% on PNG
compression before this patch, though it passes all tests.
This patch mirrors X86 for AArch64 and initializes
`TTI::MemCmpExpansionOptions()` to then expand calls to `bcmp()` when
appropriate. No tuning of the parameters was performed, but, at this point,
it's enough to recover the performance drop above.
This problem also exists on ARM. Once a consensus is reached for AArch64, we
can work to fix ARM as well.
Authors:
- Evandro Menezes (@evandro) <e.menezes at samsung.com>
- Brian Rzycki (@brzycki) <b.rzycki at samsung.com>
Differential revision: https://reviews.llvm.org/D64805
Added:
llvm/trunk/test/CodeGen/AArch64/bcmp-inline-small.ll
Modified:
llvm/trunk/lib/Target/AArch64/AArch64ISelLowering.cpp
llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.h
Modified: llvm/trunk/lib/Target/AArch64/AArch64ISelLowering.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AArch64/AArch64ISelLowering.cpp?rev=367898&r1=367897&r2=367898&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AArch64/AArch64ISelLowering.cpp (original)
+++ llvm/trunk/lib/Target/AArch64/AArch64ISelLowering.cpp Mon Aug 5 11:09:14 2019
@@ -629,6 +629,10 @@ AArch64TargetLowering::AArch64TargetLowe
MaxStoresPerMemmoveOptSize = MaxStoresPerMemmove = 4;
+ MaxLoadsPerMemcmpOptSize = 4;
+ MaxLoadsPerMemcmp = Subtarget->requiresStrictAlign()
+ ? MaxLoadsPerMemcmpOptSize : 8;
+
setStackPointerRegisterToSaveRestore(AArch64::SP);
setSchedulingPreference(Sched::Hybrid);
Modified: llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.cpp?rev=367898&r1=367897&r2=367898&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.cpp (original)
+++ llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.cpp Mon Aug 5 11:09:14 2019
@@ -618,6 +618,19 @@ int AArch64TTIImpl::getCmpSelInstrCost(u
return BaseT::getCmpSelInstrCost(Opcode, ValTy, CondTy, I);
}
+AArch64TTIImpl::TTI::MemCmpExpansionOptions
+AArch64TTIImpl::enableMemCmpExpansion(bool OptSize, bool IsZeroCmp) const {
+ TTI::MemCmpExpansionOptions Options;
+ Options.AllowOverlappingLoads = !ST->requiresStrictAlign();
+ Options.MaxNumLoads = TLI->getMaxExpandSizeMemcmp(OptSize);
+ Options.NumLoadsPerBlock = Options.MaxNumLoads;
+ // TODO: Though vector loads usually perform well on AArch64, in some targets
+ // they may wake up the FP unit, which raises the power consumption. Perhaps
+ // they could be used with no holds barred (-O3).
+ Options.LoadSizes = {8, 4, 2, 1};
+ return Options;
+}
+
int AArch64TTIImpl::getMemoryOpCost(unsigned Opcode, Type *Ty,
unsigned Alignment, unsigned AddressSpace,
const Instruction *I) {
Modified: llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.h?rev=367898&r1=367897&r2=367898&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.h (original)
+++ llvm/trunk/lib/Target/AArch64/AArch64TargetTransformInfo.h Mon Aug 5 11:09:14 2019
@@ -130,6 +130,9 @@ public:
int getCmpSelInstrCost(unsigned Opcode, Type *ValTy, Type *CondTy,
const Instruction *I = nullptr);
+ TTI::MemCmpExpansionOptions enableMemCmpExpansion(bool OptSize,
+ bool IsZeroCmp) const;
+
int getMemoryOpCost(unsigned Opcode, Type *Src, unsigned Alignment,
unsigned AddressSpace, const Instruction *I = nullptr);
Added: llvm/trunk/test/CodeGen/AArch64/bcmp-inline-small.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AArch64/bcmp-inline-small.ll?rev=367898&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/AArch64/bcmp-inline-small.ll (added)
+++ llvm/trunk/test/CodeGen/AArch64/bcmp-inline-small.ll Mon Aug 5 11:09:14 2019
@@ -0,0 +1,44 @@
+; RUN: llc -O2 < %s -mtriple=aarch64-linux-gnu | FileCheck %s --check-prefixes=CHECK,CHECKN
+; RUN: llc -O2 < %s -mtriple=aarch64-linux-gnu -mattr=strict-align | FileCheck %s --check-prefixes=CHECK,CHECKS
+
+declare i32 @bcmp(i8*, i8*, i64) nounwind readonly
+declare i32 @memcmp(i8*, i8*, i64) nounwind readonly
+
+define i1 @bcmp_b2(i8* %s1, i8* %s2) {
+entry:
+ %bcmp = call i32 @bcmp(i8* %s1, i8* %s2, i64 15)
+ %ret = icmp eq i32 %bcmp, 0
+ ret i1 %ret
+
+; CHECK-LABEL: bcmp_b2:
+; CHECK-NOT: bl bcmp
+; CHECKN: ldr x
+; CHECKN-NEXT: ldr x
+; CHECKN-NEXT: ldur x
+; CHECKN-NEXT: ldur x
+; CHECKS: ldr x
+; CHECKS-NEXT: ldr x
+; CHECKS-NEXT: ldr w
+; CHECKS-NEXT: ldr w
+; CHECKS-NEXT: ldrh w
+; CHECKS-NEXT: ldrh w
+; CHECKS-NEXT: ldrb w
+; CHECKS-NEXT: ldrb w
+}
+
+define i1 @bcmp_bs(i8* %s1, i8* %s2) optsize {
+entry:
+ %memcmp = call i32 @memcmp(i8* %s1, i8* %s2, i64 31)
+ %ret = icmp eq i32 %memcmp, 0
+ ret i1 %ret
+
+; CHECK-LABEL: bcmp_bs:
+; CHECKN-NOT: bl memcmp
+; CHECKN: ldp x
+; CHECKN-NEXT: ldp x
+; CHECKN-NEXT: ldr x
+; CHECKN-NEXT: ldr x
+; CHECKN-NEXT: ldur x
+; CHECKN-NEXT: ldur x
+; CHECKS: bl memcmp
+}
More information about the llvm-commits
mailing list